Rules and standards discussions/Archive/Archive14

This is an archive page for the Rules and standards discussions page. Please do not edit the contents. To start a new discussion, please click here.
This archive includes discussions from January - December 2016.

Archive Quick Links
Archives of old Rules and standards discussions.

1 · 2 · 3 · 4 · 5 · 6 · 7 · 8 · 9 · 10 · 11 · 12 · 13 · 14 · 15 · 16 · 17 · 18 · 19 · 20 · 21

Expanded archive listing

Grouping editor series by year

I am trying to figure out why this is a practice here. Currently, all of the issues/publications that come out in a particular year are merged into a title record by year. No one seems to know why this is done (at least the two people with whom I've discussed it). From my view, this makes it more difficult to find an issue because it is effectively hidden behind this arbitrary grouping (arbitrary in the sense that it has nothing to do with the publisher grouping them that way, but rather ISFDB choosing to do it for whatever reason).

Now, if this grouping was tread more like a regular series, where listings still showed all the issues collected in that year grouping, it would be fine. As it is, this grouping doesn't really make a lot of sense as it doesn't really seem to serve any purpose other than making a magazine issue listing shorter. I propose either getting rid of this practice or tweaking the database so the individual issues are visible in the listing (thereby making it more like a regular series in the listings). I'd prefer the former, however, as I don't see any benefit or value in putting them in year groupings. ···日本穣 · 投稿 · Talk to Nihonjoe 17:00, 18 January 2016 (UTC)

Another option would be to only apply it to series with 6 or more issues per year. For those that have fewer than 6 issues per year, it makes no sense to merge them like is currently done (we have some that are "merged" with only one issue). ···日本穣 · 投稿 · Talk to Nihonjoe 17:03, 18 January 2016 (UTC)

I think it's basically a good thing to have this feature: we not only have magazines published monthly (or four-weekly), but also some that are published weekly, so it becomes really unmanageable to have a separate title entry for every issue. And there's the problem of really long-running magazines, like Amazing. So, I think it's better to restrict the exception to magazines that are published less than four times a year: at least that's how I handled this up to now. Stonecreek 17:15, 18 January 2016 (UTC)

That would be fine, too. I just don't see the use for those with very few issues per year. ···日本穣 · 投稿 · Talk to Nihonjoe 17:44, 18 January 2016 (UTC)

When you say "if this grouping was tread more like a regular series, where listings still showed all the issues collected in that year grouping, it would be fine", do you mean how it is displayed on the title page? In fact, all the publications under this editor annual grouping is displayed, and their ISFDB records are linked to that list. But I do agree that there's no reason to create annual groupings for periodicals that publish one or two issues per year. My concern for how you were handling these had nothing to do with the unmerging of editor records, but how you were numbering them. That was non-standard, and the whole point of my post to your talk page. Mhhutchins|talk 19:42, 18 January 2016 (UTC)

I don't have a problem with that. I don't care either way on the series numbering thing. As for the "listing more like a series", I'm talking about on the series bibliography page (like this one). For all the ones grouped by year, you have to click on the year grouping to be taken to another page listing whichever issues are grouped in that year. It would be much better if they either were just all listed individually on the series bibliography page, or, if the year groupings must be kept, they were listed just like a subseries of the main series (like on this page). ···日本穣 · 投稿 · Talk to Nihonjoe 19:55, 18 January 2016 (UTC)

I'd like to point out that it would be the same number of clicks regardless. The series is going to list the title records. Even if the standard were to keep all the title records separate, you would still have to click the title record to see what publications there are of that title. The only difference would be that there would only be one publication when you click through instead of all those for the year. It's the same number of clicks to navigate to the record either way. That being said, my own preference is to keep titles grouped in annual collections which makes for a much cleaner display in the editor's bibliography page as well as the series page. I would even prefer if a single yearly title record be used even if the the editor changes in the course of a year, listing all editors that worked that year. Having multiple title records per year causes a bit of problem when linking awards that are mostly given to a year's worth of issues. This results in having to repeat awards and nominations for each title record. --Ron ~ Rtrace^Talk 20:11, 18 January 2016 (UTC)

I think there is some confusion about what I mean, so let me try to be more clear. This record shows how we currently do things (with the grouped years). What I am suggesting is that under each grouped year (if we keep the year groupings) is this:

Random Publication (View Issue Grid)

Random Publication, 1 Spring 1990 (1990) [ED] by Person One and Person Two
Random Publication, 2 Spring 1991 (1991) [ED] by Person Three and Person Four
Random Publication - 1992 (1992) [ED] by Person Four and Person Five

Random Publication, 3 Spring 1992 (1992) [ED] by Person Four and Person Five
Random Publication, 4 Fall 1992 (1992) [ED] by Person Four and Person Five

Random Publication - 1993 (1993) [ED] by Person Six

Random Publication, 5 Spring 1993 (1992) [ED] by Person Six
Random Publication, 6 Summer 1993 (1992) [ED] by Person Six
Random Publication, 7 Fall 1993 (1992) [ED] by Person Six

Something like that. Does that make sense? That way, when you view the series bibliography page you will be able to see all the issues without having to click again to go to the bibliography page for Random Publication - 1993 because I can see the contents of it directly below it (just like in a regular sub-series).

Regarding the awards, they are usually awarded to the editor, and the year given for the award should be a good enough clue about what they are for. ···日本穣 · 投稿 · Talk to Nihonjoe 00:24, 19 January 2016 (UTC)

The series is a title series, so you wouldn't see publication records, only title records. You would still have to click through the title record and click again to see the publication record. Using your earlier example, there are currently several title records that have not been merged into a single record for the year. If you click on either of the separate 1981 records, it takes you to the detail of the title record and lists a single publication which you still have to click to view. Compare this with 1990, you click on the title record for the year and see all the publications for that year. In both cases it takes you 2 clicks to get from the series page to the publication record.

My main experience with awards is with the Hugos and they are awarded to the magazine or fanzine, not to the editor. Thus in the database, they need to be linked to ta title record. If we have multiple title records for an award year, we have to either duplicate the award record for each separate title, or choose one from among the many title records from the year. Not merging title records as per the current policy would make that problem worse. --Ron ~ Rtrace^Talk 02:55, 19 January 2016 (UTC)

Yes, but the "title" record in the case of the grouped-by-year titles isn't a useful title as there could be anywhere from 1 to 52 individual records inside. That is the biggest reason why having the grouped-by-year titles doesn't make sense. Comparing them to regular titles isn't useful as a regular book title may have 100 different publications, but they are all (for the most part) the same content. Issues of a magazine are not. They are each unique titles. That's why grouping them by year makes no sense at all. Now, I have no objection if all of those by-year groupings were turned into publication series, because they are similar to book publication series in that they all have some sort of theme in common (in this case, all of them are issues of the same magazine). Grouping them like that would actually make sense. Magazines often already do that with their "Volume A, Issue 1, Number 1" numbering. ···日本穣 · 投稿 · Talk to Nihonjoe 03:04, 19 January 2016 (UTC)

Having the credits of the EDITOR title record differ from the author field of the publication record would cause a conflict in the software. The author field of a title record must be identical to the author field of the publication record. A way to get around this is to variant all of the EDITOR title records of the various issues to a single EDITOR title record crediting all the editors of that year. So you have to variant and not merge. Depending upon the periodical and the number of differently edited issues in any given year, the display wouldn't necessarily be less cluttered. Mhhutchins|talk 20:20, 18 January 2016 (UTC)

You're absolutely right and I hadn't considered that. I'd been kicking around the idea in my head for a while. The magazine that made consider it is Journey Planet which I believe always has guest editors and is frequently nominated for a Hugo. I agree, that additional variants wouldn't make things cleaner. --Ron ~ Rtrace^Talk 20:29, 18 January 2016 (UTC)

This wouldn't address clutter, but I could see moving away from merging the titles to using sub-series. Then the (optional) secondary grouping could be varied according to what makes the most sense: year, volume, maybe even editor, and would otherwise be independent of the actual titles/credits. --MartyD 20:58, 18 January 2016 (UTC)

I really appreciate being able to go to something such as Science Fiction Review and being able to see an overview of the entire series on one screen. If we put each issue into its own title entry, as suggested at the top, there would 127 titles. Too ugly. And if you DO want to see all of the issues, or being able to get quickly from one issue to another, then go to the Issue Grid! That arranges the entire series with separate buttons for each issue, but it still takes up only a screen and a half. Chavey 03:34, 19 January 2016 (UTC)

So, can we not group them if there are less than four issues per year? It seems like that's the number mentioned by a couple people. ···日本穣 · 投稿 · Talk to Nihonjoe 03:09, 21 January 2016 (UTC)

Having an exception when there are less than four issues seems like an arbitrary rule. Since my preference is for a clean display on series and author bibliography pages, I would argue that EDITOR title records should always be grouped and even if there is only one record for the year, I'd still prefer that the format of <Title> - <Year> be observed. Otherwise the singletons (or triplets) really stand out from what would otherwise be a standardized list. --Ron ~ Rtrace^Talk 03:25, 21 January 2016 (UTC)

In the past I've also seen the figure of 6 mentioned, but 4 seems to be common. But I do think it's an arbitrary number – while I think it often helps to have some in-built flexibility and a few small grey areas here and there, the annual grouping of magazines seems to be a rather larger one. I personally have no objection to grouping ALL issues by year, even with less than four, even with one. It's necessary to do so when it comes to awards, and it hurts my eyes when I see all issues listed individually interrupted by "Pub Name – 2005" and "Pub Name – 2012", because in those years that publication got nominated or shortlisted for an award, which only becomes apparent to the casual user when that link is clicked. A few months ago I checked into SF Commentary, where I had had every issue listed individually except for its award nomination years, and found that every publication had been grouped by another editor into annual listings, even those years which saw just one issue. I wondered, is this a new thing I ain't aware of, an effort to tidy up presentation? I don't actually object to it, it looks OK despite the extra clicks being necessary, and it makes invisible those awkward annual interruptions I mention above. But I would like to see a black-and-white rule about this to save ourselves from endless tinkering and re-formatting of the data and its presentation.

To go back to Michael's original point (the serial numbering of magazines/fanzines, or the avoidance of it), again seems to be something not set in stone, unless someone can point me to the rule for I've obviously not seen it. I've come across several magazines/fanzines that have listed serial numbering, and I tinkered/experimented with this with Journey Planet. One feature I would like to be looked into is that if the software were to allow hyphens in the serial number entry field, this field could then be used to display all issues for a given grouped year, eg. "1-4", "5-8", "1-12", "12-24", etc. PeteYoung 04:09, 21 January 2016 (UTC)

That would be difficult to implement because series numbers are used to sort series entries. Actually, there are two separate "series number" fields in the database to support series numbers like "3.75", which makes the supporting software somewhat complex. Adding support for values like "12-24" would make things even more complicated. Ahasuerus 05:20, 21 January 2016 (UTC)

Maybe do it like an omnibus entry: O/1,2,3,4,5 or something? ···日本穣 · 投稿 · Talk to Nihonjoe 05:45, 21 January 2016 (UTC)

My two cents. PeteYoung 04:09, 21 January 2016 (UTC)

Well, you can't eat your cake and have it, too. We either group EDITOR records into annual records or we number each record individually and each issue is displayed on the series bibliography page. Then we can get rid of the magazine grid page, because it would be redundant. I hope that Nihonjoe realizes that every thing he's asking for is already available, right there on the magazine grid page: a link to every issue of a periodical on a single page. Or we can get rid of merging altogether, and go back to listing every issue on the editor's summary page. I'd love to see John W. Campbell's summary page containing 400+ links to each issue of Astounding/Analog that he edited. Mhhutchins|talk 04:52, 21 January 2016 (UTC)

(unindent) Yes, but it's an additional click to get to it. All of these ways are currently additional clicks, all because people want to keep the page short. Some pages are just going to be long, and that's just the way it is. Look at Robert Silverberg's page, for example. It goes on and on and on forever. People who have done a lot will have pages with a lot of entries. I don't see why magazines (which are very similar to anthologies, only with a regular release schedule) should be treated any differently than an anthology. Every individual anthology a person edits is listed on their page, so why not every individual magazine? "Because it would list more on the page" isn't a good reason. People come here to be able to see everything someone has done, listed by the titles under which they were published, not grouped and hidden in some arbitrary way that no one else anywhere uses. ···日本穣 · 投稿 · Talk to Nihonjoe 05:43, 21 January 2016 (UTC)

You keep stating that additional clicks are required, but I still maintain that it is 2 clicks to get from the series bibliography page to the publication detail page. No matter if the titles are combined or not, the first click goes to the title page, the second to the publication page. What additional click are you seeing with the combined titles? --Ron ~ Rtrace^Talk 06:53, 21 January 2016 (UTC)

You're right, Ron. It's the same number of clicks. A merged annual EDITOR record appearing on an author/editor's summary page is the same as the title record of an ANTHOLOGY or a NOVEL. A publication record is always two clicks away from a summary page. That's the basic database structure. Listing every EDITOR record separately on a summary page doesn't reduce the number of clicks. You'd still have to click forward from the EDITOR record to the PUBLICATION record. So isn't it nice to have a magazine grid so that you have a single page with links to every publication record for that periodical?

Nihonjoe, we've tried to explain quite valid reasons why editor records are merged into annual groupings. From your responses, I think you believe that there should be a link from the author/editor's summary page to every PUBLICATION record, not just TITLE records. The reasons you give for such a change are not just impractical. They're basically impossible. Your reasons don't justify a change in how the database is structured, or the effort it would take to go back to 1995 and start all over. I apologize if my tone comes across too harsh here, because you are an important part of the editing team on the ISFDB. But don't take it personal. I've often found myself in a corner arguing about a change only to eventually realize that I was wrong. Just as many editors here have. Ce la vie. Vive la ISFDB. Mhhutchins|talk 18:53, 21 January 2016 (UTC)

No, you are misunderstanding, still. I think there should be a title record for each issue because each issue is a unique title with unique content, and that each of those title records should be linked on the author/editor page. Magazine issues are not the equivalent of various editions of a single book, which is how they are treated now with the annual grouping which is only used here. As for the "second click" thing, let me try to explain it again since no one seems to understand. Right now, with the yearly groupings, you get an entry for Random Publication - 1993, which doesn't tell you which issues are part of it. You have to click through the link to see the list of issues before you can then click on the specific issue to see what is in it. That is an extra click. What I am suggesting is that there be no by-year groupings and each issue be listed on its own so you don't have to go through that extra by-year listing in order to get to the issue list in the first place. The grouping by year is what is causing the extra click. To put it another way, from The Leading Edge page:

Current method: The Leading Edge -> The Leading Edge - 1990 -> The Leading Edge #20/21, April 1990 (three clicks)

What I propose: The Leading Edge -> The Leading Edge #20/21, April 1990 (two clicks)

I don't know how much more clearly I can explain it. ···日本穣 · 投稿 · Talk to Nihonjoe 06:20, 22 January 2016 (UTC)

I think what other people are trying to say is it's still two clicks to the contents of the magazine, no matter which way it's displayed. The TITLE is either The Leading Edge - 1990 or The Leading Edge #20/21. Whichever it is, you click on that and now get a list of the publications associated with that title. In the former case, you may see several publications, in the latter case you will see only one publication. In either case, you now click on the PUBLICATION The Leading Edge #20/21 to see the contents. --MartyD 17:36, 22 January 2016 (UTC)

I am sympathetic to the point about lack of visibility into the issue numbers if year-based groupings are used, particularly when a magazine's issues are not synchronized to the calendar. But with properly formatted publication titles (comma before the issue number), the grid display takes care of that and still requires just the same two clicks from the summary page (1 click to get grid, 1 click to see desired issue's contents). --MartyD 17:36, 22 January 2016 (UTC)

Yup. I need to go and normalize all of the issues in this series as they are not consistent. ···日本穣 · 投稿 · Talk to Nihonjoe 19:42, 22 January 2016 (UTC)

You're saying that unmerging editor records from their annual groupings will save you a click, and many people here are telling you that it won't. And you just don't believe us. So here's a task for you. Unmerge every issue of The Leading Edge (I'll accept the submissions), and see how it works. If that's the only way to convince you that you can't get to a pub record without clicking on its editor record first, it's a small price to pay. And a way to end this discussion. Mhhutchins|talk 19:58, 22 January 2016 (UTC)

Current method: search for and click on "The Leading Edge" -> The Leading Edge - 1990 -> The Leading Edge #20/21, April 1990 -> click on the individual pub (four clicks)

What I propose: search for and click on "The Leading Edge" -> The Leading Edge #20/21, April 1990 -> click on the individual pub (three clicks)

How is that not clearly fewer clicks? I don't need to unmerge them to see that. Regardless, I'm through banging my head against this particular wall. It's a complete waste of my time when no one pays attention to what I actually write. ···日本穣 · 投稿 · Talk to Nihonjoe 20:56, 22 January 2016 (UTC)

The problem is your count for the current method is incorrect. When you click on "The Leading Edge #20/21, April 1990", you are clicking on the pub link. That is three clicks to the pub, not four. Looking at "The Leading Edge" currently shows examples of both cases that are being discussed. If you try both, you will see it is the same number of clicks. Do a magazine search for "The Leading Edge". You will be presented with a listing of merged and unmerged title records.

Click on The Leading Edge Magazine of Science Fiction and Fantasy, #10 Fall 1985 which is an unmerged title record. There will then be a single publication record listed. If you click on it, you will be at the publication record. 3 clicks.
Click on The Leading Edge - 1990 which is an merged title record. There will then be two pub records listed. Click on The Leading Edge #20/21, April 1990 and you will be at the publication record. 3 clicks.

The counts are the same because the structure is the same for both cases (series -> title -> publication). Merging titles does not add an additional level. It just groups multiple publications under the same title. -- JLaTondre (talk) 13:41, 23 January 2016 (UTC)

(unindent) We already special case the issue grids which gives you a list of pubs instead of editorial titles. I really am not fond of the way we implement such but it is historical and works. This is a case where I believe we should use pub series (this has been brought up before; magazines were done they way they are partly due to the fact that existed here before pub series arrived). If we turn magazines into pub series much of the problems go away. This is not to say we do not need and want editorial series for editorial attribution (we can and probably should still keep those). Using pub series for magazines also simplifies things that currently get somewhat complex when magazines are later reprinted (reprints are new pub series but with the same editorial credits). Like all pub series, magazines are basically one-shots. Once published they are done. Many anthology series should likely also be handled this way too (of course we still want editorial attribution records too). The way I see it pub series are basically editorial works (without accreditation which we handle another way). Uzume 01:28, 23 January 2016 (UTC)

On the off-chance that it helps, let's compare two fanzines, one of which has been merged by year and one of which has not. The fanzine Haunted, by Samuel D. Russell only has one issue listed by us, no one has updated it to the modern "year" style, hence it appears to be in the format that Nihonjoe would like us to return to. If you start from either the fanzine page (e.g. you looked up "Haunted" in the magazine search) or from Russell's summary page, you would, in one click, get to the title page for Haunted #2. On a second click, you would be at the publication page for that particular issue. With the 1939 "New Worlds" fanzine, we could again start either from the magazine page or from John Carnell's page. In either case, we use one click to get to New Worlds - 1939. This title includes four publications, instead of the one in the previous example, but we still get to the publication of interest with our second click. There is NO DIFFERENCE in the number of clicks. There are other advantages and disadvantages to these two arrangements, but the number of clicks IS NOT IT. Some of us like the ability to see other nearby issues, as we do with New Worlds. Certainly there are times when not having the issue number quickly available means I have to go through a couple of annual titles to find the issue I'm looking for (which is why we invented the issue grid!). But any argument for change must come from those other pros and cons, not from the number of clicks. Chavey 06:44, 24 January 2016 (UTC)

Note field recommendations for citing titles

Help:Using HTML in Note Fields#Italics suggests using ... for italics mentioning the common English case of italicizing publication names (but not normally lesser works like short stories which are not published by themselves). That said, I would like to propose the recommendation be changed to recommend <cite>...</cite> which is semantically specified for citing titles and by default render as italics (much like ... does) in virtually all browsers. Uzume 04:42, 28 January 2016 (UTC)

As a recommendation I think that makes a lot of sense. Albinoflea 21:02, 29 January 2016 (UTC)

But why replace the method, other than someone (who knows who?) saying it is "semantically specified for citing titles"? Needing to use more characters when the display stays the same just isn't sufficient reason to change the standard. Mhhutchins|talk 22:17, 29 January 2016 (UTC)

The HTML standard (version 4) says that "[cite] contains a citation or a reference to other sources" while "i" is a generic font style element. I suppose the former is a better fit for our purposes, but the difference is so minor that I am not sure it's worth changing our data entry standard and making thousands of record non-complying. Ahasuerus 03:09, 30 January 2016 (UTC)

Unless someone reads the raw HTML (and why should they?) what does it matter that "cite" is used instead of "i"? Would some deus ex internetica reveal to the users that these italic words mean something different than those italic words? Mhhutchins|talk 04:20, 30 January 2016 (UTC)

Right now it may not make a difference if it's rendered in italics like an element, but it could make a difference in the future, if the ISFDB wants to. Using it opens possibilities for new features and ideas: you could simply decide to render <cite> elements differently in order to distinguish them from other emphasized (italics) text. You could also extract these data from the note field more easily, e.g. to create statistics of its usage, or to create an index of "Titles cited in ISFDB records", etc etc. However, if it's certain that features like these are never needed (but how can you in a Speculative Fiction database ;) ), then keep using . Having said that I'd personally prefer using <cite>, but it might nevertheless be better to keep the rules more simple, especially when considering the amount of already existing records using and the additional time it'd take to teach new editors the correct usage of <cite> and . Jens Hitspacebar 11:32, 30 January 2016 (UTC)

It also helps with different types of rendering. Supposed you wanted to read through the ISFDB website looking for something but then suppose you were blind. How should your browser render italics? If it is more semantically tagged then you could specify how you wanted book titles rendered (perhaps a different voice or something). I am not saying we go through and change all the notes tomorrow or reject submissions with ... tags. I am just suggesting a change to the recommendation. It should perhaps be noted that HTML5 has restricted the semantic meaning of <cite>...</cite> to represent "the title of a work" (much as it was defined in HTML2 which states used to indicate "the title of a book or other citation"). Rendering standards also vary by language. As you can image, italics does not hold the same value in Japanese (and likely other Chinese-based scripting). So if someone quotes a title in Japanese it is usually bracketed with some sort of quoting mechanism instead. Using the more semantically defined tag could allow for such as well (as we internationalize and localize our database and its contents we will likely start to attract users of other cultures more and internationalizing and localizing the user interfaces might be useful too). Uzume 15:42, 31 January 2016 (UTC)

ISFDB policy on file hosting

Can the ISFDB server host art image files or text files, whether they're copyrighted or not? I'm not speaking of book cover images, since that is considered fair use, but I'm referring to images of the original art or pdfs (and other formats) of essays, short stories, even novels, whether they've been previously published or not. If we agree that we shouldn't (and I believe most of us feel that this falls outside of the purpose of the ISFDB), should it not be documented in the policy? Mhhutchins|talk 17:50, 3 February 2016 (UTC)

I agree that expanding our hosting rules beyond cover scans may raise various technical (disk space) and legal (copyright) issues. However, we have also used the Wiki to store certain other types of files like artist signatures and the occasional table of contents scan (when dealing with particularly hairy pubs.) For this reason I think that it may be better to enumerate the allowed file types in Help rather than on the Policy page. Ahasuerus 21:11, 3 February 2016 (UTC)

Yes, the types of accepted files should be listed in the Help section about uploading, but I believe which files are accepted should also be a documented policy. It shouldn't be more than a sentence or two. Mhhutchins|talk 21:35, 3 February 2016 (UTC)

I guess my main issue with putting the same information in two different places is that it will be easy for the pages to get of out sync if and when we add more file types to the list. Not a big deal, though.

