ISFDB:Moderator noticeboard/Archive 02

From ISFDB

Jump to: navigation, search

This is an archive page for the Moderator noticeboard. Please do not edit the contents. To start a new discussion, please click here.
This archive includes discussions from June 2007 - January 2008.

Archive Quick Links
Archives of old discussions from the Moderator noticeboard.


1 · 2 · 3 · 4 · 5 · 6 · 7 · 8 · 9 · 10 · 11 · 12 · 13 · 14 · 15 · 16 · 17 · 18 · 19 · 20 · 21 · 22 · 23 · 24 · 25 · 26 · 27 · 28 · 29 · 30 · 31


Expanded archive listing


Contents

Perry Rhodan

Got one sitting in the queue for a couple of days waiting for approval or some notice of what might be wrong with it. Two previous ones approved with no changes. Any ideas? rbh 21:33, 28 Jun 2007 (CDT)

Ahasuerus placed it on hold. He's usually rather good at going back to the original editor if he has questions about the submission. It appears that he's not been on the site today. Mhhutchins 21:47, 28 Jun 2007 (CDT)
Sorry, folks, I had network problems earlier today and then had to play the exciting game of whack-a-mole with the vandalbots attacking the Wiki. Will get back to you shortly! Ahasuerus 22:50, 28 Jun 2007 (CDT)
Thanks for the help, I responded on your comments on my user page. rbh 22:04, 29 Jun 2007 (CDT)

Watchlists and Moderating

Is there a quick and easwy way to add multiple pages at once to a watchlist? CoachPaul 08:07, 3 Jul 2007 (CDT)

I don't think you can add multiple pages (somebody else might know) but there is one handy option in Preferences - Editing that places any page that you edit on your watchlist. If you ask a question of somebody on their talk page, that talk page will be on your watchlist. This comes in handy if they answer your question on their talk page. Once all the communciations are done you can manually remove their talk page from your watchlist.--swfritter 10:36, 5 Jul 2007 (CDT)
Thank you. CoachPaul 10:42, 5 Jul 2007 (CDT)
At one time I did an experiment with watch-listing *all* of the ISFDB wiki pages but did that with a script that built the page list from Special:Allpages and then sent in URLs using wget of the format http://isfdb.tamu.edu/wiki/index.php?title=(page_name)&action=watch. It hammered ISFDB for a while and I later decided that method did not do what I wanted and so I sent in another script to unwatch the pages. Marc Kupper (talk) 00:50, 13 Jul 2007 (CDT)


Do we know when new editors sign up, or only when they start editing? CoachPaul 08:07, 3 Jul 2007 (CDT)

Not that I know of. Every so often I check Special:Recentchanges, http://www.isfdb.org/cgi-bin/topcontrib.cgi, and http://www.isfdb.org/cgi-bin/topvers.cgi for new people by watching the link colors. There is Special:Listusers but the list has 2697 names at present with many of them being spam accounts. It looks like with [1] I could see the entire user list on a single page meaning I could automate detection of new people. If the block stuff worked correctly I would automate detection and auto-blocking of the spammers. For example, 1423 names in the list are 6 characters long, of that 1241 names have an upper case 4th letter, 1240 have a lower case second letter (only MAGNUS removed from the list), and I suppose I could test further but it looks like the list only has spammer accounts. Maybe the code for wpCreateaccount could be modified to just block creation of accounts matching this pattern... Ah yes, see [2] which deals with blocking account creation based on the IP block list and so this could also be fixed to block based on the name and the spammer thinks they are IP blocked... Marc Kupper (talk) 01:20, 13 Jul 2007 (CDT)

Moderating Help

I've approved several edits that I now think are in need of fixing. Should I fix them and then send a polite note to the editor telling them of the fix, or should I send a polite note to the Editor asking them to fix it with the reason why, or should I just go ahead and fix it and not say anything to the Editor? CoachPaul 10:21, 5 Jul 2007 (CDT)

Typically, I correct obvious typos and minor brainos without making any comments. However, if an error suggests that the editor is unfamiliar with some ISFDB conventions, I leave a note explaining what went wrong and how I fixed it -- see User talk:Ali for a recent example. On the gripping hand, if there is a question as to what the editor meant in the first place, then I put the submission on hold and ask the editor to clarify. Ahasuerus 10:55, 5 Jul 2007 (CDT)
For things like like forgetting to put the # in front of a catalog #, mixing up the cover artist and image URL fields, etc. I almost always just quietly edit/approve, even if it's for brand new people. I sometimes quietly deal with people making changes to title records in the contents. If someone gets in the habit of making a particular error I'll mention it to them. Marc Kupper (talk) 00:44, 13 Jul 2007 (CDT)
For my part, I usually try and think "positive and welcoming". So I do go quietly fix things, thank them for their activity, and gently introduce them to our conventions. Or if I'm not in a "positive and welcoming" state of mind, leave it to someone else. ;-)
Of course, I'll HOLD a really destructive edit, but usually I like to fix a few things to make them feel welcome and useful before I'll go point out all their errors - even if I've had to correct their price or date entry on every single edit so far, for instance. Sometimes it takes a reject with a note to go read their talk-page to make them start talking, but if that happens immediately I think it puts people off. Ahasuerus mentions User talk:Ali for a recent example - that may actually be an example of how NOT to approach it. :-/ Yes, there's two "welcomes" and a "thanks" in there, but four complaints. An apologetic "thanks for the edits, but did you realise that the YYYY-MM-DD format date is preferred, you're making me do some additional work to correct it from your DD-MON-YYYY entries?" might be a good start (not in this case, there were different errors), but a one-by-one flow of tips and advice might be better. An "ever-so-humble" attitude might keep a few more editors around - and who else is going to take over when we burn out? :-/ BLongley 18:40, 13 Jul 2007 (CDT)
Oh, I am all for welcoming new editors with open arms, "not biting the newbies", etc. Unfortunately, the Wiki is a "text-only" medium and the accompanying body language gets lost in the ether. Perhaps we could add a tea and crumpets image to our {{welcome}} template:
afternoon-tea-and-crumpets.jpg
? :) Ahasuerus 00:31, 14 Jul 2007 (CDT)

Moderation Glitch?

I went to approve an editor's edit of making a pub a variant title of another, and the "Approval" Page showed that the newer copy was going to become the parent of the older copy, so I rejected the edit. Then I went to make the older copy the parent of the newer copy myself, and the "Approval" Page, once again showed that the newer copy was to be the parent. So I approved the edit, and when I went to check on it, the older copy was the parent. Has this been noticed before? CoachPaul 13:44, 10 Jul 2007 (CDT)

Oh yes, it's our old friend DisplayBug 20098. Quite annoying :( Ahasuerus 13:49, 10 Jul 2007 (CDT)
I find crossing my eyes helps :-)Kraang 19:07, 10 Jul 2007 (CDT)
This should go on the "Advice to New Moderators" page that we haven't actually got yet. We're not very good at doing help pages yet, are we? :-/ BLongley 18:43, 13 Jul 2007 (CDT)
That's OK, I can never find anything again after I read it the first time. CoachPaul 19:19, 13 Jul 2007 (CDT)
I have added a note documenting this bug to the Moderator Help screen. Ahasuerus 00:35, 14 Jul 2007 (CDT)

Calling Any Moderator

I was processing a few Bester books when I ran into the following:
  • The Computer Connection shows VT's of Indian Giver and Extro
  • Golem 100 (in the awards display) shows a VT of The Computer Connection
Being of a sometimes curious nature, I looked at the text of the two volumes. They didn't compare in any way.

So, I guess the problem is fixing this (assuming it needs fixing). Need some assistance/guidance for this one.

