User talk:Ahasuerus

From ISFDB
Jump to navigation Jump to search

See User talk:Ahasuerus/Archive for discussions prior to 2022.

PLEASE NOTE:

If you're writing to inform me that you've either added a COVER IMAGE or NOTES to any of my VERIFIED PUBS, please follow THIS LINK and add it to the bottom of the list. A link to the pub record would be appreciated. Once the pub has been reviewed, I'll remove your note from the list. Thanks!

Author Merges & Alternate Names

I merged "J. C. H. Rigby" (148444) and "JCH Rigby" (287823) resulting in the "new" JCH Rigby. One of them (I'm not certain which at this point, but think it was the 287823 entry) had an alternate name of Charlie Rigby. When I did the merge, the alternate name was lost. That caught me by surprise. When a merge is done, the normal expectation would all the information is merged. Title merges handle variants so I would have expected author names to handle alternate names. Ideally, it should handle the alternate name. If not, there need to be a big warning that further action is required. I wonder if we have lost alternate names because of this... -- JLaTondre (talk) 09:06, 1 January 2022 (EST)

Let me take a look... Ahasuerus 11:33, 1 January 2022 (EST)
OK, I think I can see the sequence of events. Let me run the weekly/monthly backups and then I will try to recreate the problem on the development server. Ahasuerus 12:17, 1 January 2022 (EST)
I have wrapped up and deployed the cleanup reports which were in a semi-ready state yesterday. Hopefully I can recreate and fix this problem tomorrow. Ahasuerus 20:19, 2 January 2022 (EST)
I have been able to recreate the problem on the development server. Luckily, it doesn't seem to affect VTs, so any lost "alternate name" associates will appear on the cleanup report that looks for "stray" publications. Bug 795 has been created. Thanks for identifying the problem! Ahasuerus 17:11, 3 January 2022 (EST)
OK, I believe the bug has been fixed. Ahasuerus 18:06, 3 January 2022 (EST)
Out of curiosity: aren't we standardizing names anymore to J. C. H. Rigby? Regards, MagicUnk 05:32, 4 January 2022 (EST)
Originally, we always added a period and a space after an initial. As Template:PublicationFields:Author says:
  • Initials should normally be entered followed by a period and a space as "Gordon R. Dickson" or "K. D. Wentworth", even if the period or space is omitted in the publication.
However, we later realized that some authors deliberately use their initials as a "pseudo-name" like "JCH" and that the practice has been getting more common recently. That's when we updated Template:PublicationFields:Author with the following caveat:
  • However, when it is clearly the author's choice to omit the period, or when the author has a single letter name that is not an initial (e.g. "Harry S Truman") the period should be omitted. In the rare case where an author prefers two (or more) initials as if they were a name (such as "TG Theodore"), without a period or space, and is so credited, we follow the author's preference.
Ideally, our search logic would ignore periods and other punctuation in names, but we aren't there, so this s the best compromise we could come up with. Ahasuerus 10:08, 4 January 2022 (EST)
Given we can have alternate names for authors, would it make sense to create an ISFDB-standard name as an alternate for search purposes? ../Doug H 13:36, 1 February 2022 (EST)
It's been known to happen, e.g. see PS Cottier, an alternate name used by P. S. Cottier. I am not sure if this is the consensus approach at this time, though. Something to ask on the Rules and Standards page, perhaps? Ahasuerus 14:58, 1 February 2022 (EST)

Rejection Issue

I screwed up the HTML on this submission which is causing the moderator display to not show correctly. I tried editing the URL to change it to http://www.isfdb.org/cgi-bin/mod/reject.cgi?5187168, but that seems to get the ID from the post and not the URL. Need some help on how to reject this submission. Sorry for the inconvenience. -- JLaTondre (talk) 11:51, 2 January 2022 (EST)

Done. You were close -- the correct CGI script name is "hardreject" :-) Ahasuerus 12:19, 2 January 2022 (EST)

Misaligned

http://www.isfdb.org/cgi-bin/publisheryear.cgi?35762+1991; http://www.isfdb.org/cgi-bin/publisheryear.cgi?35762+1992; Another 2 cases where the ? mark after the last title causes it to be in a different row. I think you're the one who fixes this kind of stuff. --Username 11:43, 1 February 2022 (EST)