So, aside from cover scans and artist signatures, what other types of files do we currently host? Ahasuerus 23:57, 3 February 2016 (UTC)

We have author photos, and scans of miscellaneous bibliographic data (copyright pages, etc.) as well.

My question was prompted by this upload which doesn't fall into any of the commonly accepted categories. When I questioned the uploader his response was "The book, and therefore the cover, are both out of copyright." If that's true (which I doubt) the wrong license has been attached to it. Nevertheless, I believe the ISFDB should not be hosting images of original art, regardless of whether it is "out of copyright." Mhhutchins|talk 00:25, 4 February 2016 (UTC)

I don't disagree with you in principal, although I think there are cases where we could, and perhaps should, host such things. For example, I could see hosting original art where the source of artist credit is the signature seen only on the original (cropped or obscured when used as a cover or interior illustration). At the same time, I don't think we should be hosting original artwork behind every cover or illustration.... --MartyD 02:02, 4 February 2016 (UTC)

We already allow for cases like that, but only the artist signature is uploaded, not the complete work of art. Mhhutchins|talk 02:17, 4 February 2016 (UTC)

In one case we have a photo of a t-shirt. (With half of the text of a story.) :-) Chavey 09:13, 4 February 2016 (UTC)

Unpublished books deletion request

In a recent discussion, User:JVRudnick, aka the author Jim Rudnick explained that:

Pirates with a pub date of 2014 [ISBN 978-1-4700-0983-0] was NOT released -- this is an error over at CreateSpace

and that the book was published in 2015 with a different ISBN. Based on his communication, I changed the publication date of this ISBN to "unpublished" and updated the Note field. I also explained that:

That way anyone searching for this ISBN will find it and realize that it was "vaporware".

However, as per the author:

yet it still causes confusion within my readers--they all wonder if this is a real live book etc. Can you not TOTALLY delete this non-book? listing and cover please...

I don't think we have encountered this problem before, so I am posting it on the Rules and Standards page.

Personally, I am surprised that this is an issue for some users since we clearly state that the ISBN was never published, but stranger things have been known to happen. Ahasuerus 20:56, 12 February 2016 (UTC)

I believe it would be more confusing to not list it as it would tend to cause people that find the ISBN reference to think it exists without other data and editors here might tend to want to enter it again. Uzume 21:06, 12 February 2016 (UTC)

I agree. Leave it as "unpublished" and if a reader doesn't understand that, then they don't read very well. That's their problem, not ours. Mhhutchins|talk 21:12, 12 February 2016 (UTC)

I agree that we want to keep the record for the reasons outlined by Uzume. On the other hand, "unpublished" dates are kind of a special case, so I went ahead and adjusted the code to highlight them, e.g. here is what the offending title and publication records look like now. Does it make sense? Is the color OK? Ahasuerus 00:10, 14 February 2016 (UTC)

Rather bright, and aesthetically offensive, if I might add a personal opinion. But I guess it does perform its purpose, especially for those readers who need such a thing highlighted. Now we'll have explain to them what it means. :( Mhhutchins|talk 00:16, 14 February 2016 (UTC)

Yes, my first response was "my eyes!", but it does get the job done. The highlighting it is nice but I am not sure we really need it so garish. Are there plans to make similar highlighting on other pages and for unpublished titles too? E.g., this unpublished title record probably does not need such a loud underscoring (but some highlighting might be interesting): "Run, the Spearmaker" (which I credited to Lil Neville based on the SFE record). Uzume 17:29, 15 February 2016 (UTC)

That's odd, the highlighting was supposed to be applied to all pages. Let me check the code... Ahasuerus 21:28, 15 February 2016 (UTC)

The reason it doesn't apply here is that this is an "unpublished" title record, but without an "unpublished" publication record. (Please don't add a publication record just to get this blazing notification to work.) Mhhutchins|talk 22:16, 15 February 2016 (UTC)

It turns out that it was a flaw in the software. There were a few places in the regular search, Advanced Search and Title areas that didn't use the standard date handling algorithm. Everything should be fixed now.

As far as the color goes, as I said earlier, I will gladly change it to the consensus choice. I am colorblind and apparently tend to pick garish colors, which at least register as "colors" in my brain. Ahasuerus 22:27, 15 February 2016 (UTC)

How about doing nothing? I again question the necessity for it at all. Why go through all of this effort to appease a single author who is unable to understand what is meant by "unpublished"? Mhhutchins|talk 23:13, 15 February 2016 (UTC)

The author was originally happy when we changed the publication date to "unpublished". However, he subsequently received feedback from his readers who were having trouble interpreting our data. I figure that if adding background color helps "unconfuse" a subset of our users, it's a small price to pay. Besides, it uncovered the bug that I mentioned earlier, so it was a net plus from the software integrity standpoint :) Ahasuerus 23:40, 15 February 2016 (UTC)

But this highlighting doesn't "unconfuse" anything! There's no extra explanation, because it isn't necessary! How does highlighting change the meaning of the word or make it less confusing? I'd truly love to read what "his readers" are complaining about. Maybe he's got a large following of mentally challenged readers. And I'm not being facetious! Mhhutchins|talk 23:51, 15 February 2016 (UTC)

My best guess is that they were simply inattentive and missed the word "unpublished". If that was the case, then adding highlighting should have helped them. If, on the other hand, they couldn't interpret "unpublished" correctly, then... well, I am not really sure what we could do to help them, but Jens's idea (see below) may be worth trying. Coding-wise it's pretty straightforward. Ahasuerus 00:27, 16 February 2016 (UTC)

I actually think that the highlight might confuse more people than it helps to "unconfuse" a few, as long as it's not explained. Considering a casual visitor not very familiar with the site: if something stands out like this the reason for it should either be explained, or the explanation should be easy to find somewhere else, or the reason should be intuitively comprehensible (which I think it's not). Otherwise you'll probably search the site asking yourself: "Why is it highlighted in green. What does it mean?" Or, well, if you're not so curious, you'll just ignore it :) Jens Hitspacebar 17:47, 16 February 2016 (UTC)

I'll be more than happy to adjust the color based on the preferences of our color-enabled editors :-) Any color is fine although this table of 140 "named" colors is an unofficial standard of sorts. Ahasuerus 00:46, 14 February 2016 (UTC)

Another possibility instead of (or in addition to) highlighting is to add the question mark icon (the one used on the edit pages) next to the "unpublished" label and explain the meaning of "unpublished" in a mouseover hint. Jens Hitspacebar 20:55, 15 February 2016 (UTC)

At least in this case, what it means is explicitly stated in the notes to the book. Chavey 00:20, 14 February 2016 (UTC)

That's the point I'm trying to make, but it seems that we're dealing with pre-schoolers here, who need everything explained to them. Mhhutchins|talk 22:16, 15 February 2016 (UTC)

Popular Library

Does anyone know if Popular Library, Inc., a publisher of sf periodicals from 1964-1971, is the same as Popular Library, publisher of paperback books from 1944 to 1991? If so, was the publisher designation intended to separate the book publications from the periodicals? Mhhutchins|talk 00:49, 16 February 2016 (UTC)

I suspect so. I looked at some of the early books we have listed by "Popular Library", including the 1948 "The Case of the Crumpled Knave". One listing for this book on Abebooks explicitly lists it as published by "Popular Library, Inc.". So that book, at least, indicates these two publishers are the same. If we looked through more "Popular Library" books, I suspect we would find a lot of them being from "Popular Library, Inc.". Chavey 02:36, 17 February 2016 (UTC)

It appears that whoever entered the periodicals made a point of distinguishing the publisher name with the sole intention of keeping the books separate from the magazines. If that's the case, I'd just as well leave it as is. (I just now realized I posted this on the wrong page, and will copy it to the ISFDB:Community Portal to get other opinions on the subject. Thanks.) Mhhutchins|talk 06:51, 17 February 2016 (UTC)

Etiquette Question - Multiple Verifiers

I don't know that we have it formally stated anywhere, but it is the accepted standard to notify the primary verifier of a publication when submitting edits to that record. When there are multiple primary verifiers, I have always considered them to have equal status. Thus I notify all primary verifiers, provided they don't have a note that they are inactive, or they don't have instructions that they don't wish to be notified by the type of edit I'm doing. I've occasionally been the first primary verifier and received a note, only to discover that none of the other verifers have been notified. In these situations, I feel it is my responsibility to notify the remaining verifiers which the author of the edit failed to do. My question is how others approach this. Do others notify all active verifiers? Should new editors be advised to notify all? Or, do folks believe that notifying the first active verifier is sufficient? Thanks. --Ron ~ Rtrace^Talk 03:28, 18 February 2016 (UTC)

I have a different take on notification etiquette. I don't notify other active primary verifiers if someone notifies only me because I hold the PV1 position. I do a physical examination of the publication, either make the change or agree that the notifying editor should make the change. And I feel no guilt about not notifying any other active verifiers. After all, it was their responsibility to check my data before doing a second PV. If I missed it, and they missed it, what's the point of letting them know that some other editor caught it? Mhhutchins|talk 05:45, 18 February 2016 (UTC)

Similar take over here: I'd notify the others only if there's a new information, such as revelation of a cover artist. Stonecreek 07:02, 18 February 2016 (UTC)

I only notify the primary verifier, or the first one that's still active, unless I think a change or question might be contentious, or I and PV1 cannot agree on how something should be handled. Chavey 07:38, 18 February 2016 (UTC)

Apart from "technical" modifications, I usually notify all PVs (no hierarchy among them) according to their stated preferences. Not only for simple politeness but also because I suppose that all the maniac enough bibliographers that contribute to the ISFDB are likely to have their own databases into which they will happily enter new data (or correct false informations). That's what I ask of new contributors, which seems consistent with this point. Hauck 14:23, 18 February 2016 (UTC)

I generally notify all primary verifiers just so they are aware of the proposed changes. It only takes a couple minutes. ···日本穣 · 投稿 · Talk to Nihonjoe 17:01, 18 February 2016 (UTC)

I notify all active verifiers of every change, and most inactives too if there's new information (cover artist, publication date etc) from other sources. --Willem 19:43, 18 February 2016 (UTC)

Scientific American special issue

I own a special issue of Scientific American magazine (brasilian edition) that is totaly devoted to Artur C. Clarke. You can see the cover here. This issue is the number 4 of a serie of special issues called "Exploradores do Futuro"(Explorers of the future"), dedicated to well known science fiction writers like Asimov, Verne or Wells. It not contains any speculative fiction in it, it only contains articles about the author and his work, and i emphasize that is only dedicated to the author and hasn't any other kind of content, it could be seen as a publication about an author. After reading the rules and discuted it with one of the moderators i think the rules about non-genre magazines are not clear in this particular case. This non-genre magazine doesn't contain, in fact, any speculative fiction, wich makes it automatically excluded from the database, but in the other hand is a publication entirely dedicated to an science fiction author, like many other non-fiction publications in the database, and this is very diferent from the case of a non-genre magazine that has only one or two articles about an author or a book. So, my question is: do you think a publication like this has a place or is relevant to the database, or because it not contains any speculative fiction in it and it's only about speculative fiction is out of the scope of the database? Thanks.Wolland 00:17, 26 February 2016 (UTC)

If any of the content is about them, it's perfectly fine to include it here. If the entire issue is about them, then include all the content. Any content not about them or their works can be included in the notes. ···日本穣 · 投稿 · Talk to Nihonjoe 00:56, 26 February 2016 (UTC)

Here is an example of including non-genre content in the notes. ···日本穣 · 投稿 · Talk to Nihonjoe 00:57, 26 February 2016 (UTC)

Sorry, Nihonjoe. Saying "it's perfectly fine to include it" is incorrect advice. You'd have to agree that Scientific American is a non-genre magazine, so it would fall under the rules established for entering non-genre magazines into the database. An issue of a non-genre magazine which contains a work of speculative fiction is eligible for the database. If we were to allow the creation of publication records for any non-genre magazines which contain work about speculative fiction, then we would be overwhelmed. (A record for an issue of TV Guide with an article about a TV series based on a Stephen King novel, perhaps?) Where would we draw the line? Well, the line has already been drawn and the current rules are clear. If you want to change the rules, that's fine. We can have a discussion about that. If you want to make an exception for a Portuguese magazine with uncredited work which will wind up on a page with thousands of other uncredited works, then we can discuss that. But don't say a publication is eligible for the database without knowing the standards. Mhhutchins|talk 01:19, 26 February 2016 (UTC)

Which is basically what I wrote, but in different words. Since he said they issues were entirely about the genre authors, it is perfectly fine to include them. I didn't say to include every issue of Scientific American and that's not what Wolland was asking. I answered his/her specific questions. No more, no less. ···日本穣 · 投稿 · Talk to Nihonjoe 02:00, 26 February 2016 (UTC)

Then you're doubly incorrect. Please re-read what I said. I was talking about single issues. Nothing was said about allowing "every issue" of a non-genre magazine. Mhhutchins|talk 02:20, 26 February 2016 (UTC)

As for your comments about "a Portuguese magazine with uncredited work", I'm not sure where you got that from. I said nothing about the works being uncredited, and neither did Wolland. I'm not trying to change the rules. We regularly include reviews and interviews and other articles about works and individuals, so this is no different than that. Any non-genre stuff would be in the notes or not included at all. Please stop reading more into my comments than is there. ···日本穣 · 投稿 · Talk to Nihonjoe 02:00, 26 February 2016 (UTC)

The contents are not credited. I got that directly from the submission itself. And you're wrong to say this is "no different than that". Those articles are published in genre magazines, making them eligible for the database. I'm not saying you're trying to "change the rules". I'm saying you don't understand the rules. I'm not "reading more into [your] comments". That's why I'm quoting them exactly. Mhhutchins|talk 02:20, 26 February 2016 (UTC)

The specific wording is "Any issues of any periodical containing works of speculative fiction can also be entered in the same way as a non-genre magazine." I have always interpreted that to mean "Any issues of any periodical containing works of [or about] speculative fiction can also be entered in the same way as a non-genre magazine." Since we are trying to be as comprehensive as possible in our coverage of speculative fiction in any printed form, I don't see how my interpretation of that line makes things any more difficult or overwhelming. If the amount of work required of moderators is so significant, perhaps more ought to be appointed to help spread out the workload further. ···日本穣 · 投稿 · Talk to Nihonjoe 02:05, 26 February 2016 (UTC)

In this case, is not an article or two of in a non-genre magazine, is an entire publication, out of the ordinary series of Scientific American, with a diferent numbering. So the comparison with the TV Guide don't seems a suitable example. And if the name of Artur C. Clarke is included in the title of the Magazine (like " Scientific America - Arthur C. Clarke(special issue)) it will be easily tracebale. And just to remember, Mhhutchins, portuguese language is the 4th most talked language in the world and don't deserve to be minorized like you did.Wolland 01:46, 26 February 2016 (UTC)

That was not my intention. I apologize if that wasn't clear in what I said. If I minimalized anything, it was to point out that the contents weren't credited. And the title of the record would not be "Scientific America - Arthur C. Clarke(special issue)", because that's not how we title issues of magazines. That data would have to be given in the Note field. So, it would only be searchable if a user were to use the advance search, using the correct parameters, not so easy as you would think.

Since you're only now making it clear that this issue was not part of the magazine's regular run, and that it's not truly a periodical, there is a way that it can be entered into the database: if we don't type it as MAGAZINE, but as NONFICTION. Mhhutchins|talk 02:20, 26 February 2016 (UTC)

(unindent) It looks like it may be time for my semiannual reminder: the Wiki software is a wonderful tool, but, unfortunately, it doesn't transmit your body language, which can cause all kinds of misunderstandings. Sometimes our writing comes across more intense than intended. Ahasuerus 03:20, 26 February 2016 (UTC)

Wolland, I will continue our original discussion, back on your talk page. Mhhutchins|talk 04:39, 26 February 2016 (UTC)

Well on a related note, I wonder if we consider Scientific American a non-genre magazine or a non-fiction magazine or both. I recently came across some of Daniel Keyes's publications and apparently we have non-fiction titles of his (with pubs) both marked non-genre and genre. I am am sort of confused by this, however, I can understand how non-fiction works about speculative fiction might be genre and others non-genre. I straightened out the publications and titles (with regard to Japanese translations, etc.) but I did not delve into and try to rectify whether we should have such records to begin with (or whether they should be marked non-genre or not). Uzume 21:53, 26 February 2016 (UTC)

I consider Scientific American to be both a non-fiction magazine and a non-genre magazine. I interpret the ISFDB definition of "non-genre magazine" to be any periodical which isn't considered a speculative fiction-related magazine. (My interpretation may not match that of other users.) Even though Locus is a non-fiction magazine, it's still considered a periodical of the spec-fic field, thus a "genre magazine". Our rules for a "non-genre magazine" wouldn't apply to Locus, and all of its contents are eligible for the db.

As to your other point, the ISFDB non-genre flag for titles was originally intended to be used to mark fiction only. But I see that some editors (and accepting moderators) are also using it for non-fiction, as in the case of Keyes. There is a valid case for saying that a work can be both non-fiction and non-genre. Even so, I don't think it's been thoroughly discussed and documented. And if you believe it should be, I suggest starting a new topic. Thanks. Mhhutchins|talk 19:47, 27 February 2016 (UTC)

Cover art title naming

It is really necessary and/or useful to have cover art titles still be prefixed with "Cover: "? Perhaps I am wrong but I believe this was originally necessary due to a limitation of the software which we no longer have. I ask this in light of our move towards better language support and "cover" being an inherently English word, it seems limiting when dealing with foreign titles (e.g., I have to enter the cover art title of a work named "狼と香辛料" as "Cover: 狼と香辛料"). Since we now have a title type for cover art, it seems to me that this is unnecessary and I would like to propose and discuss changing our ways to remove this. Thank you. Uzume 01:50, 4 March 2016 (UTC)

I have to enter the cover art title of a work named "狼と香辛料" as "Cover: 狼と香辛料".