--Dsorgen 21:10, 1 Aug 2007 (CDT)

As far as I can tell, at some point we imported a malformed award for Golem 100. The current version of the Locus Award database lists it as simply Golem 100. Unfortunately, there is no easy way to fix the award record at this time because the award editor module (which Al started late last year) is still in its infancy and had to be taken off-line some time in January since it was doing more harm than good. Al may be able to fix the data by hand, but it's likely safer to wait for him to wrap up his move to Texas and other real life issues. Once he is again available, the award editor will be at the top of his list of priorities. Ahasuerus 21:31, 1 Aug 2007 (CDT)
That's pretty much what I suspected had happened. I hope the move goes OK. I've lived in Texas for 23 years (San Antonio) and like it. It gets pretty hot in the summer, but stays relatively warm in winter. A nice trade-off.
Regards, --Dsorgen 21:48, 1 Aug 2007 (CDT)

Python Error

While editing Bova's "Mars" I received the following from the Python interpreter:
PythonError (IndexError) List index out of range
The error occurred while trying to display the reviews (i.e., the Reviews heading was displayed). Just thought someone(?) ought to know.
--Dsorgen 22:39, 2 Aug 2007 (CDT)

This one? Sorted. Another orphaned review record. BLongley 14:31, 3 Aug 2007 (CDT)
Related to this problem, no doubt. As we don't catch them till later, I'm not sure EXACTLY what is causing it, but if a Mod sees a change of a Review record being submitted, can they please check for an orphan record left under the original reviewer's name? I suspect there's a similar problem with cover-artists too, but have no examples. BLongley 14:38, 3 Aug 2007 (CDT)

Blocking spammers

Please be extra careful about blocking spammers via the user-block mechanism. I've been contacted by several people in the past few days because they got hit by the auto-blocks. With the current layout of the ISFDB infrastructure (as provided by TAMU) and Media Wiki software user-blocks will block a fair percentage of legitimate editors.

I have been blocking out the spammers by using the page protection mechanism. It means regular editors can't edit the pages and instead need to post a notice elsewhere asking that a page be updated. Page protection is transparent to ISFDB moderators. I have played with adding {{protected}} to the top of pages that I protect. This inserts a message informing editors about why the page is protected and informing them to leave a message on the community portal.

The only pages I don't protect are:

  • Moderator talk pages so that regular users can leave messages.
  • Non-moderator user and user-talk pages.

Note - my ISFDB time is very limited for now and so I have not been monitoring things. I just removed a bunch of blocks but could a moderator with a little more free time please protect these pages as they are recent spam targets? If a page is normally empty (either it's blank or hasa red link below) then just add the {{protected}} template and protect that. Thank you.

  • (I snipped out the list as I've dealt with it - click "edit" to see the list) ~Marc

Marc Kupper (talk) 13:47, 13 Aug 2007 (CDT)

I protected the list of pages that's above. I did not add {{protected}} to the template pages used by the help system. Marc Kupper (talk) 23:43, 13 Aug 2007 (CDT)

To explain the problem more technically, when I was blocked myself, the system said that my IP was 10.95.0.241, which it patently isn't, and in fact isn't a real IP at all - 10.* is reserved for networks' internal use. Where the proxy thing comes into play.
I think that unblocking all those spammer accounts was a bit too much; the solution mentioned elsewhere, i. e. patrolling the Special:Log/block and removing just all autoblocks in the form "User:#[number, currently over 1000]" seems to be enough, and keeps the spammers out of circulation.
AFAIK newer versions of MediaWiki have better tools for this: first, you can block just an offending acount without any effect on others who share the IP; second, it has a semi-protection option, which protects the page just from the anonymous and newly-registered accounts but allows access to non-admin users of long history (so it could be used almost anywhere). I'm not sure, and it's rather difficult to find, how much of this is implemented in the newest (roughly a year old) version the hosting here allows to use, but we can hope... --JVjr 05:55, 14 Aug 2007 (CDT)
I much prefer the Protection method than the User-blocking method. The first round(s) of blocking spambot accounts as and when they were spotted led to me having to unblock MYSELF up to 6 times a day, and I don't know how many potential new editors have been put off for good. But that seemed to settle down - only SOME accounts seem to lead to auto-blocking of real users, so we could probably have left the rest blocked. However, the accounts that we HAVE to unblock do seem to return to the same pages again and again, and so can be easily foiled by protecting the page they attack. BLongley 12:56, 14 Aug 2007 (CDT)
We actually had quite a reduction in the amount of protecting/blocking we had to do for a while, but the mass unblocking/reblocking/protecting/unprotecting isn't helping us. Having some mods go back to blocking spambot accounts is causing more unblocking work for the rest, and unblocking ones that didn't need to be is causing more spam and more work. I don't know how many real users we've put off by the blocking, but the protecting seems to have caused very few people any troubles. BLongley 12:56, 14 Aug 2007 (CDT)
Still, I too look forward to learning what the next upgrade, however trivial, will give us to fight with. BLongley 12:56, 14 Aug 2007 (CDT)
I believe that if I remove an auto-block that it can reappear which is why I removed the user blocks that triggered the autoblocks. What was happening is in the past few days I got several e-mails from random editors saying they were blocked. This is the last thing I want happening as it discourages people from using ISFDB. Thus I removed nearly all of the user blocks and instead protected the page each of those user accounts was attacking. This increases the availability of ISFDB to everyone (editors and moderators) and the main thing we loose is the ability for editors to modify the protected pages. When practical I added {{protected}} so that editors understand why they are blocked.
I agree fully that newer versions of MediaWiki offer much better tools to deal with this issue. However, Al is the only one that knows the passwords. If I knew the ftp password I'd have at least patched the code to block out the spammer accounts long ago while we worked through the much harder and time-consuming issue of upgrading MediaWiki. It's not a simple upgrade as the ISFDB code accesses MediaWiki tables directly to implement a couple of features and the table structure has changed enough in the past two years that the ISFDB code does not work at all with the current MediaWiki (I tried this).
While waiting for Al to reappear, if/when I get a chunk of available time I want to look into {{protected}} more as I'd rather that not display a message until someone tries to edit a protected page. Again, this is something I know I can do with newer versions of MediaWiki but I believe I can coerce the current MediaWiki to do something similar. Marc Kupper (talk) 13:15, 14 Aug 2007 (CDT)
A minor, but perhaps relevant, note. As it was now "old news" and noise for this thread I tried to hide the list of pages above by wrapping it in <div style="display: none"> and </div>. Preview showed this would work but when I tried to save it I got this message
Spam protection filter
From ISFDB
The page you wanted to save was blocked by the spam filter. This is probably caused by a link to an external site.
The following text is what triggered our spam filter: <div style="display: none">
Return to Main Page.
I was not aware that this version of MediaWiki had any filtering at all. Marc Kupper (talk) 13:23, 14 Aug 2007 (CDT)
The solution proposed by Jan, i.e. "patrolling the Special:Log/block and removing just all autoblocks in the form "User:#[number, currently over 1000]", might be workable if we could automate the process and have a bot checking the block log every few seconds. Unfortunately, the current (manual) system of unblocking is too slow to catch these bad blocks in time and causes more problems than it solves. The page protection scheme is, admittedly, an ugly temporary workaround, but its impact on our editors is minimal compared to the inconvenience and frustration of being sporadically locked out. Ahasuerus 15:31, 14 Aug 2007 (CDT)
Patrolling probably looks good to those in the right timezones that see problems resolved fast, but I frequently come on and see several hours backlog - weekends especially. Looking at the "moderator availability" suggests that some people are understating how much they actually DO do, but I don't want people to depend on that.
I know I frequently get tired of moderating and want to go back to editing. Which probably annoys a few editors after I approve a few, comment on some, leave the rest. But I need my fun times here too. Playing Whack-a-Mole is definitely not fun any more. Teaching new editors isn't always fun either, but I'd rather concentrate on that so we get more Mods to help with self-approvals at least: but we need to resolve a few "This Mod says this, That Mod says that, what's going on here?" issues before I can see the "ISFDB: The Next Generation" Moderators appearing. BLongley 16:39, 14 Aug 2007 (CDT)
Two people have contacted me this evening about getting unblocked. I'm assuming there are more that were unable to track down my e-mail. At the moment there are 15 blocks related to spammer accounts that hit user-talk pages. I'm mulling over either removing all of them and abandoning the block system or just removing the auto-blocks as they come up and seeing if perhaps there's only one autoblock per entry (of the 15) and once removed autoblocks will not happen. Unfortunately, the Protection Log does not show the details of the original autoblock. Marc Kupper (talk) 01:25, 16 Aug 2007 (CDT)
I'm cleaning out my e-mail in-box and ran across a block noticed I had e-mailed someone. I clicked Edit to add it to the block log below and got blocked! :( So far the good news seems to be there's only a couple of blocks per day. Marc Kupper (talk) 19:49, 17 Aug 2007 (CDT)

