User talk:Citation bot/Archive 32
![]() | This is an archive of past discussions. Do not edit the contents of this page. If you wish to start a new discussion or revive an old one, please do so on the current talk page. |
Archive 25 | ← | Archive 30 | Archive 31 | Archive 32 | Archive 33 | Archive 34 | Archive 35 |
Bots crashes halfway on Murine respirovirus
- Status
- Not a bug, just a crazy huge page.
- Reported by
- Headbomb {t · c · p · b} 03:52, 23 May 2022 (UTC)
- Replication instructions
- Run gadget on Murine respirovirus
- We can't proceed until
- Feedback from maintainers
Punctuated bare URLs again
Many thanks to @AManWithNoPlan for the prompt fix of this issue, reported at User talk:Citation bot/Archive_31#Bot_fails_to_fill_ref_to_bare_URL_followed_by_punctuation. The bot is now chomping its way through a list of 1,355 articles with punctuated bare URLs.
However, I just spotted a glitch. On Belt-driven bicycle, there was a bracketed bare ref followed by a full stop: <ref>[https://www.google.com/patents/US425390].</ref>
The bot edited the page[1], but didn't fill this ref.
So I stripped the brackets etc[2], leaving a "pure" bare ref: <ref>https://www.google.com/patents/US425390</ref>
Then I invoked the bot again, and in this edit[3] it filled the ref.
I set up some testcases at User:BrownHairedGirl/sandbox198, where the bot did not do well: it failed to fill any of 5 bracketed refs[4]. BrownHairedGirl (talk) • (contribs) 17:45, 23 May 2022 (UTC)
PS As before, feel free to experiment with that sandbox. --BrownHairedGirl (talk) • (contribs) 17:54, 23 May 2022 (UTC)
{{fixed}} AManWithNoPlan (talk) 00:00, 24 May 2022 (UTC)
- Bless you, @AManWithNoPlan. That's great.
- When the bot has finished its run through the batch of 1,355, I will pre-parse the list again and run the bot again on the residue. BrownHairedGirl (talk) • (contribs) 00:05, 24 May 2022 (UTC)
- And it's working. The bot is chomping its way through Articles with bare link followed by punctuation, take 2, filling the bracketed+punctuation refs: e.g. [5], [6], [7].
- I wish that editors were a bit less creative in finding ways to mangle citations ... but thanks to your diligent programming, we are working around some of the creativity.
- Note that as a separate workflow, I have been using the bot to fill what I call "semi-bare" refs: bare URL followed by access-date, e.g.
http://example.com/Boris-Is-A-Liar, access-date 14th. Mar., 2020.
I do that by partly-filing a {{cite web}}, without title, then letting CB do the rest. the documentation is at User:BrownHairedGirl/Articles with semi-bare links and {{SemiBareRefNeedsTitle}}. - I considered asking you to include this directly in CB, but decided that a) I had bombarded you with more than enough requests recently, and b) some of it is a bit edgy, and might draw unwanted heat onto the bot.
- For example: the date formats are all over the place. AWB does an amazingly thorough job of rescuing malformed dates (like the one in my example above), but there are some so offbeat that it can't accurately decipher them, and those end up in the date field still needing attention. I reckon that's hepful 'cos that way the errors are tracked and hence more likely to be fixed, but I got a monstering before for moving imperfect data into tracked citation parameters, and I wouldn't want to make you and CB the target of that editor's pedantic wrath, esp since AFAIK CB lacks the cunning date-fixes built into AWB's powerful WP:GENFIXES#Cite_Template_Dates_(CiteTemplateDates), which does a lot more than the documentation claims.
- However, if you were interested in perhaps extending the bot to of some of the more straightforward semi-bare refs, I could set up some testcases and show you regexes which work. Only if you are interested.
- Meanwhile, my task for tomorrow is seeing whether I can tackle semi-bare refs with a {{Webarchive}} template included. It should be possible to do something with them, but I need a clear head for that. BrownHairedGirl (talk) • (contribs) 04:03, 24 May 2022 (UTC)
Invisible characters
- Status
- {{fixed}}
- Reported by
- Aidan9382 (talk) 06:07, 24 May 2022 (UTC)
- What happens
- In rare cases, when creating titles for citations, the bot proceeds to insert some invisible characters, either due to encoding issues or because the original source had them.