No, you don't. The system does that for you "automagically." :) Mhhutchins|talk 05:39, 4 March 2016 (UTC)

Actually I do (and I get a pop-up dialog when I try to submit without naming them in such a way, telling me I have to). It might name them that way for me upon entering new pubs but not when updating existing pubs and I find myself having to update many Japanese pub records to native script that were originally entered with Latin transliterated titles. The point is: can we consider and hopefully decide to be rid of this historic vestige? As an interim measure, we could probably update the software to have it always display "Cover: " before those title names (and not require such upon entry) and run an update script to remove the "Cover: " prefix from all the cover art records. Then at least visually it would be the same (but would be more flexible in that the word "cover" could be displayed differently later). Uzume 12:32, 4 March 2016 (UTC)

I'm sure at one time it was necessary to distinguish cover art records from other records. I can't say specifically why the convention was adapted since it was implemented before I started here. Personally, I like the distinction. But I'll go along with whatever is decided by consensus. (Again, you never have to add "Cover:" to a COVERART title, even when you're editing it. It's already there. You just have to change the title of the work.) Mhhutchins|talk 15:19, 4 March 2016 (UTC)

I guess the first thing to do is to find all of our Web pages that show COVERART titles and see whether they still make sense if the "Cover:" prefix is removed. I'd also have to check the internal logic which expects COVERART titles to have this prefix. Ahasuerus 16:44, 4 March 2016 (UTC)

OK, there is no particular hurry but suffice it to say I am requesting such an investigation and if there is no need to carry this baggage around any longer let's decide to dispense with it and create an FR (and then it can get prioritized however among the others). Uzume 17:35, 4 March 2016 (UTC)

On artist's pages such as this (and search for "New Maps of Hell"), the "Cover: " prefix differentiates between a) COVERART and b) INTERIORART that has been varianted to COVERART (and the other way round). I like to be able to see that difference on artists' pages. Horzel 14:44, 12 April 2016 (UTC)

I also fear that dropping that prefix would lead to more disorientation on the user's side, especially when there are reproductions of the art as INTERIORART under the then seemingly same title. I'd also think that 'Cover' is common enough to be understood by anyone using ISFDB, which requires the knowledge of some English words. Stonecreek 14:53, 12 April 2016 (UTC)

Exactly, Christian! That thought never occurred to me. I can't imagine the chaos that would ensue if we removed the "Cover:" prefix. Mhhutchins|talk 02:52, 13 April 2016 (UTC)

I certainly would also like to be able see the difference between COVERART and INTERIORART variants. However, to make this happen it is, in my opinion, unnecessary that the text "Cover: " be actually part of the database data. I think that it should be the software's job to examine whether it is about to produce output for a COVERART or INTERIORART record, and then dynamically add whatever is necessary for the appropriate visual representation. For instance, on this publication page the software currently outputs this:

interior artwork by Harry Douthwaite (variant of Cover: The Final Programme)

but it should produce this:

interior artwork by Harry Douthwaite (variant of cover artwork The Final Programme)

I don't see how this could be disorienting or confusing to the user. Patrick -- Herzbube ^Talk 10:05, 13 April 2016 (UTC)

Seems like a good idea to me. Stonecreek 12:55, 13 April 2016 (UTC)

I agree. That's how it should be handled. The current "Cover:" prefix is redundant and not necessary because there's already the "title type" COVERART. The software could decide by this title type alone if and how a "cover" label/prefix/info should be displayed. Jens Hitspacebar 18:17, 13 April 2016 (UTC)

Online pub links in title records

Do we encourage links to online publications (e.g., Gutenberg, LibriVox, HathiTrust, Internet Archive, Google Books, etc.) in title records here? I recently attempted to move a LibriVox link from a title record (T94) to the LibriVox-specific pub record (P294078) and I got a submission rejection on the link removal from the title record (see User talk:Mhhutchins#LibriVox links for original discussion). Since we seem to have many such title records I thought I would provide links to these searches:

Notice there are many title records with LibriVox, HathiTrust, and Internet Archive links but only a few with Gutenberg and Google Books links. This is the converse of moving translator, and title transliteration information from pub records to title records (I believe I got dinged on something along those lines a long time ago and there was a discussion on it). I do not see how pub-specific data constitutes title-specific data status (otherwise we would not need any pub records, just giant title records with all the pub data). Of course data from some pubs can be useful in supporting other pub records (e.g., just as some pubs mention printing information for other prints, online scans can support the existence of historic editions and prints) but I do not think that all belongs in our title records (unless they are partially documenting things we do not currently have pub records for, e.g., first edition first print information of obscure non-genre pubs, etc.). I think we should have better division of such data and consider policy on such things. Thanks for your feedback. Uzume 22:45, 8 March 2016 (UTC)

I'm not completely sure I understand your point here, because you sort of wandered between a couple of different points. But I think your question is: "Should an online 'publication' of a work be listed as a separate publication record, or as a web link in the title record?". Assuming that's the question, my preference is the latter. In some cases (e.g. Internet Archives and Google books), the online link is just a photocopied version of a publication record, and hence (IMHO) does not justify a new publication record. It would be logically valid to put a link to that online copy as a link in the Publication Record, but this would be quite awkward for the user who might have to skim dozens of publication records to find the online link they were looking for. Thus for user friendliness, I think having that link be in the title record is to be preferred. (Yes, having it be a separate publication record of type "ebook" would also make it easy to find, but IMHO, it isn't a new publication.) In some cases (esp. Librivox), you can't always tell what "publication" it is being read from, hence there isn't a publication record to which to attach the link. In the case of Librivox, I think there's a better argument to say "this is a new publication", and hence allow it to be created as a publication record. But for most of the other sources (Gutenberg is kind of in-between), they aren't new publications. Thus while I could, logically, support some of these links as being listed in new Pub records and some in links within the Title rec, I think that it would be preferable to handle these kinds of links consistently. Even you listed them all in the same set of parenthesis, implying that you also think they should be handled essentially the same. And to me, since some of these kinds of records should *only* be stored as links in the Title rec, my preference is for all of them to be handled that way. Chavey 21:49, 9 March 2016 (UTC)

While writing the previous note, I contemplate what it would mean for a photocopy / digital copy of a publication to qualify as a *new* publication record itself. Of course one thing is that that copy of an 1873 publication would have to be listed under the year it was first place online, i.e. something *much* later than the original. (We do that with facsimile reprints.) That would be odd. But then I wondered about some fanzines I have that are single sheets of paper. Then when I post a "Cover" for that fanzine, it would contain the entire fanzine, and hence would qualify as a "new publication". So then, logically, I would have to enter a new publication record for that, including its cover. Which would also contain all of that fanzine, which then would logically require a new publication record, which would ... And of course if Zinewiki, Fanlore, Fancyclopedia, or Homeville included a picture of that fanzine's cover, then that would also be a new publication, etc. My conclusion is that we do *not* want to consider direct copies of a publication to be a new publication *unless* the creators specifically marketed it as a reprint (e.g. Khatru #3-4, or various professionally produced facsimile copies). Chavey 21:49, 9 March 2016 (UTC)

I agree with Chavey. If it is just a scan, OCR dump, or other exact copy of some sort of the original record, it is not a new publication unless the publisher indicates it's a reprint (like with this reprint). Since Librivox and Gutenberg (and the others like them) are only duplicating the content, they shouldn't be considered a new publication of the content, except in the instances Chavey mentions. (For example, someone taking a public domain work and releasing an ebook for purchase on Amazon would be considered a new work, but Project Gutenberg simply making the text available on their site for free would not be a new work). ···日本穣 · 投稿 · Talk to Nihonjoe 22:22, 9 March 2016 (UTC)

Librivox & Gutenberg are not the same thing as scans and should definitely be considered new works (as they currently are today). Librivox makes audiobooks from public domain works - that is a completely different format from the original work so I don't see any logical way they could not be considered new works. Gutenberg transcribes public domain works into new editions - often with editing corrections (typos, etc.) - and produces ebooks from them. I also don't see the logic of saying the Gutenberg ebook is not a new edition, but if someone takes the Gutenberg ebook and sells it on Amazon (which is what many of the Amazon public domain works are) it is suddenly a new work. -- JLaTondre (talk) 02:42, 10 March 2016 (UTC)

I am not against developing policy on what constitutes a publication for acquisition/inclusion here (e.g., accepting Gutenberg and LibriVox as publications but not accepting pure scans of earlier now public domain publications such as HathiTrust, etc.) but my point in starting this discussion was more in organizing pub-specific data such as those links within pub records (vs. allowing them to be duplicated and lumped into title records). Uzume 13:53, 14 March 2016 (UTC)

There are several different cases being mixed here and I think it would be best to separate them:

Publication scans: I can see pros and cons. I agree an on-line scan should not be considered a new work. I am ambivalent about whether they should be linked at the title record or publication record (when the original edition can be determined). I can see pros and cons to both. Having it a the title record is probably the most user friendly as it allows people to easily see an on-line scan exists without having to dig through every publication.
Gutenberg: We already include Gutenberg editions as new publications. This should remain the case. I see no reason to have a link to Gutenberg at the title level or manually at the publication level. The ISFDB software automatically links to the applicable Gutenberg page (under "Other Sites") using the catalog number.
Librivox: We already include Librivox editions as new publications. This should remain the case. I see no reason to have a link to Librivox at the title level. I've always had the sense many of these links pre-date the Librivox publications having been added, but not sure about that. Anyhow, they are specific to the Librivox edition so belong with the Librivox publication and not at the title level. It is unfortunate that Librivox does not provide "catalog numbers" to allow easy linking to them (under "Other Sites") the way we do with Gutenberg titles.

-- JLaTondre (talk) 02:42, 10 March 2016 (UTC)

I agree. Though there could be an argument for pub scans to be new pubs, I tend to not want to do that but that was not my reason for opening this discussion. Instead of putting online pub links (be they scans of old original content, reformulated content like Gutenberg and LibriVox, or entirely new content like fanzine material) in the title records, I would rather put them in the pub records they belong to. There is the argument that they could be harder to find but I would argue putting them in the title could also be harder to find (what if I do a search for pub records with author X and publisher LibriVox, will I find the links in the title records?). I would rather put online scan links in the pub records of the original pub (hopefully we would be documenting such) even though there is an argument for making them their own pubs and the other online pub links in their respective pub records (of whatever kind they are). My point is to consider policy towards not putting them in title records. Uzume 04:24, 10 March 2016 (UTC)

If your argument is to remove data, you're not going to win. I have no problem with linking pub records to an online publication of that publication. A link given in the title record is intended to be for the work, not for the publication, so that, in fact, makes the links title-specific and not publication-specific. What actual harm is being done by keeping a link in the title record? Too much data? Mhhutchins|talk 20:50, 10 March 2016 (UTC)

JLaTondre's comments above match my own feelings regarding what should be considered a publication.

Regarding where to link the scans, I feel that they really belong in the publication record for the publication which has been scanned. Having one or two listed in the title record doesn't bother me. However, for some titles, there are dozens of scans of different editions available online. If we were to list them all under the title, we'd end up with a large collection of links with no metadata as to which publication they refer. Even if we augmented the display of links to indicate identifying information, that would result in two competing lists of publications (our pub records, and links to scans). Perhaps limiting links on the title record to the first edition (or the earliest that can be found) would prevent that problem. --Ron ~ Rtrace^Talk 00:06, 11 March 2016 (UTC)

It sounds like it would be beneficial to add a "Web Page" field to publication records. Ahasuerus 00:13, 11 March 2016 (UTC)

That would be helpful. Also, can we add a Notes field to Authors? ···日本穣 · 投稿 · Talk to Nihonjoe 01:05, 11 March 2016 (UTC)

We are in the process of moving all Wiki-based notes to the database proper. Wiki-based "Bio" and "Biblio" pages will be a part of this migration, although we still need to decide whether we want to consolidate them in a single author-specific "Note" field or keep them as two separate fields. Ahasuerus 23:06, 11 March 2016 (UTC)

Awesome. ···日本穣 · 投稿 · Talk to Nihonjoe 01:13, 12 March 2016 (UTC)

I would be against having a web page link in publication records. I can't imagine more than 1% of 1% of the publications recorded in this database would have an online publication. So an empty field for the other 99.99% would be too tempting for editors to link to other webpages, like publisher listings, Amazon (and other retailer) listings, reviews, etc. (New editors are already attempting that with the title record's webpage link.) It would be a bad idea. As for Ron's fear that there would be "dozens of scans of different editions available online", that would be so rare to be no overwhelming problem. The software can handle that many links to a title record. Mhhutchins|talk 01:22, 11 March 2016 (UTC)

I am not against having publication-based web links (despite them perhaps being tempting link spam targets), however, I believe support for other identifiers (and links for those) at the pub level would considerably alleviate much of such needs. Do moderators look at submission link field changes (those currently at author and title level) less than other changes like (e.g., links in) note fields? I would hope not, but not being a moderator I cannot really comment on what parts of that duty is harder than others. Besides being a convenience, dedicated link fields at the pub level would also potentially allow for better control of the links and security. Currently we limit where we allow image linking due to deep-linking bandwidth-theft perception issues. With dedicated link fields we could implement something similar if we wanted. We already mark external links to be opened in separate windows (using target="_blank"; I am against the way this is done unilaterally but the point is there is such control). Also, the current way the open note-field allows generic HTML markup is a potential security issue (sure we have moderation but how many moderators look at the actual markup for hidden things like Javascript injection, etc.?) and link fields could potentially allow us to tighten what we allow in note fields making it eventually be less open and susceptible to such things; (I know changes here are painful due to so many records but there have already been some changes to its interpretation in a number of instances like with regard to line breaks, etc.). Uzume 13:38, 14 March 2016 (UTC)

(un-indent) My concern was not whether the software could handle an arbitrary number of scan links. I believe it could. Rather, I am worried that the user experience would be bad in those situations. Neither was I suggesting a link field be added to the publication record. This is a case where I think the publication notes are a fine place to record the link to the scan. --Ron ~ Rtrace^Talk 03:16, 11 March 2016 (UTC)

I agree with the placement of the link in the Note field. I was responding to Ahasuerus's suggestion that we add a dedicated field to the publication record. A bad user experience by offering multiple links to as many versions which are available? I disagree with that. Again, why limit access to data when it is readily available? Isn't that one of the purposes of the database? To provide as much data as we can and let the user decide how to, well, use it? Mhhutchins|talk 03:44, 11 March 2016 (UTC)

I do not think the issue is limiting access but one of organization. We could put all the data in the entire database on a single large page but despite such "availability" (ignoring for now things like performance), I believe we all agree that would effectively make the data harder to use (even if well organized on that one page) and thus less available not more available. If we instead carefully organize and delimit things, data becomes more available not less. Moving (and not just duplicating) pub-specific links to pub-specific records does not in my opinion delete data but rather organizes it, which is as I see it the major reason for such a database to begin with. I disagree with the concept that links to Gutenberg, LibriVox, and online pub scans are title-specific data. Despite the fact that publications support the concept of works/titles, I consider publication data separately (including links to such). Uzume 13:38, 14 March 2016 (UTC)

Free publications

What's the accepted way to document that a publication is free? The documentation for PublicationFields:Price is silent on the matter, and a quick search in the Wiki didn't turn up any discussions. Thanks, Albinoflea 03:56, 11 March 2016 (UTC)

I don't believe there is a standard. In most cases, I enter a zero price in the currency of the publisher's country, e.g. "$0.00" if it is a free publication. Caveat: never enter "$0.00" in the price field just because there isn't a stated price. You must be certain that it is a free publication before doing that. You also have the option of leaving the field blank, and noting that there is no stated price, regardless of whether it's free or not. Mhhutchins|talk 04:19, 11 March 2016 (UTC)

Entering authors as credited--complex attributions

Last fortnight I added U.S. first editions of the Philippa Fisher trilogy from library copies of the first printings. Temporarily and with a vow to return, I attributed the three COVERART simply to iStockphoto, and put the following list items in the Notes:

Back inside flap credits jacket illustrations (c) 2008 "iStockphoto (flower background)" and "Simon Spoon/iStockphoto (fairy)"
Back inside flap credits jacket illustration (c) 2009 both "iStockphoto (background)" and "Gail Shumway/Getty Images (butterfly)"
Back inside flap credits jacket illustration (c) 2010 "4x6/iStockphoto (fairy)" and "New Vision Technologies Inc./Getty Images (fairy wings)"

Those are verbatim quotations of the Notes items, except underscores added for clarity in this context.

No doubt the two copyright holders for each illustration may be considered its joint authors. Entering those parties exactly as credited in the publications --as underscored above, that is-- four of the six would be new authors. All four that contain slashes; that is, all except "iStockphoto".

Regarding the strings conjoined by slashes, FYI:
Only "iStockphoto" and "Getty Images" are now in the Author directory. Both appear variously, with variety even in the use of backslash with personal names. See the reports of two author searches, search iStockphoto and search Getty Images. None of the other four is now in the directory Simon Spoon, Gail Shumway (not evidently G. Shumway A193976), 4x6, and New Vision Technologies Inc. Nor is "New Vision" now in the directory. --Pwendt|talk 19:50, 26 March 2016 (UTC)

My opinion is that stock image companies like Shutterstock, GettyImages and so on should not be entered as cover artist at all (only in the note, if at all). They don't create any artwork. They are just a service company for the artists, licensing/reselling their works, and as far as I understand the cover artist field it's purpose is not to record who owns the copyright of the artwork. Jens Hitspacebar 20:09, 26 March 2016 (UTC)

Same opinion here. Hauck 20:21, 26 March 2016 (UTC)

Does that mean you would credit the three COVERART to new authors Simon Soon, Gail Shumway, and "New Vision Technologies Inc." respectively --sole authors in place of iStockphoto, from the perspective of the current records, eg P259409 (book 1). No change in the Notes. --Pwendt|talk 20:52, 26 March 2016 (UTC)

The problem is those covers are composite artworks. For example, the first book, the named individual (Simon Spoon) is only responsible for a small portion of the cover (the fairy image). It would not be correct to credit him for the whole cover. -- JLaTondre (talk) 21:00, 26 March 2016 (UTC)

I agree with JLaTondre, but take it a step further. I credit these composite covers to the cover designer when possible. After all, they created the singular work. (I know this isn't the standard verbatim, but sometimes exceptions should be made.) Then I record the separate stated credits in the record's Note field. Mhhutchins|talk 21:34, 26 March 2016 (UTC)

I also enter the cover designer in cases like Mhhutchins describes. ···日本穣 · 投稿 · Talk to Nihonjoe 23:41, 26 March 2016 (UTC)

But iStockphoto is not a even cover designer. Wouldn't an additional cover artist entry using the famous "uncredited" be a better option to get rid of the problem that Simon Spoon gets the credit for the whole artwork? Jens Hitspacebar 01:35, 27 March 2016 (UTC)

I don't think anyone has said iStockPhoto is a cover designer. You're reading way too much into it. In many cases, there will also be someone credited as designing the cover, in addition to the listed credits. As for the original, if no designer was listed, I see no problem crediting Simon Spoon, Gail Shumway, and New Vision Technologies for the cover art as they created the images used to composite the cover. I see iStockPhoto as the distributor/provider of the images, just like Amazon distributes/provides books. They are not the creator of the images. ···日本穣 · 投稿 · Talk to Nihonjoe 15:25, 29 March 2016 (UTC)

We don't enter "uncredited" in the Cover Art field. In the case I described, the field would be left blank if the cover designer isn't credited, and the various cover art credits (all of them, including individual/corporate/collaborative photographer credits) would be entered into the Note field. Mhhutchins|talk 02:57, 27 March 2016 (UTC)

Entry of "uncredited" creates a dated title record that may be a holder for notes. Do we have a COVERART title record without "uncredited" or another generic entry? If not then publication record Notes only? For these books we have cover images from Amazon. In that case my question may be equivalent to whether we have any way to enter Notes on cover images (or Year and perhaps other data on cover images). --Pwendt|talk 22:12, 27 March 2016 (UTC)

Without an artist credit, the system is unable to create a COVERART record for a publication. We either have a COVERART record that credits the artist as "uncredited" or we don't have a COVERART record for notes. (How much more generic can you get than "uncredited"?) The prohibition of using "uncredited" for COVERART records predates my joining this project (February 2007). Although no one has ever told me satisfactorily why this is the standard, I'm not about to push for a change now. Any notes you want to add about an uncredited cover art must be given in the publication record. But you shouldn't use Amazon-linked image as the basis for these notes. Those aren't stable, and you shouldn't be adding such notes to non-primary verified records. Mhhutchins|talk 23:58, 27 March 2016 (UTC)

Note that I'm also completely baffled by the fact that we allow the use of "uncredited" nearly everywhere except for COVERART. As I was called to order for such a violation, I will say no more than that I found this "rule" perfectly counterproductive. Hauck 09:31, 28 March 2016 (UTC)

Another generic value such as "composite" or "corporate" might be permitted for the author field.

(For these three books there is a sense in which either one of those two proposed generic values would be literally true, while "uncredited" would be literally false. That sense depends on the interpretation of copyright holders as authors, however, which is routinely but not universally valid. These books do not provide any information about the jacket designs except by the copyright statements quoted above and in the first printing publication record Notes.)

These three publication records are primary verified, but transient only, so there is no requirement to notify me of revisions. --Pwendt|talk 00:09, 29 March 2016 (UTC)

What about using 'various'? (Which would be a term that is used for texts or conglomerates of texts that are credited to creators that are of no further interest to ISFDB). Stonecreek 15:06, 12 April 2016 (UTC)

Currencies

About the entering of currencies see first the discussion in my talk page titled 'Uncommon Currencies'.

The problem mentioned by Mhhutchins about replacing the former entries of "$" with "USD", "£" with "GPB", and "€" with "EUR", in fact also crossed my mind when i first sugested the ISO code, and it also seems to me, apparently, an imposssible task. I don't know if is an easy way to do it, it maybe possible through some programming, i don't know. However my sugestion primarily tends to alert to the help page about currencies. So, my sugestion for now, is don't mess with the dollars, pounds or euros, but refer in the help page that for other currencies the editors must guide by the ISO code. It's seems to me a much more simple and also democratic way to deal with this issue. And although is an english based database, and i´m glad that it is, once is the most international language and that makes possible for all us to comunicate, it seems to me that it also pretends to be an International database, as the name of the site indicates. So, 'Lit' could mean somethinhg for the italian member who add that designation, or Ft for the hungarians, but not much to me and i believe neither to the great majoraty of the people, the same way any portuguese will recognize the 'ESC' as escudos, but hardly someone for America will be familiar with that, but a simple search in google could clarify that PTE currency corresponds to portuguese escudos. I would like to ear your pros and cons about this matter. Thanks.Wolland 18:58, 8 April 2016 (UTC)

Sorry that I have to object, but to introject any standard from outside seems to me undemocratic, and non-sensical to a certain extent: 'Lit.' for example seems to be used at least for the majority of all pre-Euro Italian publications (as it was the symbol printed on the pub.s), as was DM for (West) German publications. I don't think we should aim for ISO tables as far as we contradict the essence of what was printed on the respective publications. Also, the major (national) bibliographies mostly do report the used national currency symbols. After all, we are a bibliographical site, not a catalogue of international currencies that have to follow the recent currency state-of-art. Stonecreek 19:02, 8 April 2016 (UTC)

The problem is that we, for now, are between the "enter-as-it-is" road or the "normalization" road. We're supposed to enter data "as-it-is" but also we're asked to change the prices as printed but normalizing them (not using a comma or leaving a space). It's this incoherence (that exists for other fields) that is problematic.Hauck 20:40, 8 April 2016 (UTC)

I'd think we aren't on either of these roads but on the normalization-as-we-agreed-upon-and-stick-to-the-printed-prices-as-closely-as-possible road: there seems to be no either-or but the striving for the best of the two. Christian Stonecreek 20:57, 8 April 2016 (UTC)

But my point is why, acording to the rules, a hungarian publication can have is price as 'Ft' without giving the note that it refers to hungarian fiorints, but a portuguese, a french or a japonese publication have to give explanations about the currencie. Is the Yen symbol more unsual than the hungarian fiorint? Has Hungary a more prolific publication than France for instance? My goal here is to make notice that the rules in this matter are not inclusive, and looks randomly to new editors.Wolland 20:46, 8 April 2016 (UTC)

As with authors (like 'Robert Smith') we have to make distinctions between identical names/symbols for currencies. If a symbol/currency shortcut is unique, we don't have to add a distinctional addition to any of it. Stonecreek 20:57, 8 April 2016 (UTC)

The only reason "a hungarian publication can have is price as 'Ft' without giving the note that it refers to hungarian fiorints" is because a moderator (a human, not necessarily an English-speaking one) excepted the submission without asking the editor to add the note.

I resent your saying that the ISFDB "pretends to be an International database". The "I" in the name stands for "Internet". The ISFDB accepts submissions for eligible publications regardless of their language or country of publication. That's it. Nothing more. No other claims implied or pretended, regardless of how such statements may be perceived. And with that being said, the ISFDB is the most inclusive database of its kind in existence. There is nothing else in the world that comes even close to being as inclusive as the ISFDB. Mhhutchins|talk 21:08, 8 April 2016 (UTC)

I didn't say nothinhg about the ISFDB not being inclusive. I only refered the rule as it is doesn't seems inclusive in that partucular matter. As new editors, like me, could find strange why the rules refer to a so restrict number of contries. So, should i state the price as it is in the publications and clarify it the notes, or find a standard to this particular currency(portuguese escudos) and use it all the time is necessary?Wolland 21:37, 8 April 2016 (UTC)

If you only wanted to establish a standard for entering Portuguese currency, that's all you had to say. There was no need to attack the database because you felt that a certain currency was being slighted. As I said before, if you have found records for "uncommon" currencies without a note of explanation, it was simply an oversight by either the editor or the moderator, either one of who may not have been aware of the policy. There was no intention to treat some "uncommon" currencies differently than others.