Unblock log

Moderators, when you patrol the block list and remove an auto-block could you please copy/paste the details here? I wish I had done that with the autoblock I removed an hour ago - I've copy/pasted what an editor e-mailed to me below. I want to research if the same block, the one for User talk:Tjacksonking in this case, is trigging multiple auto-blocks or if there will only be one autoblock for that block. If we are getting multiple autoblocks then I believe we should remove all of the blocks. Marc Kupper (talk) 01:25, 16 Aug 2007 (CDT)

Swfritter - on 17 Aug 2007 around 16:06 you cleared two blocks. You listed #1917 below. The second was #1913. Do you recall any details from this? I was reviewing the block log to get a sense of how often we get auto-blocks and saw this. I'm happy with what I saw in that the autoblocks are rare. Marc Kupper (talk) 14:30, 18 Aug 2007 (CDT)
BLongley reported unblocking an autoblock created by a user block for User:JhyT74 that spams User talk:JVjr. The puzzle for me is that user's contributions is empty and JVjr's history does not show any spam. Marc Kupper (talk) 14:59, 18 Aug 2007 (CDT)
JVjr does weird things with his talk page anyway, he might have removed the history you're interested in: e.g.
20 Aug 2007
* (Deletion log); 11:29 . . JVjr (Talk | block) (restored "User talk:JVjr": 25 revisions restored)
* (Deletion log); 11:27 . . JVjr (Talk | block) (deleted "User talk:JVjr": remove more spam, and mistaken (?) blanking)
Anyway, I've just unblocked myself from the User:JhyT74 block again, so go find out why you're blamed this time instead of Grendelkhan. ;-) BLongley 15:19, 21 Aug 2007 (CDT)
Thank you guys - please keep adding entries as I'm trying to see of one of these theories is true
  • When a user block is installed then later there will be one random autoblock triggered for any of the blocks.
  • The autoblocks are triggered by the spammer coming back or some other event.
I sure wish Al was around to dump the raw block log as things like the IP address from Squid still seem rather mysterious. The reports I've seen only talk about 10.95.0.206 and lately 10.95.0.241 being blocked implying very few IP addresses are in use and the triggering mechanism is something else. Marc Kupper (talk) 13:26, 19 Aug 2007 (CDT)
After looking at JVjr's weirdness I revised my own block procedure. If an empty page is spammed (check the history) then I delete the page using the comment "Spammed by xyzzy" where xyzzy is the spammer account, I then edit it to add {{protected}} and finally I protect the page both with the same "Spammed by xyzzy" comment. This wipes out both the page history and the spammer account's "contributions" history though both of these can be recovered by undelete. The "good news" is that it also wipes out the original spam from Recent Changes though that recent-changes will show your addition of {{protected}}. Thus Recent Changes currently only shows spams of pages that already have data and a history of that data being entered that we may want to preserve.
I'm thinking of one new template, {{empty}} that would be the same as {{protected}} but notes it's an empty page. The thinking behind {{empty}} is when we finally upgrade the wiki we can then look at the "what links here" for Template:Empty and can just mass-delete those pages including the spammer history. Marc Kupper (talk) 20:37, 21 Aug 2007 (CDT)
  • 01:07, 24 Aug 2007, Marc Kupper blocked #2061 (expires 01:07, 25 Aug 2007) (unblock) (Autoblocked because you share an IP address with "JhyT74". Reason "Spams User talk:JVjr".)Kraang 20:34, 23 Aug 2007 (CDT)
  • 19:05, 23 Aug 2007, Marc Kupper blocked #2065 (expires 19:05, 24 Aug 2007) (unblock) (Autoblocked because you share an IP address with "DbyLys". Reason "Spams User talk:Dcarson".)
Well guys - I had been hoping the autoblocks would die out. Which is the lesser evil? The autoblocks or spams and reversals on user's talk pages? I did a test of creating a sub-page of my user talk page to see if I would get notified of updates but unfortunately that's not the case. I was hoping I would get notified meaning we could protect our talk pages and just shift to a sub-page.
  • My personal vote would be to remove all of the user blocks and to continue with protecting pages (other than active user pages) as they become spam targets. Marc Kupper (talk) 00:10, 23 Aug 2007 (CDT)
  • I am in favor of removing blocks as well. We seem to have little control over them and the unintended consequences of the cure are far worse than the disease. Ahasuerus 20:43, 23 Aug 2007 (CDT)
  • I got e-mailed twice this evening about clearing auto-blocks and so have removed all of the user blocks. The user block list was
    • 20:54, 19 Aug 2007, Marc Kupper blocked TslSor (expires indefinite) (contribs) (unblock) (Spams User talk:JVjr)
    • 12:33, 19 Aug 2007, Marc Kupper blocked DbyLys (expires indefinite) (contribs) (unblock) (Spams User talk:Dcarson)
    • 11:56, 19 Aug 2007, Marc Kupper blocked JhyT74 (expires indefinite) (contribs) (unblock) (Spams User talk:JVjr)
    • 10:44, 18 Aug 2007, Marc Kupper blocked ZnrQ1j (expires indefinite) (contribs) (unblock) (Spams User talk:Kraang)
    • 22:22, 15 Aug 2007, Marc Kupper blocked Rl8Xw2 (expires indefinite) (contribs) (unblock) (Spams User talk:Marc Kupper)
    • 21:54, 13 Aug 2007, Marc Kupper blocked MyiG1s (expires indefinite) (contribs) (unblock) (Spams ISFDB talk:Community Portal)
    • 21:52, 13 Aug 2007, Marc Kupper blocked BbcBuc (expires indefinite) (contribs) (unblock) (Spams User talk:Thomas conneely)
    • 21:52, 13 Aug 2007, Marc Kupper blocked CesYde (expires indefinite) (contribs) (unblock) (Spams User talk:Thomas conneely)
    • 21:51, 13 Aug 2007, Marc Kupper blocked Hl2E1i (expires indefinite) (contribs) (unblock) (Spams User talk:Thomas conneely)
    • 16:20, 21 Jul 2007, Grendelkhan blocked Bk8D3k (expires indefinite) (contribs) (unblock) (spam User talk:Hayford Peirce)
    • 16:19, 21 Jul 2007, Grendelkhan blocked RbpFdr (expires indefinite) (contribs) (unblock) (spam User talk:Hayford Peirce)
    • 17:58, 20 Jul 2007, Marc Kupper blocked SspDh7 (expires indefinite) (contribs) (unblock) (Spams User talk:Ahasuerus)
    • 17:52, 20 Jul 2007, Marc Kupper blocked IczZlr (expires indefinite) (contribs) (unblock) (Spams User talk:Unapersson)
Marc Kupper (talk) 22:27, 23 Aug 2007 (CDT)


Unverifying required, or new verification type(s)?