That's right, I usually handle software issues. However, I am looking at these two Web pages right now and they appear to be displayed correctly. I have tested it under Firefox, Google Chrome, MS Edge and Tor so far. Which browser and which version are you using? Ahasuerus 12:40, 1 February 2022 (EST)
I recall that months ago I brought this up and I think it was you who discovered it was some issue with covers with that rollover thing next to the title that were causing the problem, and then you fixed it; it's on 1 of these message boards somewhere. Recently, I brought it up again after discovering more of them, but was told you were in charge of that stuff, but I don't think I bothered asking you about it then. Now that I came across these 2 problems by the same publisher (in Poland), I see that the first 2 covers are displayed in 1 row and the 3rd cover (the ones with the ? next to them) are displayed below the first two, in a separate row. I believe this is exactly the same issue as the other times I brought it up, so I guess the answer is to do whatever you did that first time to fix it. I use Google Chrome on a laptop with Windows 8.1. --Username 12:49, 1 February 2022 (EST)
Oh, you are looking at http://www.isfdb.org/cgi-bin/publisheryear.cgi?35762+1991+1 and http://www.isfdb.org/cgi-bin/publisheryear.cgi?35762+1992+1 (note the additional "+1" at the end), which display cover scans for these publications. Now I see the issue. Let me check our archives to see what was causing similar problems in the past... Ahasuerus 13:06, 1 February 2022 (EST)
OK, it should be fixed now. Could you please check if everything looks OK? Thanks for reporting it! Ahasuerus 13:21, 1 February 2022 (EST)
Yes, they're fine now. I found my previous message, http://www.isfdb.org/wiki/index.php/ISFDB:Moderator_noticeboard#Jumbled_Covers, in case you want to look at those, too. --Username 13:26, 1 February 2022 (EST)
Thanks, I have updated that discussion on the Moderator Noticeboard page.
What happened was that I fixed this issue on the Title Covers page back when it was first reported some months ago, but I didn't realize that the Publisher Year page had the some problem. And then I missed your January post on the Moderator Noticeboard.
If you come across any other ISFDB pages which display mouseover help for covers and mess up the display, please let me know. Ahasuerus 14:28, 1 February 2022 (EST)

Author:W._A._Harbinson

It looks like there was only one use made of the Melvyl template. I notice you created the author wiki page Author:W._A._Harbinson and in the last sentence said what he'd been doing for 'the last eight years'. Checking the log, seems that was back in 2008. Bit misleading after 13 years. I wonder what kind of editor let that kind of thing get on the wiki. ../Doug H 14:55, 1 February 2022 (EST)

I have moved the author-provided autobiography to the database proper, clarified that it was posted in August 2006 and deleted the Wiki page. Thanks for catching it! Ahasuerus 15:08, 1 February 2022 (EST)

Dating publications - a thought

My last point in the discussion was that of verified data being sacrosanct and all else to be annotated with source. I've been digging through multiple sources to document some non-genre Asimov most recently OCLC. I dug into entry with 4 sources with a date of [1954] and got c1954, sixth printing (in associated scan only), one doesn't have the book, and the fourth connection just hung. I'm currently wrestling with how to record such references when I create a pub, because if I don't record it, someone else might either use it to create new publications or modify an existing one. With that for context, I can see that the idea of documenting sources will need to be addressed as a community. You're probably already wrestling with it, let me know if I can help. ../Doug H 12:42, 4 February 2022 (EST)