I never had the intent of atacking anyone or anything. I just stated my opinion that i find this particular rule inconclusive and just open the discussion to a sugestion and presented my aguments. That's all. I hardly see how it can be offensive. But for some how it sounds like that i apolagize to anyone who felt it in that way.Wolland 01:00, 9 April 2016 (UTC)

So let's start over again. How do you think publications priced in Portuguese currency should be entered into the database? Mhhutchins|talk 22:26, 8 April 2016 (UTC)

As the currency symbol of escudos is the same as the dollar ($) my sugestion is to enter like this: 40$50 esc., as this is the usual form used, the '$' separates the 'escudos' from the 'centavos' (cents), it works like a decimal separator. If placing the '$' as a decimal separator represents a problem, i would sugest like this 40.50 esc. or if decimal places equal to 0 just 40 esc. And if a diferent printing appear i will mention in the note field.Wolland 01:00, 9 April 2016 (UTC)

There may be a translation issue here. In Portuguese, the verb "pretender" means "to intend", "to aspire", etc, while in English "to pretend" implies "false pretenses". Ahasuerus 23:18, 8 April 2016 (UTC)

You are completely right. And now i see how it can be offensive. My sincerely apolagize to all. Fortunaly we aren't world leaders, it could be a diplomatic incident that could end in world war! :). Once again my apolagizes. Thanks to clear it out Ahasuerus.Wolland 01:00, 9 April 2016 (UTC)

No worries, it happens fairly often. The most (in)famous example I can think of is the difference between the UK usage of the verb "to table" (to propose to start a discussion) and the US usage (to postpone a discussion). And then there is the colloquial expression "to knock someone up" :-\ Ahasuerus 15:49, 9 April 2016 (UTC)

Currencies: Improve standard for East German Mark and Deutschmark

Seeing that there's a discussion about currency standards right now above, it just occured to me that the standard for German currencies could be improved. Currently, only West German Deutschmarks are mentioned in the help. East German Marks should also be mentioned because they used a different currency letter. Suggestion:

Enter West German (and after German reunification German) Deutsche Marks as "DM", for example "DM 7.80".
Enter East German Marks as "M", for example "M 7.80".

Jens Hitspacebar 23:09, 8 April 2016 (UTC)

Yep, that would fit. Stonecreek 06:49, 9 April 2016 (UTC)

I just realized that there were three currencies in East Germany actually (see Wikipedia) which were called differently and used different currency letters over time. Does the ISFDB want to take these details into account? Here's what the rules could look like then:

Enter East German Marks (also called "Ostmark") depending on the publication date, because it had three different currencies over time,
- as "DM" for publications released between 1948-07-24 and 1964-07-31, for example "DM 7.80"
- as "MDN"for publications released between 1964-08-01 and 1967-12-31, for example "MDN 7.80"
- as "M" for publications released between 1968-01-01 and 1990-06-30, for example "M 7.80"

Jens Hitspacebar 08:01, 9 April 2016 (UTC)

I think that this would lead to more chaos. There doesn't seem to have been different currencies in the GDR, only the official naming has changed over time. As said elsewhere, we shouldn't try to represent the state-of-art in currency titling but try to have clean separattion between currencies (and our representation of them). Thus it seems that M for East Germany and DM for West Germany do represent clear distinctions between the two currencies. Stonecreek 15:15, 12 April 2016 (UTC)

I agree with Christian. Rudam 17:04, 12 April 2016 (UTC)

Ok, sounds reasonable. If there's no further objection or other comment I'd add my initial suggestion to the help page. Jens Hitspacebar 18:23, 13 April 2016 (UTC)

Clarifying SERIAL subtitles (was: Serialized stories with non-English titles)

(copied from the Help Desk)

I need to convert some stories from the Polish language magazine "Fenix" from short fiction to Serial. In general, these serialized stories have formal titles of the form "Ciężki bój (cz. 1)". My inclination (including our English-preference approach to comments and documentation) is to change that title to "Ciężki bój (part 1 of 2)", but a reading of the Serial Help page implies that since the original title is unique, I should leave it as is. What is the preferred approach to this? And would the answer be different if the original story language was English (as in this example story by Greg Bear) or Polish (such as "Egaheer (cz. III)")? Chavey 13:21, 23 April 2016 (UTC)

I see our "(part X or Y)" convention as a way to disambiguate SERIAL titles. If a SERIAL title was disambiguated by the editors, then there is no need for additional disambiguation, which is why we have this moderator-only cleanup report. Since "(cz. 1)" is already part of this SERIAL title, I think it should be entered "as is". Ahasuerus 17:55, 23 April 2016 (UTC)

Thanks. I'll add that to the Serial Help page. Chavey 13:01, 24 April 2016 (UTC)

I disagree with this change and believe it mistakenly interprets the standard. If you look at almost any novel that has been serialized in periodicals, you'll find editorial titles are almost always added in the publication to distinguish the parts in the various issues. It's been ISFDB policy to use the author's title, and add the ISFDB standard for disambiguating parts of a serial, thus ignoring the editorial disambiguation. Pull out a copy of the nearest publication that has a serial and see how it's titled. There are multiple editorial methods being used to distinguish a serial's parts. That's why the ISFDB came up with a single standard for all serializations. (For example, the serialization of Venus on the Half-Shell in this publication is editorially subtitled "2nd of 2 parts".) So the serialization of a work titled "Ciężki bój" would be entered into the ISFDB as "Ciężki bój (Part 1 of 2)", etc. [note the capitalization of "Part"]. This method is used regardless of the language, e.g. this record. I believe the purpose of the Help section (which was recently updated) was for serials in which each part of the serial had a distinctive title, thus making it unnecessary to add the "(Part 1 of X)" disambiguation. Mhhutchins|talk 18:46, 24 April 2016 (UTC)

I agree with Michael, Druga śmierć Thery (cz. I) is less clear for the average (non Polish locutor in this case) user (including me of course) than Druga śmierć Thery (Part 1 of 2). Hauck 06:47, 25 April 2016 (UTC)

Re-reading the linked Help text, it looks like it comes down to how we want to interpret the word "unique". The example used in the previous version of Help, "Butterflies in the Kremlin, Part Eight: As the Bear Turns", was clearly unique, so we entered it "as is". To use another example, "Part 1" wouldn't be unique, so we would enter it as "Part 1 of [X]". "2nd of 2 parts", which Michael references above, is currently treated as not unique, so it's listed as "Part 2 of 2".

I guess the question that we need to answer is where to draw the line between "unique" and "not unique". Perhaps we should replace the "unique" clause with something more explicit, something that would basically say:

If the SERIAL title contains a subtitle and the subtitle is some form of disambiguation that does not contain additional substantive (?) information, replace the subtitle with (Part X of Y).

Does this sound about right? Ahasuerus 21:19, 25 April 2016 (UTC)

Sounds good to me, as long as we make it clear that all titles, regardless of their language, are handled the same way. Mhhutchins|talk 22:45, 25 April 2016 (UTC)

(unindent after copying from the Help Desk) One thing that we should probably consider in this context is what we want to do when some serial installments have a "unique" subtitle and some do not. Butterflies in the Kremlin is a good example. Here is how the installment titles are currently entered:

Butterflies in the Kremlin, Part 1: A Russian Noble
Butterflies in the Kremlin, Episode 2: A 'Merican in Moscow
Butterflies in the Kremlin, Part 3: Boris, Natasha ... But Where's Bullwinkle
Butterflies in the Kremlin, Part Four
Butterflies in the Kremlin, Part Five, The Dog and Pony Show
Butterflies in the Kremlin, Part Six: The Polish Incident or The Wet Firecracker War
Butterflies in the Kremlin, Part Seven, The Bureaucrats are Revolting
Butterflies in the Kremlin, Part Eight: As the Bear Turns

Parts 1-3 and 5-8 are clearly unique, so we will keep their subtitles. But what about "Part Four"? Should we change it to "(Part 4 of 8)" or should we leave it as is? Ahasuerus 16:21, 27 April 2016 (UTC)

It seems to me we should either always add "(Part m of n)" or only add it when there is no "part" indicator. Uniqueness aside, there's some value in understanding the order of the serial parts, without having to look up the dates. A sub-question is: If we do always add it, do we replace any "existing" part indicator? My feeling is leave anything with a natural "part" indication in it alone, otherwise add "(Part m of n)" before any subtitle. But putting "(Part m of n)" on everything seems ok to me, too (the "of n" is useful info). It's a little clunky, but I think it would look ok, even in the "Butterflies" example if we put it before any delimiter and subtitle. --MartyD 01:27, 28 April 2016 (UTC)

Abyss & Apex

I haven't been paying as much attention to recent goings-on as I should have been doing, but has something changed that would allow Abyss & Apex to be included in the database? Nobody has more love and respect for that magazine than I do, but it's a semiprofessional webzine that's very much ineligible for SFWA. I just thought I'd ask if some exception had been made before I went around challenging people's contributions... Dwarzel 17:04, 16 July 2016 (UTC)

I think it's IN under the "It has been short-listed for a major award." exception, if this means "short-listed", that is. --MartyD 22:08, 17 July 2016 (UTC)

Sounds about right. Also, I have been thinking that our current inclusion rules, especially references to SFWA, may be too NorAm-centric given our increased coverage of non-English titles, but that's fodder for another discussion. Ahasuerus 23:16, 17 July 2016 (UTC)

Good point; I'd forgotten that it was indeed on the final Hugo ballot last year, but I see it now. Would that make the entire run of the magazine eligible for inclusion, or just the year's-worth that was considered for the Hugo? Dwarzel 16:55, 19 July 2016 (UTC)

I would say it gets the whole thing in. It would be silly (and confusing) to have it only get the one year in. ···日本穣 · 投稿 · Talk to Nihonjoe 17:02, 19 July 2016 (UTC)

Titles of Barrie excerpts

I would like to link to a discussion on MLB's page about the titles that were used for excerpted novel chapters in Masterpieces of Fantasy and Wonder, about whether they should have (excerpt) after them or not. Anyone have thoughts? --Vasha77 17:17, 4 September 2016 (UTC)

My opinion: We use parenthetical information to disambiguate, not to explain what something is. The reason to add "(excerpt)" is to distinguish between the title of the excerpt and the title of the full work. You should add it if the title given in the publication is the same as the title used for the work from which the excerpt is taken. The same thinking would apply to a sub-part (for example, if a chapter's title is used, include "(excerpt)" if the excerpt is only part of the chapter, but omit it if the except is the full chapter). --MartyD 19:35, 5 September 2016 (UTC)

Thank you, sounds like we should change those titles. --Vasha77 02:43, 6 September 2016 (UTC)

I think that I said most of this, still, it's good it's been codified. While my hard drive got sick and had to be fixed I found my copy of this book. Guess I shouldn't (***sob***) have bothered. Still, if you're interested, the introductory blurb states that these are selected chapters from the novel The Little White Bird. I've never read this book, I've never owned this book, and I know nothing about this novel, still I hope that this bit'o'info helps. MLB 23:58, 8 September 2016 (UTC)

Graphic novels

I've gone through the history of discussions on graphic novels. (Specifically here seems relevant). There doesn't seem to be a standard approach for labeling them, and that is leading to a fair amount of confusion on my part. Many of them are designated chapbooks and others are novels. I recently entered this pub and that highlights a fair amount of confusion. It's quite similar to both this pub which is chapbook and this pub which is a novel. In each case, the stories were initially printed as comics in a monthly format, but only as a limited series. I wanted to add Free Country to the the DC Metaverse series but could not since it's a chapbook. The publication has three distinct 'chapters' with different authors. Should this be a single Chapbook with 7 authors, or should I break it into individual stories each with unique authors. Any clarifications would be appreciated. TAWeiss 21:57, 5 September 2016 (UTC)

Alas this won't help but, AFAIC, there should be strictly NO graphic novels (even by authors "above the threshold") in the db, lest we face such problems and/or an invasion of publications we're not equiped to deal with. Hauck 07:29, 6 September 2016 (UTC)

Amen to that. As for the problem case, I think for any book with different chapters by different people, I'd be inclined to make the publication an ANTHOLOGY and the chapters separate shortfiction works. --MartyD 10:56, 6 September 2016 (UTC)

There are several graphic novels in the db and there have been a number of discussions on how to enter them correctly. I've seen moderators who have entered this content in the database such as this pub. There was specific discussion about authors above the threshold here. I guess my confusion is shared by others.TAWeiss 00:51, 8 September 2016 (UTC)

Regarding adding chapbooks to series, you instead add the chapbook content title record(s) to the series (in this case, add the four short fiction titles to the series). As MartyD mentioned, that chapbook would be better changed to an anthology. ···日本穣 · 投稿 · Talk to Nihonjoe 01:12, 8 September 2016 (UTC)

I agree that the anthology approach would work better in this case. Chapbooks are generally reserved for standalone appearances of short fiction and poems.

As far as the issue of eligibility goes, it has a long and complicated history. In the mid-2000s, some editors supported their inclusion while other editors were opposed. There was some back-and-forth on this issue and the result was a compromise which allowed a limited subset of graphic novels to be included. A couple of years ago the software was upgraded to make it easier to distinguish between prose novels and graphic novel, but the policy issue remains, to a large extent, in limbo.

We'll probably have to revisit the policy issue sooner rather than later because the number of multi-media franchises is growing. A number of popular prose series have had graphic adaptation lately while many manga series have "light novel" offshoots. I watch the numbers climb month after month since I process robotic submissions. Ahasuerus 01:22, 8 September 2016 (UTC)

Upon reflection, it seems more a case of collaborative fiction, where a number of authors have created a single story. None of the chapters stand on their own as a story. If Chapbook is the wrong designation, then I would think it falls in the novel category, which matches most of the other graphic novels that I've seen. It also seems like I should hold off on entering any further graphic novels until there is further clarification, though it leaves some odd holes in the database. There are a large number of books with considerable graphic content, which makes exclusion a difficult judgement callTAWeiss 01:35, 8 September 2016 (UTC)

Variant title dates

These statements are not direct contradictions because the first gives instruction (as befits EditTitle) and the second interprets what is displayed if instructions have been followed (as befits Title). But the second presumes a directly contradictory convention, as I understand it.

Help:Screen:EditTitle > Title: Year

When entering a variant title, enter the first date when the work appeared under any title and any pseudonym; typically, variant titles do not have separate dates.

Help:Screen:Title > Variant titles

If the title/author combination displayed is the canonical version of the title, then any variant titles will be displayed. This will list the variant title, the date of first publication under that variant title, and the author name under which that variant title appeared.

Both conventions are commonly followed in regions of the database familiar to me. I know that I have followed each one sometimes.

Concerning artwork, there may be a related point concerning publication as INTERIORART on the one hand, COVERART on the other hand. In the spirit of the EditTitle instruction, a cover illustration may be dated by its first publication as an interior illustration. --Pwendt|talk 02:14, 9 September 2016 (UTC)

It is my understanding that in practice each variant title has the date that the work first appeared with that title and author and language. This matches the second explanation. Further, the canonical title record should always have the date of first publication under any title. I'm not sure that I see the point in having the dates of the canonical and all variant titles match. --Ron ~ Rtrace^Talk 02:24, 9 September 2016 (UTC)

Our documentation and our standard practice don't agree on this point. In the past, variants were always dated with the date of the parents. Lately, more and more editors are choosing to date the variant with the first date the variant was used. This has been raised before, but while there were no objections to updating the documentation, there was not much participation in the discussion so nothing was changed. Personally, I think the documentation should be updated to match practice (especially given translations). -- JLaTondre (talk) 02:29, 9 September 2016 (UTC)

I agree with JLaTondre. This makes the database more accurate. ···日本穣 · 投稿 · Talk to Nihonjoe 02:33, 9 September 2016 (UTC)

Reading #Titles of Barrie excerpts, it occurs to me that the same issue concerns titles such as "The Little White Bird (excerpt)" and "Through the Looking-Glass and What Alice Found There [3]". First publication of the work as an excerpt, perhaps enumerated, or its first publication as part of a larger whole. --Pwendt|talk 02:41, 9 September 2016 (UTC)

Some questions about the EditPub titles for magazines

In this help entry for Publications, the Under Titles - Magazine it states "the title should be of the form Magazine Title, Date" and to see the missing and variant dates for more on formatting. That sections has "The date part of a magazine title should be given after the title, following a comma and a space. The month should be given in full and then the year in full. If the issue is a quarterly, or a bimonthly, give the date in the form given on the magazine -- for example, "Fantastic Universe, June-July 1953" or "Interzone, Fall 1979". A hyphen should be used between two months used for a bimonthly issue."

A conflict arises between "give the date in the form given on the magazine" and "A hyphen should be used", if the magazine uses a slash (/). Which is to be given priority? And should be blanks be used to separate the hyphen from the month if the magazine does so?

The second concern is which date to use - in some cases the cover and the contents page differ. The example I have is the "July/August 1992" cover of Science Fiction Chronicle which has "July 1992/Volume 13, Number 10/Issn 0195-5365/Whole #153" leading the table of contents. Which should be given priority?

Combining the two, is issue with cover "December 95/January 1996" and contents of "Issue #187 / December 1995-January 1996 / Volume 17, Number 2 / ISSN 0195-5365".

A final point - a number of magazines have been entered (e.g. Locus) using the Series number as well as the date. Is this acceptable or should it be discouraged? Doug H 16:35, 17 September 2016 (UTC)

A partial answer, with much debate and the apparition of non-US magazines (that are usually sequantially numbered), the series number is used when it's proeminently displayed on the cover (for example Interzone) but not if it's not on cover (for example Asimov's even if each issue has a whole number on toc). Hauck 17:53, 17 September 2016 (UTC)

The competitor magazine for Science Fiction Chronicle was Locus. It includes the series number even though it is not on the cover. There are no images for the first few issues but back in 1980 they had the issue and date on the cover. The existing issues of SFC seem to have been modelled on Locus. The first 60 issues of SFC are not loaded, so I don't know whether they had the issue number or not. I'd prefer to retain the issue number for various reasons (all 60 done this way, with at least 160 unsubmitted issues), gaps are easier to identify - 192 is June 1997, 193 is October 1997), but don't know if (or how, or to whom) I should be arguing the exception. Once these are resolved, maybe we can update the help page. Doug H 19:11, 17 September 2016 (UTC)

Fictional Authors of Fictional Essays

This template states that there is an exception to the rule of recording the author's name as it appears on the title page: "An exception is made for fictional essays, which are written as if by a character in the story, often as an introduction or afterword. Even if these are signed by the fictional character, they should be recorded as by the actual author of the work." I've received a request to change the author of Quidditch Through the Ages from J.K. Rowling to the pseudonym "Kennilworthy Whisp" which is how the book is credited on the title page. The argument is being made that the above exception only applies to true essays, and that essays that we list as short fiction (as is done here) should be credited to the fictional author. My reading of the above exception is that it is there exactly for this kind of work and that the variant titles already created as by Kennilworthy Whisp should be changed to Rowling. Do others have opinions of whether this book is an appropriate application of the exception about fictional authors? Thanks. --Ron ~ Rtrace^Talk 02:26, 29 September 2016 (UTC)

I do think that this really is a kind of shortfiction as it tells the story of 'quidditch' through the ages. Also, I believe the dissinction between 'fictional essays' and shortfiction a futile one as it can become really tedious to draw a line between the two: where exactly should there be a border established? At the least all this pieces tell a kind of (in-universe) story, however compressed it may be (and this can vary in many degrees). Stonecreek 03:41, 29 September 2016 (UTC)

What author name would people look for the story/book under? If there is a chance that they look for it under Kennilworthy Whisp, that should be the name associated with it. And where would it be filed - under R or under W in a library or in a bookstore (my library put a call number of 796.1 WHISP so if it is just under Rollins, I may decide that the db does not have it). I'd go for the author as written when it is an individual book and not a part of a longer book - even if it is a chapbook. May be wrong but I just do not think that this rule applies here.:) Anniemod 03:58, 29 September 2016 (UTC)

The distinction in my mind is that here, it is a publication credited to the fictional character. The exception might be applicable to the work if it appeared in conjunction with a novel credited to Rowling -- and I believe the spirit of the exception covers this example in such a case -- but here it is a standalone work, so it's not "covered" by any primary work's credit to the real person. --MartyD 10:35, 29 September 2016 (UTC)

I think it would be fine to treat it as short fiction by Whisp, and then make Kennilworthy Whisp a pseudonym of Rowling. That way, people will be able to easily find it. ···日本穣 · 投稿 · Talk to Nihonjoe 23:48, 29 September 2016 (UTC)

"uncredited" publisher

The other day I discovered that about a dozen publications, half of them verified, list their publisher as "uncredited". I don't think it was ever intended, but all that Help says about this issue is:

For self-published works, fanzines, bibliographic pamphlets and the like, use the name of the editor/author if no other publisher information is visible.

I think it would be beneficial to add the following sentence to the quoted paragraph:

In all other cases where the publisher is not stated, leave the field blank.

Ahasuerus 00:51, 3 October 2016 (UTC)

Looks like a useful addition. But in this case, would the publisher's name appear as "unknown" (as in the cover art field) or as "uncredited" (which would be more appropriate, in my opinion) ? Linguist 08:00, 3 October 2016 (UTC).