I mentioned before somewhere in here that if we disagreed on what physical verifying meant, it wouldn't be too difficult to add another verification category or two to cover the different meanings. One obvious discussion point was over whether "verification" meant you were claiming the book as something you would be willing to answer questions on for evermore - after all, nobody else can primary verify it if you have. (Yes, I know when Al "Godot" van Ruff returns he will make everything better, but I'm a bit more pressed for time than that.) Basically, I need to reclaim my sofa by next weekend so a guest can use it, which means the discard pile has to really, finally, go. And I'm thinking of putting some other less-read books into storage till I can rework the book-cases a bit. (Please don't try and tempt me into putting the discard pile into storage with them!) I'd rather NOT be bothered with questions over books I don't have any more, so should I:

1) Unverify the ones I discard?
2) Create a new Verification category for "Verified, but don't own now" and reclassify them that way? 
(Which could be useful for people verifying from library books, for instance.)

There's also the option of creating "3) Book is inaccessible at the moment, be prepared to wait for answers to questions about it" (which Ahasuerus and several others might like), but if people can wait for answers like that then we can also wait for Al to fix things, I guess. I've only recently discovered I can unverify pubs (when I started moving verifications from my other ID to this one) but as I did put a lot of effort into those pubs in the first place just plain unverifying them makes that work seem wasted. So I'm leaning towards option 2. Please, add some comments on this SOON! BLongley 17:29, 2 Sep 2007 (CDT)

Well, there are free WiFi networks in a number of countries as of 2007, so the number of "on the fly" verifications from bookstores, libraries, etc is likely to increase over time. And you are not alone in your predicament, Bill. After sporadically verifying a few hundred publications over the last 6 months, I ended up with a couple of stacks of duplicates (and one Lackey omnibus that had somehow managed to sneak into the house) which I'd like to get rid of at some point. It's not pressing, but I can see how I could end up with a thousand+ duplicates by the time I am done verifying my books.
Overall, I think that a simple "Verification copy no longer available" note will do for now given the urgency of your situation. Later on we can argue whether we need a new Verification category or whether the proposed note constitutes a tacit permission for other editors to overwrite it with a "Verified and I own it" Verification. Ahasuerus 18:31, 2 Sep 2007 (CDT)
Unfortunately notes are the slowest option, due to the approvals. It also leads to more to undo manually later if my name gets removed. I'm not going to be able to add anything more on this pass, otherwise I might be able to convince myself that such a boring set of edits are worth my time, but I've gone over each of them at least twice. I'd really prefer option 2, unless anyone can give a good reason why it's not something we'd want to do. It's easy to undo with a bit of SQL if it needs undoing (it would leave me no worse off than unverifying them all), and if people are still unsure of it we can leave it for my use only until it's decided whether to zap it or not. BLongley 12:40, 3 Sep 2007 (CDT)
The only reason why I was hesitant about creating a "Verified, but don't own now" category was that the order of "Bibliographic References" references is apparently meaningful, so there may be complications involved. However, as long as we don't create even more categories until Al comes back, an "SQL undo" solution should be viable, so I agree that it's the least painful alternative. Ahasuerus 14:43, 3 Sep 2007 (CDT)
OK, it's started. There is a "Primary (Transient)" category of verification now. The fact that the numbers for sources go "10, 13, 17" makes me suspect that SOME attempts at new reference sources haven't worked out in the past. But stick with the "don't change the reference ID" rule and even this experiment can be lost happily. BLongley 16:16, 3 Sep 2007 (CDT)
I've tried it out on a small series I own, and have done ALL the physical verifications for, but no longer want to keep, and the only downside so far is that there's no warning about any prior verifications of any kind now. Which is exactly as Verifications against Tuck or Clute/Nicholls appear, so no worries there. It may be a job for Al to add this new category to the warnings about editing verified pubs, if this level of verification is useful. But I've altered the obvious help to explain this is just an experiment - If I've missed some, please feel free to edit it and blame it all on me! BLongley 16:16, 3 Sep 2007 (CDT)
So far, so good: I have already created space enough for a guest to SIT on the sofa now! :-) My memory of having squeezed every bit of data out of them seems to have been false though, I'm going to have to add some coverart to some "Primary (Transient)" verifications still I think. When Amazon UK updates. (Why is it 2-5 minutes on Amazon US and 2-5 days on Amazon UK?) However, the remainder are the awkward ones where I will have to check whether I'm getting rid of an IDENTICAL pub or just another edition, so the last few will take much more time. I may have to just lose the rest and deal with individual queries about verifications later with a "Sorry, I think I don't own that anymore". BLongley 18:22, 4 Sep 2007 (CDT)
Anyway, it's been useful for me so far, does anyone else find it useful? It's quite weird watching my Verification scores go DOWN, and thinking that's a good thing - maybe someone will congratulate me on reaching the 2000 count going the wrong way? (They didn't on the way UP.) BLongley 18:22, 4 Sep 2007 (CDT)
The more you unverify the sooner i catch you!Kraang 19:31, 4 Sep 2007 (CDT)
Well, I'm only unverifying slightly faster than I verify... 3 more books arrived today for instance. And weekends have been quite productive recently, e.g. about 40 books last weekend. But the distance I've had to travel keeps increasing - I think it's now an hour's drive to any town or major village that I haven't already scoured the Charity shops in. I'm still very disappointed in the number of real second-hand book shops left. I've only visited two in the last year that had an SF section larger than my own home collection. Time to book a few days in Hay-on-Wye again, maybe... AFTER I've reorganised the shelving though. (I'm told the second bedroom should actually have a bed in it, but books don't need beds, do they?) BLongley 16:09, 5 Sep 2007 (CDT)

(unindent) I stopped by and noticed this thread; just to add my two cents, I'd say that I thought the verified flag meant that someone had actually looked at the physical edition and verified it. It didn't mean that they still had it, or had ever actually owned it; just that someone had checked the information against a real copy. Of course if that person wants to discard that book, they won't be able to answer queries about it, but that's not that big a deal -- it can be unverified if there's a serious concern, or a note can be added. So I'd say it's not necessary to unverify just because you don't have a copy any more. Mike Christie (talk) 07:50, 7 Sep 2007 (CDT)