- What should happen
- The best move would be to replace non-breaking spaces with regular spaces (as these are nearly the same functionally) and to remove any other invisible characters (as these are actually invisible and would never display). For the encoding issues, thats not for me to really touch on, as I'd need to get familiar with the bot myself to determine whats wrong there.
- Relevant diffs/links
- ZWS case, Encoding case. There have been more cases in the past. If I find them, ill append them. I believe this has happened with non-breaking spaces, though i may be mis-remembering.
- We can't proceed until
- Feedback from maintainers
ZWS Issue fix confirmed, encoding still not. I assume thats not gonna be possible easily within reason (info). Aidan9382 (talk) 17:46, 24 May 2022 (UTC)
- That is impossible. https://en.wiki.x.io/api/rest_v1/data/citation/mediawiki/https%3A%2F%2Fwww.zerozero.pt%2Fcoach_compet_detail.php%3Fid%3D5455%26competicao_id%3D42 AManWithNoPlan (talk) 18:16, 24 May 2022 (UTC)
- Oh wow, I see why. Ok, thats understandable. Thanks for fixing the ZWS issue at least. Aidan9382 (talk) 18:17, 24 May 2022 (UTC)
TNT volume/issue/page(s)=Online
- What should happen
- [8]
- We can't proceed until
- Feedback from maintainers
- Not fully fixed. See [9] and [10]. Headbomb {t · c · p · b} 17:35, 24 May 2022 (UTC)
- Caps added. AManWithNoPlan (talk) 17:43, 24 May 2022 (UTC)
Fails to TNT volume/issue = N/A
- Status
- Fixed by adding a bunch more white-space types to the ignore function. Also, changed title compare to be case-insensitive.
- Reported by
- Headbomb {t · c · p · b} 15:00, 26 May 2022 (UTC)
- What should happen
- [11]
- We can't proceed until
- Feedback from maintainers
Running on
{{Cite journal|last1=Dasgupta|first1=Utteeyo|last2=Jha|first2=Chandan Kumar|last3=Sarangi|first3=Sudipta|title=Persistent Patterns of Behavior: Two Infectious Disease Outbreaks 350 Years Apart|journal=[[Economic Inquiry]]|volume=n/a|issue=n/a|doi=10.1111/ecin.12961|issn=1465-7295|doi-access=free}}
does nothing. But removing the title
{{Cite journal|last1=Dasgupta|first1=Utteeyo|last2=Jha|first2=Chandan Kumar|last3=Sarangi|first3=Sudipta|title= |journal=[[Economic Inquiry]]|volume=n/a|issue=n/a|doi=10.1111/ecin.12961|issn=1465-7295|doi-access=free}}
and then running has the bot properly TNT the volume/issue information. Headbomb {t · c · p · b} 15:00, 26 May 2022 (UTC)
Normalize hyphens, and TNT volume/issue=N/A accordingly
The difference in hyphen in the title leads to a failure of TNTing. The before hyphen is - (-), the after hyphen is ‐ (‐). Both should be considered equivalent/normalized to the plain - (-). Headbomb {t · c · p · b} 15:09, 26 May 2022 (UTC)
- Status
- Implemented some thing that have moved this closer to Fixed
- Reported by
- Headbomb {t · c · p · b} 19:39, 26 May 2022 (UTC)
- What happens
- Nothing
- What should happen
- Should finish
- We can't proceed until
- Feedback from maintainers
This happens on several longer articles. A systematic fix would be nice. Perhaps the bot could have a 'there's X references to be processed, I estimate processing time to be X × Y' and then only die after that threshold has be exceeded by a certain factor (like +50%/double/triple the expected processing time). Headbomb {t · c · p · b} 19:39, 26 May 2022 (UTC)
- The PHP server does not actually support that (printing out stuff before the entire run is complete), but I will look into seeing if there is an internal timeout that is killing the run. AManWithNoPlan (talk) 21:02, 26 May 2022 (UTC)
Bot not recognising lists on my batch job pages
For the last 10 months, I have been running batch jobs using the "Linked pages" option at the bottom of the form https://citations.toolforge.org
I have several such pages for different workflows, each with a similar format: a section or two of explanatory notes, then a "lists" section where the pages are linked in a numbered list, e.g. # [[:2009 Boston City Council election]]
This system worked very well for me, because:
- The bot's edit summaries included the page name, allowing anyone to monitor what I am doing
- That same feature allowed me to track different workflows
- Each batch page had a linked list, allowing me to click on the link to a page which the bot skipped, and inspect what happened
See e.g. User:BrownHairedGirl/Articles with new bare URL refs (edit | talk | history | links | watch | logs), with over 150 versions since October 2021. Here is an example of a documented batch: Special:permalink/1089218065
Then sometime on the 24th this system began to degrade. First the webform stopped recognising the lists unless I stripped out the documentation. Then in the last few hours it stopped recognising the lists even when all the docs were removed, and the only remaining content was the linked list. See e.g. the latest version of list of new articles, which the webform said contained no list. I had to reformat the list a pipe-separate lit, and submit it through the first line on the form. This is extra work for me, makes it harder to track jobs, and removes transparency.