I don't know how frequently we run into secondary sources that do not list the publisher, but it does seem the distinction of "uncredited" vs. "unknown" is useful there. I'd rather see an explicit "uncredited" for cases where a primary verifier knows for a fact there is no publisher credit of any kind and leave blank for the cases where we're working with sources from which we cannot tell who the publisher is. --MartyD 10:59, 3 October 2016 (UTC)

ElfQuest graphic novels

I was told some time back NOT to add the ElfQuest graphic novels, but someone has added a whole bunch of them (but not all)

As far as I can figure the prose shorts (the 5 anthologies under the Blood of Ten Chiefs titles) and the novelizations plus the two chapbooks with "new" material (A Gift of Her Own and Searcher and the Sword) and maybe the Gatherums (a, b and the big gatherum) plus the other non-fic stuff (The First 20 Years) should be here but not the other stuff and not the comic books by Wendy for Beauty and the Beast (Night of Beauty and Portrait of Love). How to get rid of:

(I think I got them all) Susan O'Fearna 03:29, 24 October 2016 (UTC)

If I recall correctly, the ElfQuest universe was one of Michael Hutchins's projects and he spent a fair amount of time sorting everything out. I don't recall why he chose to keep the graphic novels, but I assume he had a reason. We probably don't want to delete them until we hear from him. Ahasuerus 15:16, 24 October 2016 (UTC)

Sounds like a plan to me... I just checked my archive (!!) and he's one of the mods who told me original graphics can be added but with rare exceptions (Sandman) collected TPBs (as opposed to gns) are not to be added Susan O'Fearna 17:23, 24 October 2016 (UTC)

I'd say "delete the lot". Hauck 17:27, 24 October 2016 (UTC)

As this is foremost a series of titles with graphic formats I can't see a reason why those should stay. Else we would also have to add all kinds of comics whose universes happen to have an incarnation as one or more prose titles (for example super heroes, Asterix etc.). Stonecreek 11:12, 3 November 2016 (UTC)

Note that this series is now a complete mess. I don't how many changes about it I've moderated (e.g. titles going from NOVEL to CHAPBOOK and back to NOVEL). There was also a lot of work in the cleanup reports (lots of novels with twice the same title included). There is still the matter of the duplicates # in the title series, can someone attend to this before I delete the whole lot? Hauck 09:19, 3 November 2016 (UTC)

If there are no objections forthcoming, I'm going to delete this series in a few days. Hauck 09:30, 6 November 2016 (UTC)

We'll just need to be careful to preserve the (relatively few) prose works that are part of the series. Ahasuerus 12:47, 6 November 2016 (UTC)

No problem. Hauck 13:33, 6 November 2016 (UTC)

Done. Hauck 09:06, 7 November 2016 (UTC)

Pseudonym direction

I need a rules check for a case. I would usually go with Tupper as the main name and Bob Tupper as pseudonym due to number of works involved under the two authors. However - it is a "just family name" against a fuller name so I am kinda inclined to go the other direction and just variant the whole lot... What do the rules say in this case? What is the correct way to pseudonym these two? Anniemod 01:17, 25 October 2016 (UTC)

In our world, "canonical" is not defined to be "most complete", but rather to be "most commonly used". See Help:How to record a pseudonym. --MartyD 11:24, 26 October 2016 (UTC)

Yeah, and I had been following the "most common" rule. That one gave me a pause as the last name only is the common one. Overthinking it (again) I guess. :) I will get it connected the usual way - it can always be reversed of needed one day. Thanks!Anniemod 20:15, 26 October 2016 (UTC)

Tami Hoag

I just noticed there's one novel by Tami Hoag on here... all of her novels, including this one Night Sins are romantic suspense/mystery with no SpecFic... delete it? There are no verifications Susan O'Fearna 05:24, 26 October 2016 (UTC)

OK, done. Hauck 05:39, 26 October 2016 (UTC)

Essay in non-genre book

I apologize for sounding like a total noob. Hauck suggested that I post here after having rejected this submission, which is a book of literary essays including one about Frankenstein -- that was the only regular-title content I included. This database doesn't seem to have a good way of of adding short stories or essays that are in non-genre publications; I hate to clutter up the place with books like The Educational Legacy of Romanticism, but shouldn't the essay be in the DB? What is the opinion of people here? (Would it be possible to just keep the title record for the essay, and have its publication info in a note?) --Vasha77 18:31, 30 October 2016 (UTC)

Not much to add to my rejection message here. If we start to go this way, we will have to accept any essay "about" spec-fic in any publication. I'm personally against such a move that will likely swamp the db. Hauck 18:52, 30 October 2016 (UTC)

I am with Hauck on that - if it is not in a genre media and it is non-fiction, we do not catalog. Otherwise we will end up cataloging articles in TLS or Reader's Digest because they happen to talk about something sf-inal. If it was fiction, what you submitted would be the proper way (we have Nature magazine cataloged like that for example - with only the fiction that is SF) but non-fiction (especially by non-genre authors) just does not belong here. Otherwise we are also opening the doors for sf art in non-SF publications (is non-fiction more important than art?) and all kinds of other borderline cases. At least until there is a different way to separate that kind of content, it clutters the DB. And fiction is what this DB is all about (name kinda says so as well) - we catalog art and non-fiction as supporting elements in my opinion. Just my 2 cents :) Anniemod 22:01, 30 October 2016 (UTC)

It seems to me that there should be room for literary criticism, even if it is in a non-genre publication. Art, no. It's just a matter of defining the line. Unfortunately, most of our rules of acquisition have fuzzy boundaries. For example, debates about the inclusion of graphic novels and letters have been going on for as long as I've been a member. Part of the problem is that the ISFDB is like a monastery, we're mostly closed off to feedback from the outside. I think we need to take a bigger view. The question is how do people use the ISFDB, and what would they like to get out of it? Do we serve only the community of SF readers? Biographers would certainly appreciate references to SF criticism, and authors published outside of the genre. I suggested adding a gender identification field a while back, and that really stirred things up, despite the fact that every published biographical author listing, Encyclopedias,Contemporary Authors, Austlit, etc., identifies the gender of authors. Maybe it's time to update the rules of acquisition with outside feedback and draw as many hard lines as possible?--Rkihara 22:32, 30 October 2016 (UTC)

IIRC this (references to SF criticism, and authors published outside of the genre) approach has been tried once.Hauck 08:09, 31 October 2016 (UTC)

As long as we draw a strict line between criticism and "this article that talks about SF-inal things/books" by someone that had never written anything else about SF... And drawing this line won't be that easy and will depend a lot on what someone is looking for - what I consider criticism may be stricter or less strict than what someone else does. Just thinking aloud. Anniemod 01:11, 31 October 2016 (UTC)

Sharp lines? As if. (To use an Americanism.) Hauck also objected to me adding literary criticism that wasn't in his opinion genre enough. Jorge Luis Borges wrote stories which in his own words "pertencen al género fantástico" and one of which was a finalist for a Retro Hugo last year. It could be argued that the "Ficciones" don't belong to genre SFF as English speakers understand it, since they weren't originally published in that context, but if Hugo voter retrospectively accept them, why disagree? Likewise most of the vast amount of criticism written about those stories isn't written by people interested in genre fantasy, but rather who are interested in epistemology, semiotics, history of literature, etc. Should criticism coming from a non-genre angle be omitted even when it's about genre texts? What about texts which, like "Tlön, Uqbar, Orbis Tertius" have been accepted into the genre but only in hindsight, not when first published? There's no easy answer to that. --Vasha77 02:26, 31 October 2016 (UTC)

About this question "Should criticism coming from a non-genre angle be omitted even when it's about genre texts?", my answer is a clear "No" (see this page where all these books have been entered in the db). I've objected (but sometimes approved them) to some of your proposed additions because a book like this one is IMHO (I never forget that it's in the end a subjective choice) outside our scope as Borges is not above the threshold. For the enough part, as I said I tend to follow a 50% rule (more than half about spec-fic, it's in, less, it's out), it's not perfect but it's a first approach. Hauck 07:32, 31 October 2016 (UTC)

[Unindent] We do our best to be completists when it comes to literary fiction. We do NOT try to do that with non-literary fiction (e.g. movies, comics, fanfic, or graphic novels) nor with SFF art or SFF non-fic, and I think that's a good policy. IMHO, we should only include art books that are entirely speculative fiction or non-fic books that are entirely analysis of, or critique of, speculative fiction. (So Hauck's "50% rule" is much more generous than me.) Partly, we don't need to try to compete with sources like The Science Fiction and Fantasy Research Database, whose goal it is to catalog such non-fiction. And, IMHO, we should try to limit ourselves because we can't really keep up with the fiction by itself!

There is *so* much fiction out there that we still haven't been able to incorporate (especially after going "International") that we seriously need to maintain substantial boundaries on what we try to catalog. I just discovered a publisher we had ignored, who had 60-some books we needed to add. There are thousands of fanzines, many of which have full contents online, that haven't been entered. Half the books in the db have additional OCLC records that we haven't pulled in. WorldCat has thousands of "Title synopses" that we haven't pulled in, and Amazon has thousands more. Most of the big reference guides (Tuck, Schlobin, Reginald, Bleiler, Currey, etc.) have been partly, but not completely entered. There are hundreds, if not thousands, of anthologies for which we have no contents, even though those contents are in WorldCat. There are ebooks, self-published books, non-English works, .... The challenges we have with the range of work we have established are so daunting that it should be clear that we cannot do a respectable job on any attempt to massively expand our mission. IMO, we are better off to keep a narrower focus, and try to do it right, then to establish an "include everything relevant" policy and do a terrible job at it. Chavey 14:54, 31 October 2016 (UTC)

We are volunteers, and if one us decides to enter something "relevant" it doesn't mean everyone else has to. It doesn't become policy because a few people do it. Being above the threshold for inclusion of non-SF books, can be pretty subjective as Hauck points out. In regard to Borges, I've read most of his translated collections and in my opinion he's well over the 50% mark.--Rkihara 15:40, 31 October 2016 (UTC)

Chavey may have a good point, and if other people agree that the DB is for fiction, then I would switch over to just adding contents to anthologies and collections -- I enjoy doing that. It can't be denied that the Cornell University library (which I'm using) is a better resource for criticism than for genre fiction; but it does have enough fiction to keep me busy for a long time. --Vasha77 20:12, 31 October 2016 (UTC)

I agree that we should be extra careful when deciding to expand the scope of the project. As we have learned over the last 20+ years, it's very important to have a good understanding of the additional complexities and the extra volume of work that scope expansion may lead to. Ahasuerus 00:56, 1 November 2016 (UTC)

Acts of Light

What is a collection of illustrated Emily Dickinson poems, Acts of Light(not verified) doing here? I don't see a genre connection at all. Yeah, one or two of Dickinson's poems mention ghosts or witches, but... --Vasha77 20:56, 30 October 2016 (UTC)

Someone cataloged it or it got pulled by the automatic processes based on Amazon tags and somehow slipped in:) I think it should be deleted Anniemod 22:03, 30 October 2016 (UTC)

Done. Hauck 07:21, 31 October 2016 (UTC)

Disappointed!

Why must we IGNORE "best practices?"

It's been said that:

"1) a copy (be it made of electrons or of paper) of a magazine is not itself a MAGAZINE (for us), it's transformed into a COLLECTION.

Yet our very own Template for Pub Type informs us quite sensibly:

COLLECTION. A publication containing two or more works of SHORTFICTION or POEMs by a single author or authors writing in collaboration should be typed as a COLLECTION. The typing of individual publications which contain works with various combinations of author credit should be discussed on the Community Portal on a case-by-case basis. The title page credit should be the major factor in determining the types of these kinds of publications. Excerpts from other works published after a NOVEL for promotional purposes do not make the publication into a COLLECTION.

It's plain to see that our helpful template makes very clear that the typical MAGAZINE with several authors writing individually, is most definitely NOT a collection!

Perhaps we might seek guidance again from a further Template bulleted item:

MAGAZINE. It can be difficult in some cases to determine if something should be regarded as a magazine or a book. Some magazines were published in book format; some books were published as series with letter columns and regular dates of publication. Borderline cases should be discussed on the magazine or book wiki pages, but generally a magazine must have a common title from issue to issue, and an enumeration or dating system of some kind. This still leaves anthology series such as New Worlds Quarterly as judgement calls. In these cases, look for a consensus on the publication bibliographic wiki page. If no discussion exists, use your best judgment and document the decision on the wiki page.

So referring back to the above assertion that a reprinted magazine is no longer a magazine, but a collection-

How is this justified without flagrantly disregarding our own guidelines?
Chrisgherr 11:32, 1 November 2016 (UTC)

There's an error involved in the discussion: most often a reprinted magazine becomes an ANTHOLOGY, not a collection (and that is just what happened in the case that started the upheaval, I think). The title in question is stated to be an ANTHOLOGY, and that's what its respective publications are listed. Stonecreek 11:53, 1 November 2016 (UTC)

Yep, see below (my mistake) all the rest stands.Hauck 12:32, 1 November 2016 (UTC)

Instead of COLLECTION it should be ANTHOLOGY (an artefact of translation), the source of the discussion is in these rejections. Hauck 11:52, 1 November 2016 (UTC)

So after correcting for the faulty translation, we have:

1) a copy (be it made of electrons or of paper) of a magazine is not itself a MAGAZINE (for us), it's transformed into an ANTHOLOGY.

Again referring to our Pub Type template we see:

ANTHOLOGY: A publication containing fiction by more than one author, not written in collaboration, should be typed as an ANTHOLOGY. For example, "Late Knight Edition" contains stories by both Damon Knight and Kate Wilhelm, individually; this is an anthology, not a collection. If a book of Conan stories contains stories which are all partly or wholly by Robert E. Howard, it is a collection; if one or more of the stories is by Lin Carter or L. Sprague de Camp, not in collaboration with Howard, then the book is an anthology.

If we were to apply this rule always, we would never have need for the pub type "magazine," since all magazines are indeed "publications" of the type described above!

Perhaps it's useful to put the two pub type definitions one after the other, for easy comparison!

MAGAZINE. It can be difficult in some cases to determine if something should be regarded as a magazine or a book. Some magazines were published in book format; some books were published as series with letter columns and regular dates of publication. Borderline cases should be discussed on the magazine or book wiki pages, but generally a magazine must have a common title from issue to issue, and an enumeration or dating system of some kind. This still leaves anthology series such as New Worlds Quarterly as judgement calls. In these cases, look for a consensus on the publication bibliographic wiki page. If no discussion exists, use your best judgment and document the decision on the wiki page.

Looking closely at the two categories it seems that magazines are a particular subset of anthologies; magazines being defined by certain agreed upon characteristics such as letter columns, periodicity, etc.

The question now arises, is a magazine, any magazine, be it "Analog" or "Newsweek," any less a magazine, for having been reprinted?

Conversely, is there any bookish anthology, be it "Dangerous Visions" or any of the "Year's Best Science Fiction," which is no longer an anthology, or is an anthology of another kind, for having been reprinted?

Speaking for myself, I am surprised that there is anything but unanimity on this matter! It seems indisputable that magazines REMAIN magazines; no matter how many times they are reprinted.

I suggest that: (after closely examining both definitions)

We must concede that there is nothing in them which would either suggest or require that we reclassify "reprinted magazines" as something other than "magazines."
Chrisgherr 13:58, 1 November 2016 (UTC)

I suspect that there are two separate issues here:

Our software design originally assumed that a magazine would always have one (and only one) printing. Later on we discovered that it was not a safe assumption, but by then it had been embedded in the software implementation.
There is also a larger philosophical question here. Is a 2013 facsimile reprint of a 1935 magazine still a magazine? At the moment the rules say that it's an anthology, although some series like Adventure House Facsimile Reprints handle it inconsistently.

Ahasuerus 14:10, 1 November 2016 (UTC)

IMHO the defining characteristic of a MAGAZINE is its distribution system. A MAGAZINE is found at a newsagent's stand (even if it can be found in some specialized bookshops). In France the circuits for books and for magazines are completely separated, and I suppose that this holds true in other countries. The fac-similes or the electronic copies are not to be found at newagents, whence their classification as ANTHOLOGIES that use a more "bookish" (if the term still applies to electronic publications) distribution system. Hauck 15:22, 1 November 2016 (UTC)

But that changed a lot in the last 10 years or so - not just because of e-books. If a newsagent is needed, then none of the e-book and print on-demand only magazines of the last decade are magazines (Is Lightspeed a magazine?). Maybe we just need to fix the help a bit (or look at practices) - because with the e-books now in the picture - is an e-book copy of a magazine created 2 months later than the print version a reprint? If so - should it be an anthology? That will make a really complicated display for that magazine/series. We are not good at handling multi-format magazines as it is anyway. And then there are some magazines that changed to anthologies mid-run (Postscripts for example - and they were never distributed as anything but a book but they were a magazine for years) - thankfully they are not reprinting their early issues or it will be fun. A magazine is a useful form to sort publications that come out more than once a year - outside of that, they are anthologies for all intents and purposes... In my mind there are two different things here: immediate format change reprints (ebooks) - which leaves a magazine a magazine; and later reprints (1935 -> 2016 ones) that we are marking as anthologies due to software restrictions. Just thinking aloud. Anniemod 15:34, 1 November 2016 (UTC)

To be frank and to measure how dinosaurish I am, I personally wouldn't have included any electronic publications in the db without a complete intellectual overhaul (e.g. what does "printing rank" means for an ebook? what is a cover artist for such publication or is there any cover to speak of? what's the publication date of a changing text with multiple iterations? If I change an octet in my ebook, is it a new publication warranting a new record? Etc...). Just for the sake of discussion, we're talking here about a 1930 publication that is not likely to have originally existed in an electronic form. Hauck 15:43, 1 November 2016 (UTC)

That would have made the DB almost irrelevant very fast - as a source of finding information on where a book exists anyway. Which is what I came for - I was looking for options for reading a story I could not get the original media for :) And as much as this will create chaos, we will see more and more of these reprints -- we live in the era of reprints and everything that people want will get reprinted sooner or later. So we might as well address it here. As I said in my last sentence above - old reprints are anthologies under the current rules (software restriction or convention - either way this is the practice). That's the current way and until it is changed, this is the best practice. :) But we do need to start addressing the e-books situation and reprints of one-time media(even when printed) in a way fall under the same hat. As much as I wish we could stop the world, technology seems to be moving faster than expected :) Anniemod 16:42, 1 November 2016 (UTC)

Ultimately a database is an attempt to model (a part of) the world. For better or for worse, the world is near-infinitely complex and encompasses a variety of borderline and corner cases. We, on the other hand, have limited resources, so we can't model it perfectly. Even if we had dozens of developers who could create and maintain a very complex database that would reflect real world complexities more accurately, it would make our data entry rules ever more arcane and difficult to apply -- consider the monster that is the MARC 21 family of standards, particularly their bibliographic standard. And so we compromise.

In addition, the world keeps changing. Of the roughly two dozen binding codes that we recognize, less than 20% existed 200 years ago. Dos-a-dos books with two different covers, which were introduced in the middle of the 20th century, presented unique cataloging challenges. ISBNs were adopted less than 50 years ago. And so on and so forth. E-publications and audio books are just the latest challenges in a very long line of challenges. We try to adapt to them as best we can, but sometimes it requires -- you guessed it -- more compromises. And so it goes :-) Ahasuerus 17:34, 1 November 2016 (UTC)

Except that for some things (e-versions of magazines/fanzines/prozines/sami-prozines for example or the realities of the reprint era) we seem to be closing our eyes instead and wish them gone and leave every magazine/editor to find its own way (some have separate series, some are mixed and have doubles in the same grid (Journey Planet) and some just ignore the e-versions (Analog for example off the top of my head)):) I know I know - I will bail out of this conversation and start another discussion for that :)

Software development is always fun like that - you are right. The world changes and the software always lags behind it. But then we should just find a happy medium - or new editors get very confused (me being a relatively new one can attest to it). Sometimes the moderators have issues explaining very easy concepts (although they all are really trying) because they get boggled into the special cases (somehow almost any early addition ends up being a special case) - and anyone that tries to start with something different from adding a novel is a bit lost in the process. And the fact that you need to understand the intricacies to be able to add anything does not help much. We really need a "easy add" kind of page that allows a lot less choices and can be used by new editors as a start - and then have a more experienced editor hold their hand through the steps to get it more complex. Anniemod 18:23, 1 November 2016 (UTC)

(Coming full circle), I asked at the top of this discussion, "Why must we IGNORE best practices?" It seems (to me) that the consensus reached herein is that IF some "best practice" is being "ignored," it's due to software design limitations. It would also seem that, due to said limitations, there has arisen a "rule" which requires that magazine reprints be typed as anthologies. Also (at the top), I copied the Pub Type Template entry for "magazine." Yet -- I find no such rule mentioned, nor referenced, therein. As a neophyte editor; I rely primarily on such templates for guidance. If the (popup) template provided makes no distinction between a magazine's original printing and any reprintings, on what basis am I to demur? I would also ask, on what grounds should any editor (or moderator) ignore the written rules, presumably arrived at thru a process of discussion and consensus, such as takes place here? Don't the moderators, especially, have a heightened responsibility to see that the rules, AS WRITTEN, are abided by? Furthermore, mustn't the moderators set aside their own subjective feelings; and play the roles of impartial referees, where rule enforcement is concerned?
Chrisgherr 21:50, 1 November 2016 (UTC)

I don't enter reprints, but I do own a couple. One is a hardcover, the other is a soft cover trade edition. I think the practice of entering them as anthologies may have started because calling them magazines would have been confusing.--Rkihara 22:35, 1 November 2016 (UTC)

The ISFDB operates off a set of community standards that have been established via discussion (primarily on this page, but can also occur elsewhere) and/or sometimes by de facto practice. Our documentation has gaps and sometimes even inconsistencies. This can be confusing for new users at times, but the solution to that is to fix the documentation when required. -- JLaTondre (talk) 22:50, 1 November 2016 (UTC)

So time to change the documentation - things sometimes are so well known that noone remembers to edit the documentation to say so - and new editors get confused. What exactly would make that part clearer? What changes in the documentation would have made it clearer now that you know what the whole thing is supposed to mean? Once we find a better way to say things and there is enough agreement, documentation can get changed. :) Anniemod 23:21, 1 November 2016 (UTC)

I mostly agree with Anniemod's conclusion, posted above. It's time to consider changing the documentation! It's apparent to me that as JLaTondre points out above, there has been established a de facto standard, or rule, thru practice and consensus, that magazine reprints are publication typed as anthologies. This de facto rule directly (I believe) contradicts the existing de jure rule, written into the website help, templates, etc. Such a contradiction lends itself at the least to confusion, if nothing more. The solution contemplated above, of changing the documentation, (i.e. the "de jure" rule), would surely be an effective solution, harmonizing these two inconsistencies. However, as a tyro, I must wonder: Why has the written rule lasted so long (presumably), unchanged(?), in it's contrariness vs. the unwritten rule? Is the written rule possibly better than its de facto counterpart? For myself, no defense is needed for the existing written help, templates, etc. (regarding this specific matter). I've not seen a spirited defense of the de facto standard. If we're to rewrite the help, to conform with practice, WON'T SOMEONE HERE ARGUE THE BENEFITS OF DOING SO? (By which I mean: How is what we do now so much better than what the Template advises us to do -- and vice versa.
Chrisgherr 02:20, 2 November 2016 (UTC)

How often do you read documentation for a process that you know? Had you tried editing a document you had read 100 times already and where you know the process in your sleep? Both require fresh eyes :) And a will to help with identifying the gaps and missing "common sense" and "but everyone knows this" parts. And the moderators are very good at explaining things to new editors when things are confusing - may take a few attempts or get a bit complicated but everyone is amazing in explaining and helping. Which makes the help pages a bit... rusty. At least this had been my feeling around here.