Sorry, I missed this message when it was posted. I will try to respond tomorrow. Ahasuerus 00:17, 8 February 2022 (EST)
Yes, secondary sources like OCLC and other online catalogs present a variety of challenges. At one point we tried to describe them in Help:How to parse data in library catalogs and Help:Using Worldcat data, but it's been a number of years since the instructions were last updated and they are still incomplete. For example, the former says "Need to add an explanation of the MARC family of standards, SUTRS and the OCLC guidelines here", which hasn't been touched since 2008.
If you have ideas/suggestions re: improving these Help pages, creating new ones or documenting third party information in Notes, please don't hesitate to post them on the Community Portal! Ahasuerus 09:01, 8 February 2022 (EST)
I've added some text to the Help Entry for parsing (sans discussion on a Portal as it is background and context) and came to the realization that I don't know why you'd want the MARC family of standards (et al) on there. OCLC won't give you the MARC version (as far as I can tell) and all it does is format what's displayed. So unless one plans on automating the pull, I don't see why one would want the machine-readable format. Particularly since it is notoriously difficult to parse. ../Doug H 15:53, 9 February 2022 (EST)

Mismatched Double Quotes

The Mismatched Double Quotes cleanup report might need an ignore option. A single double quote is a valid abbreviation for inch (which is the usage in the two current records). Or have it ignore word boundary + number + double quote (\w\d+"). -- JLaTondre (talk) 21:37, 7 February 2022 (EST)

Good point. FR 1485, "Allow ignoring records on the Mismatched Double Quotes cleanup report", has been created. Ahasuerus 08:54, 8 February 2022 (EST)
After examining the code, I see that this is a bigger mess than I realized. The report is actually limited to records with notes which contain mismatched double quotes AND the string "http". In other words, it's looking for malformed URLs. The logic works most of the time, but notes which contain a URL plus a separate mismatched quote throw it off. In addition, we have another cleanup report, "Invalid HREFs in Publication Notes", which does something similar -- but not the same -- for Publication notes only.
Let me see if I can expand the second report to cover all record types and then fold the first report into it. Ahasuerus 10:45, 8 February 2022 (EST)
Done. Ahasuerus 17:16, 8 February 2022 (EST)
Thanks. -- JLaTondre (talk) 18:39, 8 February 2022 (EST)

Info for authors

If you were to talk to a group of authors about ISFDB, what would be the first things you'd share with them? I'm going to be doing that, and I'd be interested to know the things you'd think are most important for them to know. Thanks! ···日本穣 · 投稿 · Talk to Nihonjoe 17:20, 8 February 2022 (EST)

Off the top of my head:
  • Eligibility - what cannot be added to the DB. Their site or blog are not producing a "published" story. That magazine in Lithuania is.
  • The fact that we are a historical DB - if you EVER published it, we won't delete it. No matter how ashamed you are of it.
  • They won't see the magazines and anthologies they have stories in on their pages - they will see the stories themselves.
  • If they need help, we will assist them by adding or editing or teaching them how to add to the DB. They are welcome to add their own stories and novels or have someone else add for them or ask for help for someone to add (and that includes translations and we have multi-language editors to assist).
For example... I may have done some of it with some of authors :) Annie 17:34, 8 February 2022 (EST)
Apologies for intruding, but a couple of observations/suggestions, based on stuff I've seen on social media:
* Re. eligibility, I don't think it's obvious to newbies that the way to add stories is via the pubs they appear in. See this exchange for an example
* It's not at all obvious to visitors that the primary means of communication is via the Wiki, hence stuff like that prior example or this one of authors crying into the void for help.
* I've also seen ISFDB described as a wiki, which I guess it is in the general sense of a community of contributed info, but not if you're thinking of stuff like MediaWiki, Confluence, Sharepoint (or whatever MS offer in that vein these days). In particular, it might be worth pointing out that having all submissions be moderator approved means that data should be more accurate than Wikipedia, Goodreads, etc ErsatzCulture 18:42, 8 February 2022 (EST)
How about asking them to make sure their full names, correct birth dates/places of birth, and updated websites (if any) are in their records; today I corrected over a dozen "Calfornia" misspellings for places of birth, plus many records have just a year for birth dates with no day or month (many death dates are also questionable, but you can't ask those authors because they're dead). --Username 18:48, 8 February 2022 (EST)
I would probably present it as two separate questions. The first one is "Why is bibliography important when you are an author?" The second one is "What makes ISFDB stand out among other online bibliographies?"
The primary answer to the first question is "Assistance with discovery". When you finish reading an exciting book/story, one of the first questions that you ask yourself is whether there are more books by the same author or similar books by similar authors. A comprehensive bibliography can answer the first question and help with the second question. For example, Mountaindale Press specializes in LitRPG, so it's a good place to check out if you are interested in LitRPG. Ditto Yen On, Airship, and J-Novel Club for light novels. Tor and Baen have their fans who pick books based on the publisher. Major media franchises like Star Trek, Star Wars, Warhammer, etc are good places for new authors to put their name on one or more bestseller lists. Etc. A bibliography which lets readers hop from author to series to publisher to author ad infinitum is a useful discovery tool.
The answer to the second question is that no single bibliography is perfect, but ISFDB has certain advantages compared to other online alternatives. Goodreads has more ISBNs and ASINs, but they are not as well organized and searching for short fiction can be challenging. Amazon has a LOT more ISBNs and ASINs, but, again, organization and short fiction can be problematic. On the flip side, they have reviews and more robust tag systems.
Finally, I have seen editors comment on how useful the ISFDB database can be when putting together anthologies and collections. It may not benefit authors directly, but it's likely an indirect benefit. Ahasuerus 18:56, 8 February 2022 (EST)
That's a good way to look at it - as long as the discoverability is paired with the "but we keep records of all the books we know of, even if you they are not available or you do not want them to be listed or if they were authorized editions". :) Annie 19:27, 8 February 2022 (EST)
Re: Eligibility: Thus my third point - which pivots into explaining how you add your stories :) Annie 19:27, 8 February 2022 (EST)
This is all great information, everyone. Thanks! ···日本穣 · 投稿 · Talk to Nihonjoe 16:19, 9 February 2022 (EST)