I agree the intent, but it seems some people consider it almost "taking possession" of an ISFDB entry. Which is useful for questioners, but if that's the way it's going then I for one want to separate the ones I'm happy to discuss from ones I just want to say "I had access to a physical copy once, and checked it". Most of the recent entries on my talk page are questions about my verifications now, and most of the questioning I do is similar, which must be improving data quality but will just waste time eventually. Still, I expect Al will eventually sort out multiple verifications or whatever we really want. In the meantime, this gets me off the hook for a few books and allows Kraang to attempt to catch up on the "I've got all these books and want to talk about them" list! (Shame he can't see the recently changed ones from me, I'm sure there's some he could "take over".) BLongley 13:44, 7 Sep 2007 (CDT)
It's gotten to the point where I usually ignore the verification flag when I make modifications. Virtually all of my queries have gone unanswered even by editors who are still active. One editor actually unverified a pub rather than reply to my question. As long as there is only a single verify and no auto-notification of modification to verified pubs then the verified flag should be used by someone willing to answer questions.--swfritter 08:07, 8 Sep 2007 (CDT)
That's too bad someone would unverify and not even say why. I would not "ignore" the verification flag but would add notes if a verified pub does not match my book. Then I can post a message to the original verifier explaining why I've updated the record and move on. It also means I'm not putting the original verifier under any time pressure to locate the publication.
I would be interested in adding a "additional" verification flag where for pubs already verified I can say that it matches my copy. This would both increase confidence in the data and also give us two or more people to ask should a question come up. Of the stack of books I picked up this weekend about 1/2 of them were already verified which I thought was pretty cool as it means ISFDB is getting pretty good coverage. We could do a quick hack with the Bibliographic References table and just add rows for secondary physical but I was thinking along the lines of a publication log where there's a log associated with each publication that shows an edit history and would also allow for logging verifications.
A while back I had suggested a "My Library" thing where you could flag that you had a book. Maybe I'll revisit this but also add an "access" flag where you could flag how painful it would be for you to get the thing.
  1. I own the publication and it's on a shelf meaning I can find it more or less immediately.
  2. I own the publication but it'll take an hour, day, week, month, or year to get to it.
  3. I don't own the publication but I estimate an hour, day, week, month, or year to get to it.
  4. I don't own the publication and the odds are low that I'd be able to locate it again.
The main hassle is keeping ISFDB up to date should I shift books from one access state to another. Marc Kupper (talk) 04:15, 10 Sep 2007 (CDT)
The people who don't respond are generally those who did not do a thorough job in the first place. I don't use the ignore option unless the editor has a track record of A) Doing an unreliable or incomplete job and B) Not responding. That generally does not apply to anyone involved in this discussion. There are a number of user for whose response I would wait weeks, a few for whose response I would months and a couple for whose response I would wait years. swfritter
By the way, I fall into the "does not do a thorough job" or "does an incomplete job" categories on several title-types, but hopefully not the "unreliable job" category. I don't want to pressurise people into thinking they have to do EVERYTHING to our current standard if they can at least do something useful in the meantime. As mentioned elsewhere already today, if I feel like adding Reviews but not Interior-art, that saves someone else some work. If someone else adds the Interior-art but doesn't care about Reviews, that's useful too. If Primary Verification eventually means that everything is added to current standards, and that the Verifier will respond to requests for anything else decided later (e.g. if we eventually add some details about the actual Printer) that's fine, but it means we need the "lesser" verifications more. We have to keep the number of verification categories LOW though, or it will scare people off. Some categories may even become COMPLETE, in which case they're no longer useful to even show. (How many Tuck-owners do we need to work through the entire set before we say it's done? We should be REPLACING Tuck as a Source eventually.) BLongley 15:09, 10 Sep 2007 (CDT)
I have to disagree about replacing Tuck as a source. How many of us are in a position to own 40+ year old books in order to verify their existence? I began on the first page of Tuck a few months ago, and have slowly worked my way up to page 39 (Ben...) and have added literally hundreds of new publications based on his efforts. That leaves 400+ pages to check, meaning the figure for new pubs could run up into the thousands. (And I'm disregarding foreign language publications entirely.) Why give up such a valuable resource? As for verification, I think the new Primary (Transient) status works as well as it is intended to work. I also see the need to lock a pub, requiring more than one person to agree to any changes. Mhhutchins 16:26, 10 Sep 2007 (CDT)
The "eventually" is the important word here. I think Tuck was used to help create the original ISFDB, but if you're finding new works still then it obviously wasn't completely used. At least two other editors are Tuck-Verifying: but if you, Al and Ahasuerus (for instance) said "OK, I've finished ALL Tuck verifications" then I think we could say that that's reliable enough and NOBODY ELSE should bother attempting to use it anymore. I wouldn't want to recommend people repeating work already done multiple times already. If we're really that far off, fine, I can go buy them myself and help out: but I DO believe that EVENTUALLY we'll have all that data here, and our triple (or however many Tuck-Verifiers we decide are needed), in addition to our Primary Verifications, will make the ISFDB the definitive source over and above Tuck. Yes, I know we're years away, but share the dream, OK? BLongley 16:52, 10 Sep 2007 (CDT)
I don't want to encourage too many quick hacks - yes, all of us Mods could create our own "Bibliographic Source" for Verification and work around the Single Primary Verifier problem to some extent. This is why I've asked so many times about what Verification actually means, and dithered so long about this one (which at least is programmably removable/changable whereas Notes aren't). This SLOW Hack allows us to work around the "I Own It" versus "I Checked It from a Primary Source" differences that have become apparent. It's actually quite quick as you can undo the Primary and set the Primary (Transient) in one submission and no approval required. But I know it still only allows two "Checked it" verifiers at most so I doubt this will be a final solution (and really HOPE it doesn't become one). What I think we DO need is some agreement on what we really want. More comments on THIS Feature request maybe? Or a different Feature Request? BLongley 14:47, 10 Sep 2007 (CDT)
I would actually encourage people to do partial data entry. They might be discouraged from working on a pub if they notice that is Verified especially since we don't have an agreed definition for Verification. And eventually we will need a Locked flag that will require the editor to submit substantial justification for modifying a pub. Years away, of course. Just try submitting an update to IMDB. No matter how much justification you provide, the request seems to go into a bit bucket with only an automated response.--swfritter 15:53, 10 Sep 2007 (CDT)
I know I avoided verified pubs when I started, even when I disagreed with them. (But I always expected to come back to them later on pass two or three.) We don't use Publication discussions much either it seems. It's the balance between "I can't touch that" and "It's not verified, I can stomp all over it with my assumptions" that we need to find, I think. I know I sometimes get impatient with some titles that appear several times here where NONE are verified, except by ME, and do an update that will affect them ALL - but I'm manually checking to see whether the others are Tuck-Verified or Locus-Verified or such first. So another feature request might be that we're warned about all other verifications, not just primary. BLongley 16:31, 10 Sep 2007 (CDT)

Cartoon Submissions

The standard for cartoons has been developed only recently and a number of recent entries have been done incorrectly. This is the current standard:

Rules for including artwork. If artwork illustrates a particular story, it should be included. If it does not, but is a significant piece of artwork, or is signed by or credited to a well known sf artist, then it should be included. Credited cartoons are always included. Uncredited full-page cartoons in digest magazines or at least 1/3 page cartoons in pulp and bedsheet size magazines are always included. The title should be "Cartoon: " followed by the caption, in the original case, between quotation marks. If there is no caption the words "no caption" should be used without quotation marks. See the February 1957 issue of Dream World for examples (http://www.isfdb.org/cgi-bin/pl.cgi?DRMWRLDFEB1957) If an article is illustrated with diagrams, or with photographs, these do not need to be included; they are not "artwork" in the sense that we are indexing.--swfritter 12:35, 11 Sep 2007 (CDT)

Amazon Images

I've put together this bookmarklet in case any one else finds it useful. To use you need to visit the page and drag it to your browser toolbar. Will need testing in IE which is not available on Linux :-) Just click it when on the edit page to insert the Amazon image URL into the image field and open a window with the image. The only downside, if the open window is blank then you have to manually clear the image url field. --Unapersson 12:38, 14 Sep 2007 (CDT)