I did ask you above - what changes would have made you understand how the magazines are handled better? Which parts confused you? Anniemod 02:52, 2 November 2016 (UTC)

I do hope that someone can very specifically address, (or provide an answer to), my two queries (in bold font) above (just slightly above, from November 2). I don't necessarily expect a thorough or detailed response, it's just that I've noticed that no one has commented specifically on the wonderfulness of the current undocumented practice, vs. the awfulness of the older written guidelines. Remember that I'm a newcomer, for whom things obvious to many here, are complete mysteries to me!

I haven't meant to neglect the two (and now four) queries from Anniemod; I suppose I thought that those answers would reveal themselves more fully in time. As I've now been twice asked, I'll now answer (once). ;)

What exactly would make (the documentation) clearer? I think that I, or any editor, could write one or two lines to add to the Template bulleted item (MAGAZINE) at the very head of this discussion, which would make clear the exception for reprints. I hesitate to write those lines myself, just now, as no one has offered here a compelling argument for the superiority of the de facto practice.

What changes in the documentation would have made it clearer now that you know what the whole thing is supposed to mean? As above.

How often do (I) read documentation for a process that (I) know? As a novice here, there're few that I know- so- not very often! ;)

Have (I) tried editing a document (I) had read 100 times already and where (I) knew the process in (my) sleep? The answer here is no.

I will say that if someone could take a crack at the two bold font queries above, I might well understand things better. :)
Chrisgherr 04:33, 2 November 2016 (UTC)

I was semi-serious with the two of the questions. As part of my day job I need to deal with documentation and two things you will never change - the inability of the human brain to properly read a paragraph they had read 100 times already and the fact that if you know what you are doing, you do not read docs. How does this connect to you and your post? Simple - people that had been here for long enough either never read the template page anymore or if they do, they read what they think is there (in order to copy it). Thus practice changing while documentation does not. :)

So you are basically proposing a sentence/paragraph to be added to the "New Magazine" description that says something like: " If you are cataloging a reprint of a magazine (for example a 2011 reprint of a 1935 issue of a magazine), it needs to be cataloged as an ANTHOLOGY and not a MAGAZINE". That's all that is missing from the documentation, correct?

PS: Editing help pages - please do not that on your own - ISFDB is a collaborative effort; so are the help pages. Discussing changes before making them is the process to use (which is why they do not get updated when someone notices an issue all the time but still...) Anniemod 17:08, 3 November 2016 (UTC)

Your bolded questions: are you asking someone to explain why a de-facto rule is better than the written rule? Because this is what everyone here was explaining above - the help page is outdated, it never got updated when practice led to changes. Making a case on why we should edit an outdated help page instead of fixing the said page is a bit of unneeded. It is not about what is so much better - it is really about what had happened while people had really worked with the records and the rules that had evolved with time. If you believe the written rule is better, show some examples and/or argue it - but without bringing the "the docs say so" argument. If the argument has merit and everyone agrees, rules can be changed. But the de facto policies did not come into being because someone decided to just go off script - ISFDB is a living and breathing organism and sometimes practice trumps even the best laid plans. If you are asking something else - I am really not sure what else remains unsaid :( Anniemod 18:59, 3 November 2016 (UTC)

I very much like "the while people had really worked with the records" bit when you put this in perspective with that which shows that our new contributor has less then 10 edits of the database to its credit . Hauck 14:39, 7 November 2016 (UTC)

What counts as a publication for an e-book?

This question has probably been asked before but I couldn't find it. An indie author creates an ebook and uploads it to Smashwords and Amazon simultaneously (and maybe other platforms)-- do these count as separate publications? If there is an identifiable ISBN at all, it is the same one. --Vasha77 19:02, 1 November 2016 (UTC)

I'm not sure we have a definitive standard on this, but in my observation, we generally create a single record for such an ebook. If, however, the books are published with different ISBNs or as by different publishers (per the ebook title/copyright 'page'), then they would be different publications in my opinion. Fortunately, I haven't seen many cases of those two (other than perhaps people reprinting Project Gutenberg texts), but I'm sure it's bound to happen. -- JLaTondre (talk) 23:02, 1 November 2016 (UTC)

Early on, many e-books were issued with one ISBN per format. I haven't seen much of it lately, at least in the fiction market. Perhaps it's due to the fact that ISBNs are relatively expensive. In most cases Smashwords, Barnes & Noble, Amazon, etc have the same e-books these days.

The main bibliographic problem with the way Amazon lists e-books is that it doesn't display their ISBNs even when they exist. There is a way to look them up programmatically, but human users are out of luck. It makes it harder to confirm that the Amazon version is the same as other places' versions. Ahasuerus 00:37, 2 November 2016 (UTC)

A preview/sample of any kindle book can be obtain for free if you have a kindle (or kindle reader) and it contains the first few pages and almost always the ISBN... Now - publishers that stick the ISBN at the end of the book are a problem but still... And more often than not in newer editions, a printed version of the book will contain the e-book one as vice versa specified on their copyright pages. Anniemod 01:14, 2 November 2016 (UTC)

That makes sense. In that case, though, there is a problem with Amazon Digital Services, CreateSpace, Smashwords, etc. named as publishers, if the same ebook is on several simultaneously. Should such books be treated as self-published with the author's name as publisher? --Vasha77 01:54, 2 November 2016 (UTC)

CreateSpace, which is primarily an Amazon-run printer/distributor, is kind of tricky. As Lawrence Watt-Evans once explained:

If you let CreateSpace provide the ISBN, instead of using your own, they'll provide library distribution, but they then show up as the publisher of record.

Some of these CreateSpace-facilitated books have no publisher credit anywhere in the book and end up in our database as published by CreateSpace. Others do have a separate publisher credit, which can be found using Amazon's Look Inside. ReAnimus Press was like that for a number of years, but they seem to be using their own ISBNs now.

I am not sure how Smashwords works these days. Ahasuerus 02:31, 2 November 2016 (UTC)

To take a real-world example, a random book I added to the DB recently was A Dragonlings' Haunted Halloween by S. E. Smith. This exists in the form of a paperback from CreateSpace and an ebook. The copyright page of the Smashwords ebook reads:

IMPRINT: Science Fiction Romance / Science Fiction Romance / A Dragonlings’ Haunted Halloween: Dragonlings of Valdier Book 1.2 / Smashwords Edition / Copyright © 2014 by S. E. Smith / First E-Book Published October 2014

The Smashwords page gives a date of Oct. 20, 2014, and an ISBN of 9781311508980. On Amazon, the ebook is listed with a date of Oct. 19, 2014, and no ISBN. There is no space labeled "Publisher", but instead "Sold by: Amazon Digital Services LLC". The Kindle preview gives the following copyright page:

So... I do not think that Amazon Digital Services (in this instance at least) should be named as a publisher at all; if I was only entering the book from Amazon, I would want to say that the publisher was "S. E. Smith". But what to do with it being on Smashwords? Does the notation "Smashwords Edition" on the copyright page make that a separate publication, and would Smashwords be the publisher? --Vasha77 14:01, 2 November 2016 (UTC)

Addendum: More complexity. This book is on Kobo with the publication info "S. E. Smith LLC, February 2015", with the same copyright page as Amazon; and an "ISBN" of 1230000305247. (Smith incorporated herself as an LLC?! Surely not all authors would do so. That shouldn't be necessary in order for the author to be listed as the "publisher".)

And on Barnes and Noble, we have "ISBN-13: 2940151275286 / Publisher: S.E. Smith LLC / Publication date: 06/04/2015"; and again the same copyright page.

These Kobo and B&N "ISBNs" aren't that, but rather ID codes. I want to know what the Amazon edition's ISBN is -- if it has one. --Vasha77 14:43, 2 November 2016 (UTC)

An interesting example. Using the previously mentioned programmatic method of querying Amazon, I have checked ASIN B00OOWTW1I and ISBN 9781311508980. Neither lookup returned any matches. Of course, 1230000305247 and 2940151275286 are not, as you noted, real ISBNs since 13-digit ISBNs must start with 978 or 979.

As far as "Amazon Digital Services" goes, I agree that it's not a publisher. Here is what Helps says about these types of situations:

For self-published works, fanzines, bibliographic pamphlets and the like, use the name of the editor/author if no other publisher information is visible.

I am not really familiar with Smashwords' self-publishing model, but I note that their list of "Global retail distribution" does not include Amazon. Taken together with the fact that Amazon doesn't recognize the Smashwords ISBN, it suggests that there are two different editions involved. Ahasuerus 14:59, 2 November 2016 (UTC)

P.S. On the other hand, another self-published e-book available on Smahswords, Rise the Renegade: A Rork Sollix Space Opera Adventure, is also available on Amazon with the same ISBN, 9781941939062. Curiously, Amazon's publication date is 2016-02-23 while the one on Smashword is 2016-10-24. Ahasuerus 15:15, 2 November 2016 (UTC)

It seems that even if an ebook has an ISBN, most retailers will display their own ID code in place of it. That does make it hard for a casual user to tell how many editions are out there. Cutting through the Gordian knot would be nice, (all ebooks are the same?) but that might miss some actual differences in publication... --Vasha77 17:00, 2 November 2016 (UTC)

This book is on Smashwords and also has a publisher ( Gone Writing Publishing) listed, whereas the other two examples have no mention of a publisher. Do you think the other two should have Smashwords listed as their publisher or the author's name? --Vasha77 01:32, 3 November 2016 (UTC)

My gut feeling is that the author's name would be safer, but I don't know enough about the way Smashwords facilitates self-publishing to be sure. As our experience with CreateSpace and some other online places shows, a self-publishing company can support multiple models at the same time. Ahasuerus 17:07, 3 November 2016 (UTC)

This is one of the reasons I always thought it was a poor design to separate author authorities and publisher authorities as different record types (but it seems we are stuck with the design now; unless we want to add new authority types like translator but we also include artists in our author records so things are confusing). I do not see the difference when a company is credited with writing a work or a single person is credited with publishing a publication. Perhaps we want to separate authorities by whether they contribute to a work, or a publication of such a work. Regardless this seems like it introduces redundancy (and that always spells maintenance issues). Uzume 21:28, 12 April 2017 (UTC)

Discussion continued at Community Portal

Improve "How to enter foreign language editions" page regarding the publication date of translations

Proposal for a small improvement based on existing rules. This is what bullet point three in Template:TitleFields:Date says regarding the publication date of variants/translations: "When entering a variant title, enter the first date when the work appeared under any title and any pseudonym; typically, variant titles do not have separate dates. However, if the text was heavily revised (or translated) when it appeared as a variant, the date should be that of the first appearance of the revised version of the text."

Based on this I think it would be good idea to add a paragraph like this as third bullet point to the Help:How_to_enter_foreign_language_editions page:

The publication date of the Variant Title should be set to the first appearance of the translation, not to the first appearance of the title of the translation. This means that if a translation of the same translator has been released with different titles, each title would receive it's own Variant Title record, but all of these Variant Titles would have the same publication date. See Template:TitleFields:Date for more details.

Jens Hitspacebar 21:30, 7 November 2016 (UTC)

I think that had been done exactly this way (so we just need to get the help page to agree with practice :) Which is to say "Agree, that page needs an update" :) Anniemod 22:21, 7 November 2016 (UTC)

Yes, but wasn't there just a discussion where most people agreed that under current practice, the date of a variant title IS the first appearance of that title? --Vasha77 22:59, 7 November 2016 (UTC)

If there was a discussion about this and it produced undocumented results the documentation should be updated. Links, please! :) Actually, the idea for my proposal also comes from a (though very brief) discussion on my talk page. Quoting myself from there:

"I always thought that's the way we enter translated titles: use the first appearance of the same translation with the same title, not of the appearance of the translation of the same translator (regardless of the title). No idea where I got that idea from - I probably deduced that from other titles which already existed in the database. A quick search brings up several examples which have been entered the same way, see this and this, or this and this, or this and this. I'm sure there are many others.

However, the TitleFields:Date help says: When entering a variant title, enter the first date when the work appeared under any title and any pseudonym; typically, variant titles do not have separate dates. However, if the text was heavily revised (or translated) when it appeared as a variant, the date should be that of the first appearance of the revised version of the text.

Therefore it looks like you're right. And that we have lots of records which have not been entered the correct way."

Jens Hitspacebar 23:44, 7 November 2016 (UTC)

Can you post a link (or links) to an example so we have a real world reference for what you are suggesting? I am not completely understanding what you want to do and I think having an example would clear it up. ···日本穣 · 投稿 · Talk to Nihonjoe 23:10, 7 November 2016 (UTC)

Sure. Look at the German variant titles of Perdido Street Station: there are three different German titles, but all of them are the same translation by the same translator. The latest German edition is from 2014, using the same translation as the others before, but the title Perdido Street Station for the first time. Nevertheless, its title record has a date of 2002, not 2014. Jens Hitspacebar 23:44, 7 November 2016 (UTC)

Variant dating is another case where our documentation and practice do not agree. The trend has become to date the variant at the first appearance of the variant. This has been raised before (see one and two), but with no resolution. A quick check of the last database dump shows that, where both the parent and the variant are English, only 48% of variants have the same date as the parent (I quickly scripted that so I won't swear it's absolutely correct, but am pretty confident). While each conversation has only minimal participation, there have been no objections and people are not adhering to the current documentation. I think it is time to make the change. -- JLaTondre (talk) 23:45, 7 November 2016 (UTC)

My improvement proposal was especially to clarify how to enter translations, not about variants and parents of the same language. The current documentation (I quote above) treats these two types of variants the same. I'm not sure, but: is all this probably a hint to treat translations differently (i.e. like documented) than same-language variants? I'm not advocating this, because it may overcomplicate things. Just wanted to think aloud. I'm actually fine with either of the solution, as long as we get the current standard documented correctly and up-to-date. Jens Hitspacebar 00:20, 8 November 2016 (UTC)

I entered an example today of same translation, different title. La región más transparente by Carlos Fuentes, translated into Italian by Germán Quintero and Luigi Dapelo, appeared under the title L'ombelico della luna in 2000, La regione più trasparente in 2011. I was planning on giving those variants different dates. But then that would contradict the Perdido Street Station example. Clarification is much needed. --Vasha77 00:33, 8 November 2016 (UTC)

A translation is a different version from the original. Which is why we are varianting. If we keep the same date as the original, we are losing the information about the translation (and even if you may be able to find it by looking at the publication, we have that field that should be usable for something (otherwise we should just remove dates for variants altogether... A translation is a revision in a way - so I think the current rule is covering it but adding it explicitly won't hurt. Now, I think that any variant should carry its earliest date - so it is visible in a glance how the dates and variants progressed (and that seems to be the de facto policy anyway) but I am more concerned about the translations than the straight variants. Anniemod 02:32, 8 November 2016 (UTC)

Translations are dated separately already per our documentation. As I understand it, Hitspacebar is talking about occurrences of the same translation (i.e. by the same translator) under different titles and/or author credits. -- JLaTondre (talk) 11:45, 8 November 2016 (UTC)

Exactly. Jens Hitspacebar 13:01, 8 November 2016 (UTC)

@Hitspacebar: I recognize that your proposal was about translations. However, it is the same principle. Either we believe there is merit in recording the first time that a particular title & author credit was used or we don't. If we do, than it should apply to translations as well. Given most users seem to think there is, the documentation (which would apply to multiple pages) should be updated to state that variants (including translations) should be dated to the first appearance of that specific title and author credit. -- JLaTondre (talk) 11:45, 8 November 2016 (UTC)

I agree with both of those points. It is the same principle, and I think there's merit in recording the date of the first appearance. That said, I would be interested in seeing an explanation of the reasoning behind the always-use-the-canonical-date rule. Is that doing something we would lose if all variants used their origination dates instead? --MartyD 12:05, 8 November 2016 (UTC)

I currently don't have an opinion about what's best for same-language variants, but as for translations I think it feels more intuitive (especially for the occasional user of the site) to record the first appearance of a variant title under its title and not (as it's currently documented) the first appearance under any variant title. This already seems to be the current practice of several editors if I understood the discussion correctly so far. This would mean that the date of the title record of the Perdido Street Station example I posted above should be changed from 2002 to 2014. The information lost with this practice would be the immediate answer to the question "when was a translation by certain translator first released". However, this information can still be retrieved from the NOTE field of the title or publication records as long as the translator is stated there. Jens Hitspacebar 13:01, 8 November 2016 (UTC)

(unindent) I think it would be best to consider all variant-dating rules, including the rules that apply to same-language and translation VTs, at the same time. If we handle them piecemeal, there is a good chance that we will forget some permutations.

Re: the original reason why Help says "typically, variant titles do not have separate dates", as I recall, it had to do with the way variant dates were displayed in the Contents section of the Publication page. The software has changed in the last 10 years and I think that it handles different variant dates correctly now. We'll need to double-check, though.

My current take on the larger issue is that in most cases (including this one) it is best to capture as much granular information as possible. In this case it would mean updating Help to something like "A new date is assigned to the first appearance of a new version of a text [including new translations] and/or a new title". Ahasuerus 14:15, 8 November 2016 (UTC)

I think that the last paragraph is the way to go - update the help page, have the date the first time the variant started to exist as a date -- this way you can see at a glance when a new title started to be used for example... Anniemod 23:12, 8 November 2016 (UTC)

I looks like this seems to be the consensus in all answers if I didn't overlook something. Is more feedback by other moderators needed or should we go ahead and update the documentation? Jens Hitspacebar 15:23, 11 November 2016 (UTC)

Anyone...? Hitspacebar 18:23, 12 November 2016 (UTC)

If there are no objections, I will change Help in 24 hours. Ahasuerus 19:03, 12 November 2016 (UTC)

Template:TitleFields:Date has been updated. Hopefully the wording is clear. Ahasuerus 20:44, 13 November 2016 (UTC)

Proposal to create a rule changelog

With so many rules split across lots of pages, and with so many changes going on in the Wiki every day it can be pretty tough to find out if important things have changed - especially regarding rules and if you're not here on a daily basis. I think it would help a lot if we'd do two things:

Create a changelog for the rules (similar to Development/Recent_Patches) which should contain the following information for each rule change:
- Date of the change.
- Very short description (e.g. "Date of variants").
- A link to the wiki page which conatins the rule.
- What was the rule before?
- How has it changed, what has changed?
Ensure that every rule change is logged there. We could create a small wiki template which would be added at the beginning of every thread started on this discussion page in order to verify its outcome:
- Is the thread about a rule change at all? [yes, no]
- Is there a result? [yes, no]
- Has the documention been updated to the changed rules? [yes, no, not applicable]
- Has the rule change been logged in the rule changelog? [yes, no, not applicable]

Jens Hitspacebar 16:03, 11 November 2016 (UTC)

I like the idea. When an editor comes back after a hiatus, it can be hard to determine what, if anything, has changed. Ahasuerus 16:44, 11 November 2016 (UTC)

P.S. I should add the we already have What's New, which is mostly dormant. We'll have to think of ways to integrate it with the proposed changelog page. Perhaps we could make the changelog a sub-page or a section of the "What's New" page. Ahasuerus 16:47, 11 November 2016 (UTC)

I think it shouldn't be section of the What's New page but a new separate page because rule changes are (mostly) a different thing than new or changed features (though they can overlap of course, e.g. if a new feature needs a new rule). If you want to revive the What's New page and put all on one page there might again be too much clutter to detect which rules have changed. Jens Hitspacebar 18:21, 12 November 2016 (UTC)

Well, we could create a new Wiki page for the proposed changelog and then link it from "What's New" as well as from the main Help page. Ahasuerus 20:45, 13 November 2016 (UTC)

Yes, that's better. We might want to wait a few days for more editor feedback, and then I could create the page (and the templates I mentioned above) over the forthcoming weekend. Jens Hitspacebar 17:16, 14 November 2016 (UTC)

I'm not sure what y'all mean by a 'rule', but it seems to me that if it's a rule, it's documented. If it's not documented it's a habit. So if there's a discussion of a 'rule' - be it an interpretation or request to change / create / delete - you should be able to reference that rule. So to automate your log (generally the only hope of keeping it accurate and complete), you need to track changes to the places where rules are kept and track references. I presume the wiki pages that contain the rules have tracking, so what would be useful is a way to reference a rule in wiki discussions that will allow for automatic tracking. Doug H 03:39, 14 November 2016 (UTC)

It's true that the Wiki software keeps a history of every Wiki page, including Help pages. However, although you can use Wiki history to find policy changes, it can be time-consuming once you get past the last few changes. Also, some changes are minor and don't affect policy, so you may have to wade through a lot of editorial changes to find what you are looking for. Finally, some policy changes, e.g. translation changes, applied to multiple Help pages, which would make reverse-engineering the policy change from Wiki history very hard.

In addition, we don't have unlimited disk space, so we have to purge Wiki history (all but the last 50 versions of each page) periodically. Ahasuerus 14:52, 14 November 2016 (UTC)

"Omnibus" whose only novel is non-genre

I would like some other opinions on the subject of this discussion -- the situation just seems weird. There is an omnibus that contains only a non-genre novel and a genre novella, so what title type is it? (BTW I don't want to add La muerte de Artemio Cruz to the DB-- Fuentes has too many non-genre works and too many editions and I don't want to even start letting any of them in.) --Vasha 16:51, 13 November 2016 (UTC)

Our publication type definitions make no distinction regarding contents being genre or non-genre. And that to me seems to be proper. From a bibliographic perspective, the publication remains the same type of publication even if we choose to not list certain contents. Our cleanup scripts are designed to improve the quality of the database. In a case like this, we shouldn't degrade the quality just to satisfy the script. Instead, a better solution would have been to ask Ahasuerus if he could have the script ignore this record (this is uncommon enough case that adding a moderator ignore capability is probably not worth the effort). -- JLaTondre (talk) 13:01, 14 November 2016 (UTC)

We have lots of Collections that only contain one genre short story. We do not want them listed as "chapbooks". The same philosophy would apply here. Chavey 20:38, 14 November 2016 (UTC)

I agree that collections, anthologies and omnibuses that contain only one SF work should be entered as collections, anthologies and omnibuses respectively. Ahasuerus 20:56, 14 November 2016 (UTC)

Thanks for that. Shall I make the change to OMNIBUS or do you want to do it? --Vasha 21:31, 14 November 2016 (UTC)

Well, there is no hurry, we can wait and see what other editors think and then adjust Help accordingly.