Ossuary

http://www.isfdb.org/cgi-bin/pl.cgi?647581; You entered this more than 4 years ago and today I saw it on FantLab. I added cover and fixed publisher, but there's a photo with contents that differ slightly from what you noted, plus there's a cover art credit but I don't see a name on the cover so I didn't enter it; if you're interested you might want to update some things. --Username 11:50, 9 February 2022 (EST)

Updated, thanks. Ahasuerus 13:24, 9 February 2022 (EST)
BTW, there's only 1 Kotarbinsky on ISFDB, Vasily, http://www.isfdb.org/cgi-bin/ea.cgi?271194, with an interior art credit on nearly the same date, so that's probably who it is. --Username 13:31, 9 February 2022 (EST)
Updated, thanks. Ahasuerus 16:47, 9 February 2022 (EST)

Just double-checking

To make sure my understanding is correct. Thanks! ···日本穣 · 投稿 · Talk to Nihonjoe 14:38, 9 February 2022 (EST)

Response posted. Ahasuerus 16:26, 9 February 2022 (EST)

Fateful Lightning cover - non /P/isbn cover

I've just attached a cover to [1] (when the edit goes through) which might match the printing you've PVed. --GlennMcG 01:26, 10 February 2022 (EST)

Yes, my first printing has the same cover art plus the following thin line:
  • ROC * 451-LE5196 * (CANADA $6.99) * U.S. $5.99
at the bottom. I have also confirmed that it's the same cover as the one used on the fifth printing published in 1998-01, which Locus1 attributes to Sanjulian, so now we have the name of the cover artist. Thanks. Ahasuerus 10:20, 10 February 2022 (EST)

Account Recovery

Hello!

Attempted to login yesterday and today with no success. Granted it has been a long while... Used the Wiki feature to reset my password, and it claims an email was sent, but I've been checking my inbox and spam folder all day to no avail.

From what I recall the Wiki was always... quirky. I suspect that hasn't changed, so curious if this is a glitch of some sort or something actually wrong on my end. Regardless, not sure what my options are, or if I should just plan on soldiering on with a new username. Wasn't sure where else to turn.