OK, I just got this working: I'm not sure why I couldn't before, maybe I was dragging to the wrong toolbar (it's the Bookmarks one rather than the Navigation one in Firefox 2 on Windows XP, if anyone else got stuck at that point.)
However, I can't see myself using it: I do not trust the stability of "01.LZZZZZZZ.jpg" image URLs and only use them as a last resort when I can't find a definitive one. Amazon seem to use it for "latest printing" now (I HAVE seen XX.LZZZZZZZ.jpg URLs eventually resolve to the one I want, but it's not worth the search anymore.)
Still, it's a good idea, and a single-click solution from the edit page to the relevant Amazon Pub page would be useful: knowing WHICH Amazon could be tricky still though. But please do continue to try! BLongley 15:27, 22 Sep 2007 (CDT)
Actually, it's not that hard, so here's my effort. It'll choose the US or UK site based on whether the Currency symbol is "$" or not (so works even when the "L" has been corrected to "£"). BLongley 09:40, 23 Sep 2007 (CDT)
Is there any recognisable system for user generated images? I've not really played around with Amazon enough to test. --Unapersson 08:56, 23 Sep 2007 (CDT)
I haven't spotted a system, but then I haven't really looked for a pattern. A lot of the images I want aren't attached to a properly ISBNed entry anyway... I should look into a quick title/author search next. BLongley 09:40, 23 Sep 2007 (CDT)
OK, that looks a bit more tricky, as I want to repeat the search with vaguer parameters until I actually get some results. And we need to strip some characters or even words to make sure of such (e.g. remove a "The" or a series title in brackets and suchlike before searching). Still, as this sort of Javascript coding doesn't require Al to do any work, are there any other quick-links people would like? Basically, anything someone uses that has a simple "The parameter you search by goes in the URL" should be quickly convertable, if the rest of the URL is fairly fixed. BLongley 15:13, 23 Sep 2007 (CDT)
I'm confused - how do you install and use this? I went to http://www.eclipse.co.uk/sweetdespise/isfdb.html and see a white page that says "Amazon Image" as a blue underlined link. How do I drag this to the bookmark toolbar? I gave up on dragging and did a right-click, new-Bookmark, and added ISFDB Amazon. Now what? Marc Kupper (talk) 19:05, 27 Jan 2008 (CST)
Well, it's literally as described. In Firefox: click on it and keep the mouse-button down, the mouse-pointer becomes a circle with a bar across it: move the mouse over the bookmark toolbar and it'll change to an arrow, release the mouse-button and it'll appear on the tool-bar. All that fooled me first time was that I wasn't actually DISPLAYING the Bookmarks toolbar, as I've never needed Firefox Help or Firefox Support often enough to constantly lose a few pixels to an otherwise useless toolbar. I keep it enabled now as I do use MY version still (and should probably add a few more utilities, e.g. linking to your ISBN linker in one click). What browser are you using? BLongley 16:15, 28 Jan 2008 (CST)
I'm using FireFox 2.0.0.11 on win-2000. When I right or left click the cursor stays as an arrow and there's no change to its shape as I drag the mouse. I'm using the FireFox default theme and the only add-ons are something called "DOM Inspector 1.8.1.11 and Talkback 2.0.0.11 both of which come with FireFox. Marc Kupper (talk) 22:19, 28 Jan 2008 (CST)
Stays as an Arrow? When you're over the link, the cursor should change to a hand. Then you left-click and hold and drag. If it's staying as an arrow all the time it sounds as though you're not over the link, so aren't actually on anything to drag. BLongley 12:59, 29 Jan 2008 (CST)

Wikipedia images - please be careful

Just a reminder that Wikipedia/Wikimedia do not allow direct linking to their images. They threaten to block any "deep-linking" IPs, which would prevent us from accessing Wikipedia biographies. To quote Bugs Bunny, please "be very very careful"! :) Ahasuerus 16:45, 23 Sep 2007 (CDT)

And here is a list of all of our Author records that point to a Wikipedia image as of August 11, sorted by first name:

I've done a second pass and I think I've removed all the links.Kraang 19:24, 23 Sep 2007 (CDT)
Excellent, thanks a lot! :-) Ahasuerus 20:20, 23 Sep 2007 (CDT)
Alexander Beliaev
Alexander Bogdanov
Alice Sebold
Andreas Eschbach
Anne Frank
Anthony Horowitz
Arthur Koestler
Ayn Rand
Bentley Little
Bram Stoker
Brian Keene
Charles Stross
Cormac McCarthy
Cory Doctorow
E. Nesbit
Edgar Pangborn
Evangeline Walton
Gail Carson Levine
Guy de Maupassant
H. G. Wells
Hans Christian Andersen
Hayford Peirce
Ian McDonald
Italo Calvino
J. K. Rowling
Jack Coggins
L. Ron Hubbard
Larry Niven
Lew Wallace
Mark Z. Danielewski
Megan Lindholm
Michael Jan Friedman
Michelangelo Merisi da Caravaggio
Neal Stephenson
Nichelle Nichols
Nigel Kneale
Olof Johannesson
Philip K. Dick
Ray Bradbury
Raymond E. Feist
Robert A. Heinlein
Sean McMullen
Sergei Lukyanenko
Shelley Jackson
Sinclair Lewis
Stephen King
Tad Williams
Tanith Lee
Tanya Huff
Terry Goodkind
Terry Pratchett
Vladimir Nabokov
William Styron
Zacherley

Ahasuerus 18:16, 23 Sep 2007 (CDT)

Oops - can this scan be re-run? I'm pretty sure I recently deep-linked an image or maybe I thought about it. I'm getting a little pissed at Wikipedia for their ever changing image policies and the robots that dump incomprehinsible things on the image pages and so it's likely I did not do the deep link but suspect it's a good idea to re-run the scan. Marc Kupper (talk) 19:13, 27 Jan 2008 (CST)
I don't think we have any links to Wikipedia listed on ISFDB:Data Consistency/Disallowed URLs, but it only covers Publication records, not Author records. I will re-run my scan in a couple of hours and see what it finds... Ahasuerus 20:08, 27 Jan 2008 (CST)
To be fair, Wikipedia has had a policy against deep-linking to images since at least 2004, and this has been consistent. They have been changing the degree to which they tolerate "fair use" images (mostly by reducing it) which annoys some editors, including myself, but the policies on image re-use and linking have not, as far as i know, changes significantly. -DES Talk 10:44, 29 Jan 2008 (CST)
Scan re-run, one offending URL found - ttp://upload.wikimedia.org/wikipedia/commons/thumb/8/8c/James_Thurber_NYWTS.jpg/462px-James_Thurber_NYWTS.jpg Ahasuerus 00:11, 30 Jan 2008 (CST)

Watch for incorrect legal name entries

I have been unable to get Alibrarian's attention. He is continuing to update legal names incorrectly. They should be last name first with a comma then the first name. If you notice any such entries pleas put them on hold.--swfritter 13:53, 10 Oct 2007 (CDT)

Which reminds me that I have been planning to write more data consistency scripts for some time. I'll try to spend the next few days on them and only jump in if the submission queue starts getting ugly. I'll be monitoring the Wiki, so yell if you need me :) Ahasuerus 22:55, 10 Oct 2007 (CDT)

Self-moderating vs. other-moderating

As a newish moderator (I went away from isfdb for a while right after getting moderatorized) I've mostly been using my moderation powers to clear the queue of no-brainers, while leaving my own edits for other mods to approve --- I like having a second pair of eyes on my submissions, even the ones that I think are no-brainers, because I know I'm prone to making the occasional silly mistake ( despite the vast wisdom implied by that moderator flag ;) ). I'll approve my own subs if I need to make a multi-step edit but otherwise I'll leave them to be doublechecked.