I should also add that I am not the ultimate authority as far as policy and data entry rules go. I just tweak the software to support the community consensus. There have been times when I proposed new features only to drop them for lack of support. Ahasuerus 21:48, 14 November 2016 (UTC)

Unless there are objections, I will adjust the cleanup report to let moderators ignore "OMNIBUS pubs with no NOVEL titles" tomorrow. Ahasuerus

The software has been updated. Moderators are now able to "ignore" OMNIBUS publications without NOVEL/COLLECTION/ANTHOLOGY/NONFICTION titles. Ahasuerus 23:38, 17 November 2016 (UTC)

(outdent) I've update the record to be an OMNIBUS. It can be ignored in the next cleanup report. -- JLaTondre (talk) 00:17, 19 November 2016 (UTC)

Print and Ebook Editions of Clarkesworld Magazine

Hi all,

I have been adding the 2016 issues of Clarkesworld magazine and just noticed today that there is also a magazine Clarkesworld (print). These issues mostly do not have the table of contents filled in. From the issues that are filled in, it seems that the ebook and print editions are nearly identical. The only differences are 1. price 2. the inclusion of page numbers in the print edition. Why do we need both of these series? Is the print series being automatically generated from Amazon and that is why the table of contents is blank on non-verified issues?

Sorry if these questions are easy to answer; I am new here. m (I also don't know why this isn't showing up as a new section...sorry person whose section I have hijacked) Amoeba of horror 05:06, 14 November 2016 (UTC)

Yes, the print issues are straight from amazon via Fixer (our robot) and massaged into shape by a moderator (usually me) that, alas, has no time to enter the contents without a physical copy. Don't hesitate to populate the print issues by using the "Import Content" link on the left and entering the titleID# of the electronic issue in the top part of the screen (in this case do not use the bottom part). Hauck 07:15, 14 November 2016 (UTC)

I think you meant "publication ID ... in the top part of the screen" :-) Ahasuerus 14:42, 14 November 2016 (UTC)

To add to what the guys already said (and if you know that already, ignore it) - a series grid contains publications. So if we add both the publications in the same series, you end up with something like this grid: Journey Planet. Which works for smaller series but gets too crowded for ones that have issues every month (or more often) and had had a lot of issues. So a second series is sometimes created for the second format - we kinda do not have a specific rule on that but that is the only way to have them and not mess up the formatting of the grid. :)

PS: For making a new section, when you are adding, use the "+" on top of the screen and not "Edit". That will ask for a title and make a new section for you. Anniemod 05:28, 15 November 2016 (UTC)

Collected Works as a series?

Do you think it's a good idea to put the volumes of a Collected Works edition in a series or not? There aren't very many examples of collected works in the DB to compare. Ambrose Bierce has a series made, G. K. Chesterton does not. --Vasha 03:35, 16 November 2016 (UTC)

If they are published as a series and they are unique selections and are connected in some other way besides being written by one person (such as this one) - yes. But if one of the volumes is a single novel for example (that can be published outside of the series), then I would make a publication series instead.. Looking at your examples - both can be either. Just my 2 cents :) Anniemod 03:47, 16 November 2016 (UTC)

Usually it would be only a pub. series. There really are only a few examples of collected works as a title series; and it's only meaningful if they are published for the first time, such as Philip K. Dick's collected shortfiction (but as Anniemod wrote: if they had enclosed his novels also, it likely wouldn't have been catalogued as a title series). Stonecreek 05:21, 16 November 2016 (UTC)

At the moment we have a few dozen regular "Collected" series and six "Collected" publication series. Translated "collected works" like this one are usually entered as publication series.

Another thing to keep in mind is that support for publication series is relatively recent. Early on editors had no choice but to use regular series. There is a good chance that many of them would have entered regular series like The Collected Stories of Robert Silverberg (Bantam) and The Collected Stories of Robert Silverberg (Subterranean Press) as publication series if the functionality had been available in the 2000s. Ahasuerus 18:52, 16 November 2016 (UTC)

Additional thoughts after sleeping on it:

Suppose we converted The Collected Stories of Robert Silverberg (Subterranean Press) to a publication series. At this time the series is up to 9 volumes, each one with a different title, e.g. "Trips: 1972-73" or "Hot Times in Magma City: 1990-95". If the title series were eliminated, these 9 titles would be interspersed with other Silverberg collections on the Summary page. There would be no easy way of telling that they are related, which would seem like a step back. Ahasuerus 18:29, 17 November 2016 (UTC)

That's what I had been thinking about with my "unique selection" comment above -- there is a difference between collected stories and collected works in my mind - collected works should almost never be proper series - because of the inclusion of novels and/or preexisting collections -- while collected stories will almost always be unique and grouping them makes sense on the author page - which leads to a proper series... Anniemod 19:01, 17 November 2016 (UTC)

Proposal to group pubs without a publication series

Hi,there are some publishers (such as this one) with hundred of titles grouped into several publication series. Only a very few titles (see this) are not assigned to any pub series (for whatever reason). That's why these titles are hard to find in the publisher's book list: it is necessary to click through every single publication year to view the column 'publication series'. My proposal is: If a publisher has one or more publication series then pubs not assigned to a pub series are grouped into a special 'pub series', e.g. [no publication series]. This will be very helpful. Boskar 10:00, 18 November 2016 (UTC)

Having thought a bit about that problem, I'd think that such an installation may lead to some problems: 1) It would contradict the help pages: A Publication series is a group of publications marked out by the publisher in some way. (that's why I removed such a 'pub. series' in the first place). 2) Even more grave is that when we state [no publication series] we'd assume something that we don't know about, at least for the loads of unverified publications: they may have a pub. series stated, but as no one of us has looked into the publication, we can't just state that there isn't any. Or would there be a way to circumnavigate this? Stonecreek 15:54, 18 November 2016 (UTC)

If the problem is finding publications that are not assigned to a publication series, then we can add a new link, "View publications not in a publication series", to the Publisher page. Ahasuerus 16:57, 18 November 2016 (UTC)

If that'd be not much trouble, it'd be wonderful! Stonecreek 17:03, 18 November 2016 (UTC)

OK, FR 948 has been created. I hope to finish the monthly upload of Fixer-identified ISBNs tonight, at which point I should be able to go back to development. Ahasuerus 17:29, 18 November 2016 (UTC)

The software has been changed to support this functionality -- please see the Festa page for an example. Ahasuerus 23:46, 19 November 2016 (UTC)

Hey! That's not just great, that's perfect! Thank you! Boskar 13:33, 20 November 2016 (UTC)

Excellent! :) Ahasuerus 18:01, 20 November 2016 (UTC)

Reviews of Magazines

Under the Rules of Acquisition Policy it is stated:

Reviews of media products, stage productions, magazines and fanzines (regardless of their genre), and books that are not eligible for inclusion in the database, should not be entered into the "Reviews" section of the data entry form. A record should be created in the "Regular Titles" section typed as ESSAY.

However, this appears to be contradicted by the Help Screen for New Pubs, which states:

Note also that only books, magazines, and short fiction are entered; if the column reviews fanzines, you don't need to enter the review records for these, only the ESSAY record.

Albinoflea 06:03, 19 November 2016 (UTC)

Minor quibble - call them Regular Title records, not ESSAY records to be consistent. You're mixing interface and database terms. Doug H 16:11, 19 November 2016 (UTC)

Minor quibble(2) - why call them EDITOR records? The screen I use to get the record number to link to calls it "ISFDB Title Record #". The word EDITOR shows up 4 lines down as the Type: Doug H 19:06, 21 November 2016 (UTC)

Title records have different "title types". Some pages call them "Entry types", which may be confusing. The full list of supported title types is available as a drop-down list in Edit Title. It is also available in the Regular Titles sub-section of the Content section of the Edit Publication page as well as on the Advanced Search page. Similarly, publication records have "publication types".

Each publication record must contain a "reference title". In most cases the type of the reference title matches the publication type, e.g. "NOVEL"/"NOVEL" or "COLLECTION"/"COLLECTION". The only exception is magazine and fanzine publications. Their publication types are MAGAZINE and FANZINE respectively, but the title type of their reference title is "EDITOR". Ahasuerus 04:10, 22 November 2016 (UTC)

I was taught/advised long ago that even though entering reviews of magazines was not in itself bad form, since they can't be linked to their TITLE records (but rather their EDITOR records) that it was better to leave these as ESSAY entries rather than REVIEWs.

I'm wondering what the current consensus is on this as there are a group of LinkReview requests in the moderation queue now attempting to link REVIEWs to EDITOR records, and I'm not sure how best to proceed.

It would appear the three basic options are:

a) Enter reviews of magazines as ESSAYs
b) Enter reviews of magazines as REVIEWs, but leave them unlinked
c) Enter reviews of magazines as REVIEWs, and link them to the corresponding EDITOR records

What's the thinking around each of these options? Thanks, Albinoflea 06:03, 19 November 2016 (UTC)

Certainly a) for the reasons you state, plus it may be that the magazine/fanzine is not in the db, in which case you would not have a record to link to anyway. PeteYoung 09:12, 19 November 2016 (UTC)

If a book reviewed is not present, you have nothing to link to. Is it normal to create a stub in such cases? Doug H 16:11, 19 November 2016 (UTC)

If the book falls within the Rules of Acquisition, yes. I can't tell you how many hundreds of new titles I've added just from reviews in SF Commentary alone. PeteYoung 22:09, 20 November 2016 (UTC)

(After edit conflict with Pete) When an editor comes across a review of a book that is not in the database, it is usually the starting point of an investigation. For the most part it turns out that the book ought to be in the database. If that is the case, we use the review as well as other secondary sources to create a record. Other times it turns out that the book is not eligible for inclusion. Occasionally we run into more complex cases, e.g. a book reviewed under its pre-publication title, a cancelled book or even a joke review of a non-existent book, so we have to be careful. Ahasuerus 22:16, 20 November 2016 (UTC)

Regarding b), I would consider this bad form. To my eye it looks like an unfinished publication record to a database user, who may not understand why no link is provided.

Regarding c), If a magazine or fanzine is grouped by year (as is thankfully more common now), the review as it appears on the Editor record will not be seen to refer to a specific issue. I've added a fanzine review to the bottom of this pub (which I will remove later) so you can see how it appears here: the specific issue reviewed does not show up on the Editor record, it just looks like a review of the whole year's publications. Also, sometimes a magazine or fanzine review can refer to more than one issue, eg. Review of the fanzines Broken Toys #17 & #18 by Taral Wayne so, if we're lucky enough that they are in the database, which would you link to? PeteYoung 09:12, 19 November 2016 (UTC)

We've arbitrarily decided to group magazines by year and should expect some fallout. This one is that reviews of all issues are combined in the list of links. The review's title could/should specify the particular issue(s). If a review covers two book titles, wouldn't you separate them into two reviews? Couldn't you do the same for issues that span years? Doug H 16:11, 19 November 2016 (UTC)

I agree that the word "magazines" in the Help Screen for New Pubs is misleading and should be removed. My two cents. PeteYoung 09:12, 19 November 2016 (UTC)

It looks like the referenced section of Help is out of date. It says:

Non-sf works should be entered but if an onerous number of non-sf-related works are reviewed in a column you are entering, discuss the situation on the Bibliographic Rules page to decide what can be eliminated.

even though the relevant policy was changed a while back. The new policy is explained on the same Help page:

... books that are not eligible for inclusion in the database (graphic novels, nongenre novels by authors that are below the threshold, nonassociational nonfiction works), should not be entered into the "Reviews" section of the data entry form. A record should be created in the "Regular Titles" section typed as ESSAY.

Ahasuerus 22:23, 20 November 2016 (UTC)

I think the question is - Does the new policy include magazines as 'books' to be put in the "Reviews" vs "Regular Titles" section? IF they go under "Reviews", they must be linked to the title record # (that's what the screen says) for the issue's publication entry. This loses the issue information, which could be included in the review title. Multiple issues would mean multiple review entries. Missing issues would not be a problem if there are issues for the year. If the year has no issues - a stub would be required. If the magazine is not present, it's a whole new layer of stubs. Doug H 19:06, 21 November 2016 (UTC)

The Help excerpts that I quoted immediately above are not related to the magazine issue. I posted them as an example of this Help paragraph being out of date in other ways as well. Ahasuerus 03:28, 22 November 2016 (UTC)

The idea that "The review's title could specify the particular issue" has its charm. Up to now I've only added reviews for magazines that have single title entries (for example because of the schedule of publishing: only once or twice a year). Stonecreek 17:04, 21 November 2016 (UTC)

One problem with this approach is that the way the software currently works any title assigned in the Review is stripped out when it is posted on the TITLE/EDITOR record; see Dead of Night - 1995 for an example; even though the title of the review is "Terminal Fright #10, Winter 1995" it only displays as "Review"; Pete alludes to this in his comment above.

If the desire is to discourage the linking of reviews to records with type EDITOR, then perhaps we should flag those on the moderation screen? I assume they would never be matched by the system and are always going to show up as manual LinkReview entries in the queue.

Or is it a situation where if someone wants to make the added effort to create these links, even if they're not perfect, moderators should pass them through? Albinoflea 01:30, 22 November 2016 (UTC)

(unindent) I have been thinking about this for the last couple of days. The underlying issue appears to be that we merge EDITOR records for magazine publications. If we didn't, then we could link REVIEWs to individual issues and the problem would go away.

Now, the reason that we decided to merge EDITOR records back in the late 2000s was that some authors/editors like John W. Campbell Jr. and Stanley Schmidt are responsible for hundreds of issues. Our concern was that displaying one line per issue would result in very long Summary and Series pages that would be difficult to navigate. Unfortunately, the implemented solution has proved problematic in other ways as seen here.

Perhaps what we need to do is to take a step back and examine whether it would be practical to:

convert to a "one magazine issue = one EDITOR record" system, and
handle any display issues with extra-long Summary/Series pages in the software.

The "unmerges" that will be required to change to the new system can be done automatically. Ahasuerus 03:22, 22 November 2016 (UTC)

I would definitely be in favor of that; the collision of presentation and data structure that was created when this decision was made seems to consistently trip up editors working on more complex magazines. Which is not to say the decision didn't serve some purpose when it first came about.

It's good to hear that the unmerges will be able to be automated; a few other things that come to mind:

We'll likely need a cleanup report to identify reviews that are currently linked to EDITOR type records with multiple pub records associated with them, so they can be re-linked to their proper issue.
There will be numerous reviews of magazines that have been entered as ESSAY type titles under the current guidelines that will be eligible for conversion; these should be somewhat easy to programmatically identify if they're titled with the current naming convention.

I suspect the display issue will be the trickiest bit of the process. Albinoflea 07:36, 22 November 2016 (UTC)

The one area makes use of a single (or as few as possible) EDITOR records per year is that of awards. We should consider the best way to handle those, not to mention how to handle the awards that are currently linked to annual EDITOR records. We could convert them to untitled awards, or perhaps change the software to enable us to link award records to magazine series. --Ron ~ Rtrace^Talk 11:14, 22 November 2016 (UTC)

Ahasuerus says the unmerge can be automatic. Does that include the linking of existing reviews to a particular issue? This isn't always obvious since the title may say June 2010, but the review says Issue #123. It's also possible that the particular issue doesn't exist - no stub was created because it would be the same EDITOR record. I don't know how big an problem (almost said issue) these present. The mis-matching of details in the titles also presents a problem when creating stubs for a review - presuming we do the same thing for an issue as we do for a book stub which is to fill in as much as you can using as many sources as you can reasonably find. This is easier for new issues of an existing magazine series, but harder if no such series yet exists. BTW - will unmerging affect how the magazine issue list / grid are assembled and/or presented? Doug H 14:16, 22 November 2016 (UTC)

I don't know how good of an idea this is, but that's never stopped me before: It seems that the variant title mechanism -- plus some display changes -- might give us a lot of the mechanics we want. Really, we'd like one title (EDITOR) record per issue (which would also then let us handle reprints/reissues and multi-format, BTW). Such a record could be linked for a review of a specific issue. We could then have a parent record that represents some logical grouping (might be calendar year, might be volume, might be something else) to a set of these could be varianted. Not unlike our handling of serializations of novels. That parent record could be used for issue-less situations, such as awards. --MartyD 14:25, 22 November 2016 (UTC)

I had been typing something in that direction as well and walked away to find coffee and MartyD posted :) We may have a "best of the two worlds" solution -- leave the existing yearly records in place but create variants for each of the separate issues. That allows the awards to stay on the year level, any review of a whole year can stay attached as well but gives us the individual entries. It will also open the door for a new user preference "Show individual issues" which will allow the user to tweak if they want the current view or a view that shows the individual issues as well. Although I am not sure if you can enter a review to a variant? If we cannot, we need this tweaked anyway - to allow a review of a translated work to be linked to the proper variant (and then allow the display to show it on the parent record)... Anniemod 14:36, 22 November 2016 (UTC)

Yes, you can link reviews to variant titles. For example, this review is currently linked to a VT. Reviews of variants are listed on the variant page as well as on the parent page.

The more I think about this approach, the more I like it. It should be easy to suppress the display of EDITOR variants on Summary and Series pages. You can have your garash cake and eat it too! :) Now we have to think through the proposed solution's implications and consider any unintended side effects. Ahasuerus 20:29, 22 November 2016 (UTC)

And here is the fist unintended side effect. Consider Charles Coleman Finlay's Summary page. He has been editing The Magazine of Fantasy and Science Fiction as "C. C. Finlay" since mid-2014. Under the current system, his EDITOR titles are attributed to "C. C. Finlay", then varianted to "Charles Coleman Finlay". His Summary pages displays them as:

The Magazine of Fantasy & Science Fiction, July-August 2014 (2014) with Gordon Van Gelder [only as by Gordon Van Gelder and C. C. Finlay ]
The Magazine of Fantasy & Science Fiction - 2015 (2015) [only as by C. C. Finlay ]
The Magazine of Fantasy & Science Fiction - 2016 (2016) [only as by C. C. Finlay ]

Under the proposed system the variant EDITOR titles will still be attributed to "C. C. Finlay", so there will be no loss of information. However, if variant EDITOR titles are hidden on Summary and Series pages, you will have to drill down to the parent title level to see them.

I guess we could tweak the display logic some more and replicate the current "only by" messages based on authorship. Ahasuerus 21:19, 22 November 2016 (UTC)

Or change the rules a bit - and have the title records for EDITOR to always contain the canonical name as we do for interviews (I think?). We won't lose information because the publications will have the correct names. If you want to limit to ones that have publications so the yearly records can be varianted that is fine as well I guess. But that will solve the immediate problem. Anniemod 22:26, 22 November 2016 (UTC)

Reviews and interviews are a special case. Regular titles capture one sets of authors/editors per title record. Reviews and interviews capture two sets per title record: one set for reviewers/interviewers and another one for "reviewees"/interviewees. The data entry rules for reviewers and interviewers are the same as for regular titles. The rules for "reviewees" and interviewees are different because they not really "title authors".

Also, AFAICT changing the data entry rules to use canonical names for EDITOR records would have the same impact as simply hiding variant EDITOR titles on Summary and Series pages. Ahasuerus 23:02, 22 November 2016 (UTC)

Yes, I understand why the special cases are as they are. I am not sure I follow why it is the same impact (or I am misunderstanding what is proposed to be hidden). If the editor records carry the canonical name, they will show up on the page authomatically (which won't happen if they are on the pseudonym as we cannot variant as they are already variants). If the change proposal is to do that behind the scenes, then yes, that would be enough. Anniemod 02:28, 23 November 2016 (UTC)

(unindent) For now I have removed the offending Help paragraph. The information that it contained was either duplicative or obsolete. Ahasuerus 21:24, 22 November 2016 (UTC)

This is interesting but I have two question: How do I enter magazine reviews now? Will the system and moderators accept requests that follow this process? Doug H 20:07, 28 November 2016 (UTC)

In re-reading this in search of answers for the same question Doug H asks, I notice it never was answered. Are we still in limbo on this or was it answered some place I don't know about? Thanks, Doug / Vornoff 06:35, 3 March 2017 (UTC)

Many R&S discussions never come to a resolution.... IMO, you are best off thinking "EDITOR" == "TITLE" and following the general review-entry rules. If the magazine being reviewed is eligible for inclusion or is already in the database, link to the closest available EDITOR record. If it's been merged and is for a full year, so be it. The reason I suggest that is because if any of the above suggestions are ever implemented, we'd easily be able to find such a review and do something with it. If it's entered as ESSAY, we'd have a much more difficult time finding it. --MartyD 12:30, 4 March 2017 (UTC)

Weird Heroes: publication or title series

We have this as a title series, probably entered long before we had publication series. This prevents Ron Goulart's Quest of the Gypsy (two novels and one story) to be in a title series of their own. I would like to convert Weird Heroes to a publication series, but that wolud leave the 2004 i-books reprint of volume 1 out of the series. any ideas (or other problems)? --Willem 20:33, 20 November 2016 (UTC)

Given the fact that these books contain titles which belong to various unrelated title series, I think it makes sense to convert it to a publication series. Ahasuerus 21:56, 20 November 2016 (UTC)

Looks like there were no objections. Conversion is done. --Willem 21:14, 26 November 2016 (UTC)

Are Baudelaire's prose poems POEM or SHORTFICTION?