Cheers, --Albinoflea-Spurned 21:24, 22 February 2022 (EST) (The user/former moderator once known as Albinoflea)

Welcome back :) Mails are... not working with all servers. Let's hope Ahasuerus can help! :) Annie 22:13, 22 February 2022 (EST)
Welcome back! Let's start with the basics -- are you still using s******.f*****@gmail.com as your email address? The reason I am asking is that I don't see this email address in the ISFDB server's outgoing mail queue, so I wonder if the address may have been changed. Ahasuerus 23:22, 22 February 2022 (EST)
Yes, that email is correct, and I still have copies of emails I received from ISFDB users and even an old password reset email from 2014 that were sent to my inbox. Albinoflea-Spurned 14:41, 23 February 2022 (EST)
I see. I have just changed your password using a Wiki maintenance script and sent the new password to your Gmail account. Could you please let me know when you get the email message and whether the new password worked? Ahasuerus 16:18, 23 February 2022 (EST)
Thanks, that one went to spam but I retrieved it without any trouble. Logged in, updated password and all seems good. Thanks! Albinoflea 16:52, 23 February 2022 (EST)
Excellent! :-) Ahasuerus 16:54, 23 February 2022 (EST)

Importing Content Labeled as Proposed Clone

Please see this submission: This is an import content submission, but the title at the top of the moderator screen is listed as "Proposed Clone Publication Submission". On the public view, it correctly states "Pending ImportExport Submission". Thanks. -- JLaTondre (talk) 08:38, 24 February 2022 (EST)

Yup, it's a known issue on my list of things to fix -- Bug 787, "Import/Export submission review has wrong header". Ahasuerus 11:31, 24 February 2022 (EST)
And fixed. Ahasuerus 17:58, 24 February 2022 (EST)
Thanks. -- JLaTondre (talk) 18:06, 27 February 2022 (EST)

Nav bar: something a bit disorienting for a new self-approver

When I clicked on a "Self-approver view" link (e.g. http://www.isfdb.org/cgi-bin/mod/submission_review.cgi?5243123) this added the "Moderator Links" section to the bottom of the nav bar. I wasn't sure what delights might await on those pages, so I clicked on each of the links, but they all give a "Moderator privileges are required for this option" page. My initial thought was that there was maybe some cookie/session magic involved, so I logged out and back in, and was a bit confused when that Moderator Links section had disappeared.

Having gone back through my browser history, and checked out the code, I can see now that that stuff gets added to the nav bar for any page with code in the /mod/ subdirectory. Assuming there aren't any cases where those sidebar links do work for self approvers, it might be nice to wrap the coder that outputs that section to be wrapped in a 'if moderator:' test, and so not render them for editors who can't make use of them?

This would be a very low priority change, as I guess the number of people affected is in the single digits, and they'll all work it out, but maybe it's a case of adding the if test and indenting the block of lines that output those links?

(I can see the code a bit further down mod/isfdblib.py references moderator, self_approver and SelfApprovalAllowed(userid), so perhaps there are complexities involved in the privileges that make this not as trivial as I might have made out though?) —The preceding unsigned comment was added by ErsatzCulture (talk • contribs) .

The "Self-approver view" link was added to the Pending Submission page just last week, so it's probably an oversight. Let me take a closer look... Ahasuerus 17:55, 25 February 2022 (EST)
OK, I think I got it. Could you please try again and see if everything looks OK on your end? Ahasuerus 18:20, 25 February 2022 (EST)
Thanks - I've just been through the edit/approval cycle a couple of times and didn't see the mod links appear. ErsatzCulture 18:32, 25 February 2022 (EST)
Excellent! :) Ahasuerus 18:35, 25 February 2022 (EST)

Reports Paging

Hi Ahasuerus

I've been working through my list of unstable amazon images and I noticed an entirely inconsequential bug with the paging. I just got the report down to one page, 200 rows, and there was still a link to the second page of records. Clicking the link brings you to an empty page. I've since dropped below 200 rows, and there is no link to the second page, as one would expect. I'm guessing that this is the difference between >= vs >. Although it's completely trivial, I thought I'd bring it to your attention. Thanks. --Ron ~ RtraceTalk 07:12, 3 March 2022 (EST)