Swfritter's recent comment on my Talk page, and observing the mod queue, makes me wonder if other mods are taking exactly the opposite tack, and mostly approving their own submissions (and non-moderator submissions) while leaving mods' submissions to be approved by their own submitters. Is there a custom or convention here? Should I approve my own subs and/or avoid approving other mods' subs? --WimLewis 16:03, 15 Oct 2007 (CDT)

I'm not sure if anything's written in stone, but here's what I do. I approve all of my own submissions, and in between will review the submissions of non-mod editors. If these submissions are over my head or out of my comfort zone, I will leave them in the queue. I never touch a submission by another mod...unless I have a senior moment and forget who has become a mod. Which may explain why I approved some of your edits a day or so ago. I didn't remember that you were made a mod. In fact, until I read this notice, I still had not recalled! Mhhutchins 16:18, 15 Oct 2007 (CDT)
I'm glad you mentioned it, as we've (well, me at least) been wondering if we're leaving new moderators in the lurch. From what I've viewed so far, most new moderators approve their own uncontroversial edits and ask for help on the controversial, BEFORE they submit such. (I think there's still a lot of HELP! messages from me on various areas of the Wiki still, not all answered, and some even archived now.) I'm fine with a "two sets of eyes are better than one" if you want to be the second pair of eyes on other people's edits, but still want your own checked. For potentially massively-destructive merges, unexplained make-pseudonyms, etc, it's probably best to explain WHY you think such an edit is OK somewhere. There seems to be an unwritten "Don't approve other Mods edits without being asked" convention, but if you want us to approve things you leave for us, then a stated carte-blanche permission will do for me. I'm still wondering if Dissembler is going to Sulk after me approving his stuff over the weekend, when Al normally does it... ;-) BLongley 16:33, 15 Oct 2007 (CDT)
In this case only we may want to set a "length of time in queue" standard, perhaps and put WimLewis name in the noticeboard list with the information that he is a self-moderator who leaves items in the queue for perusal by others? Thank goodness for editors who care about the data.--swfritter 16:54, 15 Oct 2007 (CDT)
"Clearing the queue of no-brainers" is a valuable task - I think that's EXACTLY the reason I got black-jacked into being a mod, for instance, I was just too active an editor. ChrisJ finally submitted to moderatorship for similar reasons I think. I'm happy for self-moderators to continue in that as long as they wish, but yes, notification that that's what they're doing should be made clearer. (I know they have the powers to notify us themselves, but we're not necessarily making it clear that that's something they should do - the Moderator help still seems to be "All or Nothing" and makes no allowances for cautious Self-approvers.)
Back to the communication issue - just because you've become a moderator doesn't mean you can't ask for help still, and I for one am happy to help, or admit that I have NO idea whatsoever about that particular problem either. There's more Mods to poke questions at now - hopefully Mods can point to other Mods with the right specialities, or at least point to the right places to ask, if they don't know themselves. Even a new Mod can still point at the people that nominated or supported him/her, and say "It's their fault! Ask them! I'm still learning!" :-) BLongley 18:03, 15 Oct 2007 (CDT)

Submission gone awry

Something happened (a parse error?) to one of my submissions (twice) and it's showing up on a strange location on the Moderators' New Submissions page. The submission took on the third attempt, but how do we clean up the page? Mhhutchins 17:50, 3 Nov 2007 (CDT)

Bug Al, I guess. :-/ BLongley 18:11, 3 Nov 2007 (CDT)
There's still the useful dumpxml script (why isn't that documented more widely?) that can give clues as to the original problem though. Spot the difference between these three submissions: 1 2 3
From what I can see, there's an unwanted "< ul >" on the first two submissions for page 7. BLongley 18:11, 3 Nov 2007 (CDT)
So that's what happened. I'd created an HTML list in the notes of the submission, and a cut-and-paste slip-up placed it into the wrong field. Mhhutchins 18:58, 3 Nov 2007 (CDT)
Yes, it's useful being able to add HTML to overcome some limitations here, but there are dangers still. :-/ Here's the entry I remembered about previous problems like this: Al did make it slightly more idiot-proof, but I proved we can always find more creative idiots whatever he does! ;-) BLongley 20:16, 3 Nov 2007 (CDT)
The submisions page is still vry strange, several entries by unknown for unknown that give python errors when you look at them and the body of a submission showing up before the submission list. Dana Carson 13:06, 4 Nov 2007 (CST)
Yes, it'll be like that till Al can zap the two badly-formatted submissions. (Unless anyone knows different and there's an undocumented ZAPSUB script or something?) BLongley 13:18, 4 Nov 2007 (CST)
There is an undocumented ZAPSUB; it's not linked anywhere in the navbar: hardreject.cgi. If you give it a submission number, ala: http://www.isfdb.org/cgi-bin/mod/hardreject.cgi?123456, it will force the submission into the rejected state from a URL. (Used when I can't get to the SQL interface) Alvonruff 14:21, 5 Jan 2008 (CST)
Thanks, I'll add it to the Moderator Help page! Ahasuerus 14:46, 5 Jan 2008 (CST)

Eye Test/Frankenstein entries

I've left a dozen submissions on the queue that I've constructed from 1) Our complete lack of publications for a given title 2) Amazon images 3) Alibris and Abebooks for dates, ISBNs and prices 4) Wikipedia and other reference data and 5) all sorts of websites your mother wouldn't want you viewing. Well, maybe not the last one. Generally they look fairly useful but I'm sure that someone with better eyes than me might squeeze a LITTLE more data out of the pictures. Or have a reference book that shows that I've mixed some completely different editions. So, for once go have a moan at MY submissions! (Or praise me for such diligent detective work, I don't mind.) I'm leaving them for now as I should at least review the pictures in daylight. (Which I will not see at home till next weekend now.) BLongley 16:42, 18 Nov 2007 (CST)

I've created a number of these. If I find several sources that agree than I go with the info. Until someone can update or confirm the data I add into notes the word "Placeholder".

submission someone may reject

I slipped and incorrectly went to put the wrong record into a series. If someone rejects it I'd be pleased (but if it gets accepted I'll fix it up.) The title is "Traditional Values", & it's 806049. (Hmm. Odd. Looking at my pending submissions, I see three for that title. There may be a good reason for that, but off hand I don't remember doing three things to it. Did I do something else too?) -- davecat 12:51, 28 Nov 2007 (CST)

There were three identical submissions for "Traditional Values" in the queue. A browser malfunction, perhaps? In any case, I have rejected all three for now. Ahasuerus 13:09, 28 Nov 2007 (CST)
Thank you! (At a guess, then, a browser problem - maybe refreshed at the wrong point?)

Problems in approvals?

About a half dozen of my approvals have produce an error messages and when I back space and re-approve them they've gone through. Has this happened to anyone else? Kraang 21:45, 22 Dec 2007 (CST)

This has happened to me when I've occasionally double-clicked the approval button. The first click approved it, the second click brings up the error message that "Submission xxxxx is not in NEW state." This is probably what's happened to you as well. Mhhutchins 15:40, 5 Jan 2008 (CST)
I've seen ones where my approval caused a horrible Python error first time, but back-spacing and approving it again let it through. It's not a "Submission xxxxx is not in NEW state" error, it's one of those really alarming purple messages with lots of dump data. I haven't been able to reproduce it on demand, but it does seem to occur mostly when I update a pub and then the associated title (for publications where you can't do both in one edit) but then approve the title edit before the pub edit (as approving the pub edit leaves me with the chance to look at the result with one more click and the title edit DOESN'T).
I know approving things out of order can and probably WILL occasionally cause problems, but I can't see why some of these do as they're for totally different database fields. (And submitting updates then making more edits before the first are approved is known to cause problems as well, e.g. merging titles then updating the one that isn't going to remain.) I'm not too worried unless we're leaving unapprovable entries or stray titles or mismatched titles/pubs (SOMEONE is creating a lot more of those than I like, and I just hope it isn't ME, posting drunk) - but it is a temporary worry when your screen goes multi-coloured and long with error messages. Anyone else seen this? BLongley 16:47, 20 Jan 2008 (CST)
Hm, the only Python errors that I recall seeing lately were generated when I was abusing the "Go Back One Page" functionality within the browser. Depending on the timing, it would occasionally resubmit the form with no-longer-valid parameters and cause an ugly looking Python error. Ahasuerus 17:17, 20 Jan 2008 (CST)