Currently there are seven of Baudelaire's "Petits poèmes en prose" in the DB; "La chambre double" and "Le joueur généreux" have the title type POEM, the rest SHORTFICTION. --Vasha 21:50, 20 November 2016 (UTC)

As I recall, at least one editor worked on sorting out Baudelaire's works a while back. Unfortunately, I don't remember who it was. Ahasuerus 21:58, 20 November 2016 (UTC)

Various people verified publications with one of these prose-poems in them: Rtrace, Don Erikson and Taweiss verified Le joueur généreux as POEM, and according to a notation by ChrisJ, it may be listed as a poem in Contento too. Only Linguist verified one of them (Mademoiselle Bistouri) as SHORTFICTION. Personally, I think POEM makes more sense. Everyone seems to agree that a "prose poem" is a thing that exists. --Vasha 22:10, 20 November 2016 (UTC)

"Prose poems" are on my list of things to add to the set of recognized title types, but we'll need to have a comprehensive discussion of various permutations first. Ahasuerus 22:19, 20 November 2016 (UTC)

OK, another data point then: many critics describe the little snippets of prose that Jorge Luis Borges published in El hacedor (a collection of half verse, half prose) as "prose poems"; currently they, and a few similar pieces he included in other poetry collections, are listed as SHORTFICTION. (To complicate matters, they were included in Andrew Hurley's translation Collected Fictions but also some of them found their way into the companion volume Selected Poems as well.) They certainly are rather more lapidary in style than Borges's earlier and later "fictions". --Vasha 22:37, 20 November 2016 (UTC)

I don't see why there needs to be a separate title type for prose poems, though. They are already unproblematically recognized as poems when they occur in anthologies of poetry (like "Paint" and "I Last Saw My Mother" in this anthology). Sure, there is plenty of confusion in individual instances as to what exactly an item is; but that would only carry over to a new distinction, with people not sure whether to call something SHORTFICTION or PROSEPOEM. I think what is needed, instead, is a note added to the page on title types confirming that prose poems are poems, and leaving the problem of recognizing them (which will only occasionally be difficult) up to individual editors. --Vasha 21:37, 21 November 2016 (UTC)

Unless anyone strongly objects, I'm going to change Baudelaire's "shortfictions" to POEM to make them all consistent, but leave the other questionable cases be for now. --Vasha 02:01, 27 November 2016 (UTC)

Multi-volume "Omnibus" works with only one novel

On an issue similar to the one recently discussed, we have at least a few examples of "The Collected Works of XXX", published in several volumes, but with only a few volumes containing genre works. Should the entire "Works" be entered, or only the volumes containing genre works? This certainly applies to many editions of the Works of Shakespeare, for example, but the one that jumps out at me is Jonathan Swift. All 5 of the English editions of Gulliver's Travels that we have listed between 1735 and 1765 are 1-2 volumes of "Collected Works" editions. Four of them were entered (by me) as novels with just the volumes for Gulliver's Travels; one was entered as an omnibus with all of the volumes mentioned. (In addition to Gulliver's Travels, the "Works" also includes the "Battel of the Books", of genre interest.) Which is the correct way to enter these works? Chavey 21:56, 21 November 2016 (UTC)

I would enter just the volumes containing genre works. I would then create separate publication series for each "Collected Works" version. Ahasuerus 22:15, 21 November 2016 (UTC)

Thanks. Chavey 23:19, 21 November 2016 (UTC)

It would also be good to include a clear note on the series indicating why some volumes weren't included. ···日本穣 · 投稿 · Talk to Nihonjoe 20:35, 22 November 2016 (UTC)

Composite "story length" values for OMNIBUSes

Re-checking my recent changes to Template:TitleFields:Length, I see that the previous version of this template read:

Other possible uses include "/1N/2C" to indicate the omnibus contains a novel and two collections

The current version reads:

For other types of omnibuses, enter the count of included titles, e.g. "/3N" or "/2N+2C" where "N" stands for "Novel" and "C" stands for "Collection"

Note the difference between "/1N/2C" and "/2N+2C".

This was not an intentional change. I am so used to plus signs being used as separators that it didn't register that the 2008 version suggested using slashes.

Checking the database, I see that we have 12 titles that use slashes and 94 titles that use the plus sign. Given the current usage, I suggest that we make the plus sign our official choice and convert the 12 titles that use slashes. Ahasuerus 17:40, 29 November 2016 (UTC)

I think that plus is more logical. Anniemod 18:07, 29 November 2016 (UTC)

There isn't the slightest bit of consistency in how people describe the contents using the plus notation. (Go through half a dozen items in that search list and you'll get an eyeful.) But in all the ones I looked at, I could figure out what they meant, so I don't think it's much of a concern. Do you want to give more detailed instructions, though?

One thing that might be plain wrong is that some people wrote something like "/1-4+nv" meaning volumes 1-4 of a certain series plus a novella. --Vasha 18:12, 29 November 2016 (UTC)

Well, "/1-4" is standard usage for omnibuses containing volumes one through four. How would you propose entering an omnibus which collects volumes 1-4 plus a novella? Ahasuerus 18:28, 29 November 2016 (UTC)

Give the novella a proper number in the series and then just use /1-4+3.5? More seriously: we also may have options like /1-3+4ss for example. I would say something like - list the numbers of the known elements and then after the + add all other elements in the order N, C, nv, nvt, ss, sf (So /1,2,3.6+1N+2nv+2ss) - which means series members 1,2 and 3.6 and then one novel, 2 novellas and 2 stories that are not numbered? Anniemod 18:40, 29 November 2016 (UTC)

I think it may be an overkill considering that only 106 titles are affected by this rule. As Vasha pointed out, the codes are usually self-explanatory. Ahasuerus 22:18, 29 November 2016 (UTC)

Probably a valid argument :) Always can be revisited when the number of strange editions increase. Anniemod 22:22, 29 November 2016 (UTC)

What do you do if something is both an omnibus and juvenile? Two kinds of data to record in the same field. --Vasha 22:00, 29 November 2016 (UTC)

Omnibus data takes precedence in my opinion, jvn gets re-delegated to the notes (but this is exactly why I think that the non-length identifier should be sent to their own checkboxes. Plus jvn as length is a bit... funny. :) Anniemod 22:02, 29 November 2016 (UTC)

It's also possible for a title to be "novella", "juvenile" and "novelization" at the same time. Which is why FR 163 was created a while back. I hope to implement the FR soon (tm) and convert the "storylen" field to a drop-down list for SHORTFICTION titles. Ahasuerus 22:17, 29 November 2016 (UTC)

(unindent) OK, so the plus sign is the new standard just like the new Help text states. I have converted all eligible slashes to plus signs. Ahasuerus 16:34, 2 December 2016 (UTC)

Publication format clarification, please -- moved from CP

I've been using the format "pb" to describe the European "pocket" size of paperback, which is 19-20 cm tall (and contrasts with non-pocket paperbacks which are generally 23 cm or more). But I now notice that the template description for PubFormat says "For books as tall as 7.25" (19 cm)... use "tp"." Is that really the rule? It doesn't even fit the new practice of making American mass market paperbacks 7.5 inches tall. --Vasha 16:48, 10 December 2016 (UTC)

I've come across this as well several times regarding German paperbacks. The current size rules are already tight for old German paperbacks (which are often exactly 11.5 cm wide), but newer German books published as paperbacks are often exactly or a bit bigger than 19cm and very often almost (or more than) 12cm wide. Based on the current rules they must all be entered as "tp", though they are "just" paperbacks actually (at least if we think of a trade paperback as a "higher-quality paperback book"). Jens Hitspacebar 17:18, 10 December 2016 (UTC)

In publishing the term "mass market paperback" describes paperbacks which are not returned to their publisher if the retailer fails to sell them. Instead the cover is removed and returned to the publisher while the rest of the book is pulped. In most cases what publishers call "mass market paperbacks" are smaller than trade paperbacks, but there are exceptions.

However, the ISFDB uses the term differently. We use the book's dimensions as the only criterion when deciding whether it's a "pb" or a "tp". Thus all previously mentioned "tall mass market paperbacks" are considered "tp" rather than "pb".

We have discussed this issue a few times and there are concerns about the viability of the traditional "pb" format going forward. However, no changes have been made so far.

My personal take on it is that we shouldn't change the current definition of "pb". Instead we may want to consider adding one or more new codes for different types of "tp"s. Ahasuerus 19:27, 10 December 2016 (UTC)

If you don‘t want to change what pb means (and I agree that that would be problematic because it would alter the meaning of so much data that‘s already been entered), then there's definitely a need for an intermediate category for paperbacks that have a distinct small size that distinguishes them from trade. Let's say, between 7.25 and 8 inches/18-20.5 cm, or whatever exactly seems best. I do think it would be worthwhile distinguishing that size category which corresponds to a marketing category. How hard would it be to implement this? --Vasha 20:14, 10 December 2016 (UTC)

(afetr resolving an editing conflict) Adding new codes will be tricky though - no way to know where the current tps will fall. One option may be to leave the current "tp" as is and as a catch all and allow other "tp" ones -- so the old ones remain as is, one can still use "tp" if they do not know where it falls and allow better separation... I would love to be able to differentiate between small tps and the real big ones. Maybe instead of playing with the type, add a new field to allow dimensions to be added? This way the tp will basically mean "softcover and not too small" and the sizing will allow different international formats to start getting grouped together... Anniemod 20:20, 10 December 2016 (UTC)

Yes, the real objection to any sort of change is how to convert old data -- much too difficult, aside from doing things like finding German books that are published by a publisher that specializes in Taschenbücher. Maybe just an additional data point for books that are tp? "Is it under 21 cm" and for existing data this would be unspecified? Does that make sense at all? --Vasha 20:32, 10 December 2016 (UTC)

Let me first mention that it would be very easy to add another supported "binding code". It would be significantly more time-consuming to add a new field.

Having said that, one thing that we may want to do is compile a list of more granular paperback formats that we have to deal with. Off the top of my head, there are:

UK: "B Format" -- 130mm x 198mm (5.12″ x 7.80″)
UK: "C Format" -- 135mm x 216mm (5.32″ x 8.51″)
US: "Tall but thin mass market paperbacks" -- ?? dimensions

Are there other standard sizes? Ahasuerus 22:21, 10 December 2016 (UTC)

Here is a summary of the sizes that German publishers use. The standard Taschenbuch is 130 mm x 190 mm (5.12″ x 7.48″).

Here is a French printer listing formats they offer; there apparently are a variety of small formats, and googling suggests that there may be even more than the three listed on that page (11 x 17 cm, 12 x 18 cm, 11 x 20 cm). I actually found a journalist, asked to research the standard size of French paperbacks, writing: "Je n’ai pas trouvé non plus de normes sur le format du livre dit de poche" (I didn't find any standards for the format of so-called "pocket" books). --Vasha 22:54, 10 December 2016 (UTC)

Most Turkish publishers use a height of 195 mmm (7.68″), while the width varies from 125 to 135 mm (4.92″ to 5.32″). --Vasha 22:41, 10 December 2016 (UTC)

Bulgarian genre books come in 2 standard formats (except for special cases like a few series in pocket formats or a publisher trying to be interesting which can be handled separately) - both end up as tp here but they are quite distinctive (the smaller one is ~130 mm/200 mm so it will fit with the UK "B" format if we give it some leeway; the bigger one is ~160x240, which is bigger than the UK formats. I am getting exact dimensions of the two formats but these are close to it. Anniemod 22:50, 10 December 2016 (UTC)

As for the German paperback sizes linked by Vasha above: these are the current ones. But they have been smaller in earlier years (the current one for "Taschenbuch" is a lot bigger than the one from the 70s or 80s). It looks like the "tp" and "pb" sizes change over time and also vary between countries (if we only want take the size of the books into account for the "tp/pb" distinction and not the print/paper quality). The current "tp" and "pb" size rules seem to have been set for US/UK formats, probably at a time when the database contained mostly English publications. Is it possible that the distinction between "tp" and "pb" should be made language-dependent (or country-dependent)? Though this might be a bit over the top regarding rule or software changes... If I may ask a heretical question: is it necessary to record the "tp" and "pb" distinction at all? Why not just "hc" versus "sc" (softcover)? Just asking... :) Jens Hitspacebar 23:12, 10 December 2016 (UTC)

We first made the distinction between "pb" and "tp" in the mid-1990s during the initial design phase. At that time mass market paperbacks still ruled the publishing world, although their reign was coming to an end.

I think the "pb" format is worth preserving in order to accurately represent the way things were done in the 20th century. Based on current trends, mass market paperbacks will soon become a shrinking ghetto and "tp" will become the de facto equivalent of what Jens has described as "sc".

Given the number of possible permutations that we were able to identify in under an hour, I am not sure that coming up with additional "binding codes" for different dimensions would be viable. If we were to add a dozen+ new codes, it would make life more difficult for editors as well as for users. Ahasuerus 23:41, 10 December 2016 (UTC)

Actually, if you allow a slight amount of leeway, a width of 13 cm (plus or minus .5) and a height of between 19 and 20 cm covers a vast number of books being currently printed: tall American mass market, B Format, modern Taschenbuch, Turkish, the smaller Bulgarian size... --Vasha 23:58, 10 December 2016 (UTC)

That might work, but too much leeway could blur the lines between pb and tp. I just did some samples with Germam tp, and they can be just 20.4 x 13.0 cm. Jens Hitspacebar 00:23, 11 December 2016 (UTC)

How about, larger than pb, but no taller than 20 cm and no wider than 13 cm? That would probably leave out a few books that call themselves pocket size, but not very many. --Vasha 00:35, 11 December 2016 (UTC)

(unindent) It looks like the latest version of the proposal is to redefine the paperback formats as follows:

"pb" stays the same and covers paperbacks up to 185mm tall and 110mm wide
"tp" covers paperbacks that are larger than "pb", but we don't know the details
"new format X" covers paperbacks that are between 186 and 200 mm tall and between 111mm to 130mm wide
"new format Y" covers paperbacks that are >200mm tall and/or >130mm wide

Is this about right? Ahasuerus 05:02, 12 December 2016 (UTC)

That‘s so, yes--Vasha 15:35, 12 December 2016 (UTC)

In this definition the "pb" format doesn't stay the same. Its current maximum width is not 110 but 115 mm (Template:PublicationFields:PubFormat). Jens Hitspacebar 20:36, 12 December 2016 (UTC)

Oops. Sorry about that! Ahasuerus 21:31, 12 December 2016 (UTC)

The "pb" size should probably keep this maximum width (or leeway), otherwise at least a lot of the older German paperback pub records need to be changed to the "new format X". They are often exactly 115mm wide (e.g. the pubs from the big Heyne Science Fiction & Fantasy pub series). Jens Hitspacebar 20:36, 12 December 2016 (UTC)

With the non-change in pb, that sounds about right (meanwhile got my confirmation from back home - Bulgarian big format has two varieties (145×210 (the new big) and 165×240 (the old big)). Both will comfortably fit in the biggest format. The "normal books" are confirmed at 13×20 (which is where the smaller new format go) so these will be perfect - biggest book that can go in that category) :)

We may also want to start a new page "per country of origin" or something to help new and not so new editors find what format to use easier... with examples of books from that country. Anniemod 20:46, 12 December 2016 (UTC)

@Jens: that extra quarter inch was to allow for inexact trimming, and I think we need to keep allowing that; we should be able to tell (I hope) whether a book is larger because of the format or just because of trimming. If those books were always 115, then yes, they belong in the intermediate format... --Vasha 21:36, 12 December 2016 (UTC)

(By the way, whether or not this new format is added, I need to go back and find all the 19-20 cm books I mistakenly put in as pb -- would there be some efficient way of searching my edit history to pull up all the NewPub, AddPub, and ClonePub?) --Vasha 21:19, 12 December 2016 (UTC)

There is a Feature Request to "Add the ability to search submissions", but it hasn't been implemented yet. For now the most direct way to get this information is by paging through "My Recent Edits". Alternatively, if the problematic submissions followed a pattern, e.g. they were by associated with certain publishers, you can try Advanced Publication Search and limit it by publisher name and binding code. Ahasuerus 01:18, 13 December 2016 (UTC)

(unindent) I guess we'll need to have names for "new format X" and "new format Y" mentioned above before the proposal can be put up for consideration in a new section. I am not sure what they should be since they cover a number of different scenarios. Ahasuerus 01:32, 13 December 2016 (UTC)

Small tp and large tp? Anniemod 01:36, 13 December 2016 (UTC)

"tp" stands for "trade paperback", which the smaller proposed format is not. --Vasha 01:51, 13 December 2016 (UTC)

Based on what definition and in which country? The B format of UK is technically a trade paperaback. Both of those are now called "tp" around here already, we are just splitting them in two categories.:) Anniemod 01:58, 13 December 2016 (UTC)

True... the term "trade paperback" doesn't make sense in an international context. Will this be like the AARP which, when they wanted to emphasize that not all their members were American (or even retired), stated that the letters didn't stand for "American Association of Retired Persons" but were just letters? --Vasha 02:07, 13 December 2016 (UTC)

(unindent) So - are we going to be adding new formats and if so, when? :) (I am sitting on a batch of books that will fall under one of those new ones - so want to know if I should add them (and add a note so I can find them for changing later) or if I should just sit on them a bit longer. Anniemod 21:22, 15 December 2016 (UTC)

I think the next step is to do a short write-up and post it as a separate section. That way editors will be able to decide whether to vote "yes" or "no" without having to follow the intricacies of the debate.

Before we can do that, we'll need to clarify a couple of points. First, just how tall are the American "tall" mass market paperbacks? Second, what are the proposed names for "new format X" and "new format Y"? Ahasuerus 22:21, 15 December 2016 (UTC)

7 1/2 x 4 1/4 inches according to this. I have a tall mass market paperback on my night stand at home at the moment so can confirm the exact when I get there in a few hours (unless if someone has one handier) :) There might be a third format as well (wider (5 inches or a bit more than that) but not as tall as the tall mmpb - but I am not sure how many of these are around - ha, found one of my sources here). For the names - see my proposal above? Anniemod 22:32, 15 December 2016 (UTC)

That thread you linked says that those wider books are digest size, and also that they may be produced only for Wal-Mart? anyhow I can't find what the exact size is -- couldn't find them on the current Wal-Mart site. --Vasha 22:58, 15 December 2016 (UTC)

Yeah, looks like that. And just to be annoying, the tall mmpb I can get easy access to is 7 1/4 x 4 1/8. But I know that I have a few that are 7 1/2. Anniemod 01:48, 16 December 2016 (UTC)

OK, so they are in the 7.25-7.5" range.

Next, we need to clean up our math. PubFormat says "For books as tall as 7.25" (19 cm)... use "tp"." However, 7.25" is 18.145 cm rather than 19 cm; 7.5" is 19.05 cm.

We need to decide whether the cut-off point is 7.25" (and change 19 cm to 18.15 cm) or 7.5". Since we have been entering 7.25" paperbacks as tp, I vote for the first option. Ahasuerus 16:56, 19 December 2016 (UTC)

It almost starts sounding as we should rename the pb to pocket book (no rename - the name works like that) and the tp to sc (for softcover) and be done with - this way the current data will be valid, we will be closer to the hardcovers (because we do not split them there) and cause less confusions. The only issue will be books in the same series being sc and tp depending on who adds them but it will always be happening with people needing to measure their books (and we can give some guidelines per language for the big series I guess)...

On the math, I vote for going with the current inches values. Annie 17:14, 19 December 2016 (UTC)

Except see here, where Christian tells me that he has been adding German Taschenbücher up to 19 cm as "pb" all along -- lots of them, I think. There's at least one vote for redefining pb to be 7.5 inches/19.05 cm. I would be OK with that -- in that case, the B format would just be tp (or sc per Annie).

I think that defining pb as less than 19.1 cm would resolve some unclarity in existing data; not only the Taschbücher issue, but also some people, misled by the occurrence of "19 cm" in the help, may have added some books over 7.25 already -- the space between 7″ and 7.5″ is a grey area. Such a redefinition would bring "pb" a little closer to a general "pocket book".

I am not sure that having three categories would be a good idea, but if clarifying the definition of pb was to make it 19cm instead of 7.25″ I think that would be an improvement. --Vasha 05:27, 20 December 2016 (UTC)

Non-Genre Magazine Cover Art

Where do we stand on non-genre magazine cover art? The help says they are not to be entered unless they illustrate a genre story or are by a genre artist. However, we seem to have quite a few of them entered. I have an edit pending on hold related to this. The editor's note to moderator states "Not sure if I should download a cover image, but all the listed issues of this magazine have cover images." I think their confusion is justified so am seeking broader opinions. Thanks. -- JLaTondre (talk) 20:39, 17 December 2016 (UTC)

The last time the question came up, the majority was against including non-genre cover art. As I recall, one of the arguments was the sheer number of non-genre artists that we would be adding. Ahasuerus 20:48, 17 December 2016 (UTC)

The only reason we have the non-genre magazines at all is so that we do not need to invent another way to add the stories they contain (please correct me if I am wrong). So unless the cover is the story's art, it really does not have much of a space here -- otherwise how is that different from a non-genre article in the same magazine? Or a non-genre story? Anniemod 21:04, 17 December 2016 (UTC)

As somebody who has been called into the principle's office a number of times for this, I'm ambivalent about non-genre covers, unless they are directly related to the listed speculative content(s). I have posted cover images for some non-genre items because either the covers have something speculative about them, or they illustrate something speculative inside. On the other hand, I have listed something speculative in any number of Alfred Hitchcock's or Ellery Queen's and except for the covers of about two, the rest have cover images that I didn't post, wasn't aware of until recently, but which are still there. Listing the cover artist, I think, is fine, but if we don't accept cover images, somebody is going to have to go back and do a lot of deletions. Maybe a cut-off point? Like nothing past the fifties. Don't know, just asking. I suspect as more editors get involved with this site, and more non-genre sources, like the slicks and pulps that are constantly being looked at, this question is going to come up again and again in the near future. MLB 22:40, 17 December 2016 (UTC)

I'm also opposed to including cover photos or cover artists for non-genre magazines, except for those rare circumstances where the cover art is for the genre story. I would be rather uncomfortable with including a bunch of Playboy covers just because the magazine included an SF story. Chavey 06:59, 18 December 2016 (UTC)

It's a likewise standpoint with me. It'd still be possible to link to amazon, if the cover's there available. Stonecreek 19:42, 18 December 2016 (UTC)

Based on the discussion, the submission was rejected, the image was deleted, and additional images from other issues removed. Thanks. -- JLaTondre (talk) 00:10, 21 December 2016 (UTC)

Web or online exclusive supplements to a printed book

This is pretty common lately with the journals and reviews published from the various colleges. Most of them are not genre so I am not sure if it ever came up (cannot find anything - if it was already discussed, please let me know). The latest issue of Conjunctions is Speculative Fiction one (most of it for sure, they claim the whole of it - had not read it yet so cannot say). And they do have a few stories in a "Web-exclusive supplement" as usual for them lately and at first glance at least one of the stories is eligible for inclusion (if not all of them actually). So what is the policy for this kind of cases? Annie 01:59, 20 December 2016 (UTC)

As to whether to include all the contents, the policy page says to include "non-genre speculative fiction" so I guess if Conjunctions says they think these stories are speculative, they're in. --Vasha 03:15, 23 December 2016 (UTC)

I am more concerned about what to do with the online stories (do we include or not and if yes, how), not the content itself - the printed ones will go in in full. :) Annie 03:21, 23 December 2016 (UTC)