Thanks, I'll take a look. Appreciate the heads-up! Ahasuerus 10:20, 3 March 2022 (EST)
OK, I see what's going on. Basically, the SQL query says "Give me the next 200 matches". If the database engine returns less than 200 matches, then the software doesn't display a link to the next page. If the engine returns 200 matches, then a link is displayed.
The problem with this logic is that it doesn't distinguish between "there are exactly 200 matches on file" and "there are more than 200 matches on file". What the software needs to do is request 201 matches. I'll see what I can do after finishing the round of changes that I currently have in the works. Ahasuerus 13:39, 3 March 2022 (EST)
OK, the 3 "My Primary Verifications" tables have been fixed. I still need to fix a bunch of other tables that have the same issue, but it may take a bit since they are all somewhat different. Ahasuerus 16:55, 4 March 2022 (EST)

Top Contributors et al

Hi Ahasuerus

I suspect that the Top Contributors and similar reports are stuck again. The most recent Last User Activity is for March 6. As before, this isn't urgent, but I thought you'd want to know. Thanks. --Ron ~ RtraceTalk 07:10, 8 March 2022 (EST)

All "Database Statistics" reports were moved to a weekly schedule last week -- see Nightly processing getting split for details. Most of them say "This report is generated once a week" at the top of the page, but I see that "Top Contributors" ones do not. Let me add that real quick... Ahasuerus 10:12, 8 March 2022 (EST)
Done -- see Top ISFDB contributors (All Submission Types). Thanks for the heads-up. Ahasuerus 12:01, 8 March 2022 (EST)
I won't say that I'm not disappointed, but I do understand the necessity of the change. I recall when these reports were on demand, and I did adjust after they went daily. Thanks for the explanation. --Ron ~ RtraceTalk 21:41, 8 March 2022 (EST)
In a way, we are victims of our success. The more records and submissions we have, the longer it takes to generate reports. Unfortunately, many reports draw data from multiple tables and lock them while the data is compiled.
Ideally, we would move all of our report-generating processes to a different server and then shuffle the backups and the compiled data back and forth between the production server and the backup server, but it would require a fair amount of juggling. Ahasuerus 22:23, 8 March 2022 (EST)

Patches, testing and a proposal

Hi. Sorry for not submitting any further patches after the last one. Offline life caught up to me and suddenly I have a lot less spare time. /Lokal_Profil 18:08, 13 March 2022 (EDT)

No worries whatsoever. Real life happens. Ahasuerus 19:47, 13 March 2022 (EDT)

I struggled a fair bit with setting up a local development environment. I simply don't have space for a full virtual machine running the right version of sql etc. so I aimed at just getting the python bit up and running. After some hacking I managed to get a virtual environment up and running for Python 2.5.4. My plan was to introduce some unit tests so that I could at least do some initial refactoring with a fairly high certainty that it wouldn't have any undesired effects. What I then came across was that the imports rely on the required python files from the common directory being copied across to the other directories when the system is built. This put a stop to my unit test plans.

What I wanted to check with you is if it would be of interest if I re-arranged the import and folder structure as a package instead. This would make unit testing possible but should also simplify both a future migration to a newer python version and make the codebase easier to understand for new developers. Even if it might be possible to split the patch up into a few incremental changes the end result still touches most of the codebase and as such would require extensive testing. I can therefore see how it might not be a desired change. If you want me to go ahead with it I will try to get it done in my spare time and let you know when it is ready for testing. If you don't then I fully understand =) Cheers, /Lokal_Profil 18:08, 13 March 2022 (EDT)

At one point I looked into restructuring our copy/import system. What I found was that the parts of Python which support packages and imports were significantly changed and expanded in Python 2.6 and 2.7. I figured it would be better to wait until we could upgrade to 2.7 and then revisit the issue. More recently, Al started looking into a possible Python 3.x upgrade -- see Development/HTTPS -- which may add another wrinkle. Ahasuerus 19:47, 13 March 2022 (EDT)
I believe the main changes are to how relative imports are handled. Since absolute importa are generally recommended I would have gone for that (should then still work for python 3). I believe the change is desirable before the python 3 migration (since it allows for easier testing) but can probably happen before or after a python 2.7 migration. Happy to wait until after such a migration but I'm also equally happy to go ahead as a preparatory step for auch a migration. /Lokal_Profil 05:07, 14 March 2022 (EDT)
Retrieved from "https://isfdb.org/wiki/index.php?title=User_talk:Ahasuerus&oldid=629333"