ANTHOLOGY not COLLECTION

My NewPub | The British Fantasy Society: A Celebration should be ANTHOLOGY, not COLLECTION; after submitting I realized I clicked the wrong link to create it. I I'll fix it later. --Roglo 12:00, 5 Jan 2008 (CST)

The Naming of Names

I see my submission MakeVariant | The Naming of Names is still pending. I'm trying to remove it's relation to Dark They Were, and Golden-Eyed (to which it is currently a variant title) - they are two different texts, with The Naming of Names being 1 page poetic description of changes in The Martian Chronicles, and Dark They Were, and Golden-Eyed a completely different story (not published in Chronicles) with dialogs and action, even though there are Martians in it. I will later make 2004-05: The Naming of Names a variant of The Naming of Names (in Bradbury Stories the date from The Martian Chronicles is part of the story title). Thanks. --Roglo 02:42, 10 Jan 2008 (CST)

Spindrift (title update)

I was trying to add series info to the Spindrift title, and after submission noticed that somehow I pasted too much into the Series input box; the submission looks like:

<?xml version="1.0" encoding="iso-8859-1" ?>
<IsfdbSubmission>
<TitleUpdate>
<Record>211411</Record>
<Submitter>Roglo</Submitter>
<Subject>Spindrift</Subject>
<Series>Coyote Rising

Coyote Universe</Series>
</TitleUpdate>
</IsfdbSubmission>

and Series should be Coyote Universe only. So think you should reject this, and I'll resubmit correctly later. Thanks. --Roglo 11:51, 19 Jan 2008 (CST)

Done. BLongley 12:43, 19 Jan 2008 (CST)

Request to edit Help:How to verify data

This page is locked. I have placed a proposed additional section on Help talk:How to verify data. Please review it when you have a chance. -DES Talk 07:24, 25 Jan 2008 (CST)

Looks mostly reasonable. It needs a little rewriting, e.g. "I rarely verify..." is a personal comment from one moderator, and help should be a bit more detached and reflect the consensus. The original discussion doesn't look quite finished yet either, so no hurry. BLongley 13:45, 25 Jan 2008 (CST)
I have rewritten the "I rarely" comment. I do think that we should make the changes to the helps screen fairly soon, if the points addressed have even a reasonable consensus. Because the new editor is otherwise left to try to find existing discussion or ask for new answers when this question arises, as it is likely to. We can always change the help screen text again if the consensus changes. -DES Talk 14:15, 25 Jan 2008 (CST)
Looks fine to me.--Rkihara 14:54, 25 Jan 2008 (CST)
OK, it's improved: I've corrected a few typos (in the section based on my own comments it seems! :-( ) and made it easier for people to look at the (LONG!) page you're referring to. We don't seem to have any hard and fast rule about how long changes should be discussed for - the only time-limit I can think of is for Becoming a Moderator, and that goes back to when most active editors where active "most days" rather than "most weeks". 5 days seemed a bit rushed to me when some editors only appeared once a week, I'd have preferred 7. But if people aren't disagreeing then all it takes is a brave moderator to take the initiative and sort it out sooner rather than later. We're getting rather good at dithering indefinitely though. :-(
Still, if people are discussing Verifications again, can I take it that nobody wants my "Primary (Transient)" verification category removed, and that it's no longer "experimental"? That was one of my braver changes and seems to have worked to some extent. BLongley 15:39, 25 Jan 2008 (CST)
I am using the "Primary (Transient)" for verifying pubs for Bibliographic Tasks because someone else has verified the pubs previously. It's a very handy tool. At some point in time we may also want to think about setting a policy for resetting the verification flag if it is has been set by a user who has not been active. This would allow someone who is still active and available for queries to reset the flag in their name. Not critical right now but might be more important in the future.--swfritter 17:42, 25 Jan 2008 (CST)
  • There have been no changes to the proposed text, and no comments here, for 4 days. What needs to be done for this wiki-edit to be considered? By the way, I don't know how many vandal edits you get here (I'm sure far fewer than Wikipedia). Could semi-protection be considered instead of full-protection for some of those pages? Semi-protection allows only editors who have been registered for a certain number of days, and who have made more than a certain minimum number of wiki-edits, to edit the page. (The numbers are specific to the wiki installation, and I don't know how you have them set. I think the defaults are 3 days and 25 edits, but I'm not sure of that.) -DES Talk 10:51, 29 Jan 2008 (CST)
Oh, I'm sure it's being considered: just not discussed very much. We're a rather smaller active base of users and some people are weekdays only, some weekends only, some only occasional. 4 days for a consensus would be remarkable. BLongley 14:57, 29 Jan 2008 (CST)
I guess, based on Wikipedia practice, that I didn't expect a need for a real consensus, merely a sanity check. Frankly I expected the first mod who saw the request to make the edit. My reasoning was: "The page is protected not because it is a basic policy page that requires consensus to change, but to protect against spam. This edit is clearly not spam. Therefore it can be made right off and others can change it further (if they are mods) or ask mods to do so on their behalf, just as with any wiki page. But if people here feel that such edits should not be made without achieving consensus, so be it. Would a mod (who could technically make such an edit at any time) be expected to propose it on the talk page and wait for consensus? Or would a mod simply make such an edit directly? -DES Talk 15:24, 29 Jan 2008 (CST)
As lots of activity here is usually considered a sign of INsanity, asking for a sanity check isn't common! ;-) If it was an obvious mistake being corrected, a Mod would probably have done it straight away. If considered a highly controversial one, then they'd probably ask for advice. In between: well, I normally leave it a week and see if anyone disagrees. A minor change to Help can mislead ALL new editors and not be noticed by those of us that don't actually READ help any more as we THINK we know it all already. Most Mods do read the discussion pages occasionally though, so it's always worth mentioning a proposed change, or that a non-trivial change has been made. We're just getting to the stage where some Mods are going from reading ALL changes to the Wiki to just SOME changes - yes, it has been that quiet in the past. BLongley 17:40, 29 Jan 2008 (CST)
As to vandals - well, it's vandal/spam BOTS we protected against, and at times those were coming at a rate that took up all my time if I was the only Mod online to deal with them. Not fun. It looks a bit draconian, but the software was rather limited at the time, an old version that we might in fact still be on? MediaWiki 1.4.5, it seems. BLongley 14:57, 29 Jan 2008 (CST)
We might well want to get upgraded to 1.6 or higher when we can -- there are significant bug fixes and additional features in the post 1.5 world (I think the current version is 1.12). But I think that semi-protection would serve to foil such BOTS, and I think it was available in version 1.4, although I'm not sure of that. -DES Talk 15:24, 29 Jan 2008 (CST)
The bots haven't bothered us in many months, so I went ahead and unprotected that page for now.
We tried upgrading the Wiki last year, but it didn't work because of various interdependencies on the back end, which is maintained by our gracious hosts at TAMU. They were going to do a major reshuffling of "our" servers earlier in January, which would have improved performance and allowed us to upgrade the Wiki software to a more recent version, but then they ran into some administrative issues and the whole thing is up in the air at the moment. Hopefully, they will be able to implement their changes in the foreseeable future so that we could go forward with our upgrades. Ahasuerus 15:42, 29 Jan 2008 (CST)
Oh, I see. Well, when we live on soemone else's machines, we must take what is available, and be glad that it is. -DES Talk 15:51, 29 Jan 2008 (CST)
I have made the edit myself, since the page is now unlocked. -DES Talk 15:59, 29 Jan 2008 (CST)
What you added looks fine - thank you! Marc Kupper (talk) 00:20, 30 Jan 2008 (CST)

Personal tools