What is going on? BrownHairedGirl (talk) • (contribs) 06:39, 26 May 2022 (UTC)
- Can you check again. Seems to be working for me. AManWithNoPlan (talk) 15:35, 26 May 2022 (UTC)
- Any luck BrownHairedGirl? AManWithNoPlan (talk) 13:46, 27 May 2022 (UTC)
- Sorry for the slow reply, @AManWithNoPlan. I was trying to find the notes I had made when I last used the listed-pages line on the webform, about 12 hours ago.
- User:BrownHairedGirl/Articles with new bare URL refs - initially failed on MS Edge, but worked after a hard refresh
- User:BrownHairedGirl/Articles with probably fixable bare links - rejected by both MS Edge and Opera, even after a hard refresh; initially failed on Google Chrome, but worked after a hard refresh
- I have left both pages untouched since then. BrownHairedGirl (talk) • (contribs) 15:04, 27 May 2022 (UTC)
- The bot died today and I have done a reboot about little bit ago. AManWithNoPlan (talk) 15:16, 27 May 2022 (UTC)
- @AManWithNoPlan: I just tried restarting User:BrownHairedGirl/Articles with new bare URL refs, using Google Chrome. Despite multiple hard refreshes of the webform, I repeatedly got the response "!No links to expand found" ... but there are 1,039 links. BrownHairedGirl (talk) • (contribs) 15:27, 27 May 2022 (UTC)
- @BrownHairedGirl works for me: https://en.wiki.x.io/wiki/Special:Contributions/Citation_bot AManWithNoPlan (talk) 16:09, 27 May 2022 (UTC)
- @AManWithNoPlan: Which browser did you use?
- I get difft results from difft browsers. BrownHairedGirl (talk) • (contribs) 16:10, 27 May 2022 (UTC)
- Chrome Version 101.0.4951.64 (Official Build) (x86_64) AManWithNoPlan (talk) 16:14, 27 May 2022 (UTC)
- Maybe I need to update Chrome. It'll see if that helps. BrownHairedGirl (talk) • (contribs) 16:32, 27 May 2022 (UTC)
- Chrome Version 101.0.4951.64 (Official Build) (x86_64) AManWithNoPlan (talk) 16:14, 27 May 2022 (UTC)
- @BrownHairedGirl works for me: https://en.wiki.x.io/wiki/Special:Contributions/Citation_bot AManWithNoPlan (talk) 16:09, 27 May 2022 (UTC)
- @AManWithNoPlan: I just tried restarting User:BrownHairedGirl/Articles with new bare URL refs, using Google Chrome. Despite multiple hard refreshes of the webform, I repeatedly got the response "!No links to expand found" ... but there are 1,039 links. BrownHairedGirl (talk) • (contribs) 15:27, 27 May 2022 (UTC)
- The bot died today and I have done a reboot about little bit ago. AManWithNoPlan (talk) 15:16, 27 May 2022 (UTC)
- Sorry for the slow reply, @AManWithNoPlan. I was trying to find the notes I had made when I last used the listed-pages line on the webform, about 12 hours ago.
- Any luck BrownHairedGirl? AManWithNoPlan (talk) 13:46, 27 May 2022 (UTC)
Yachting World is a magazine
- Status
- Fixed
- Reported by
- BrownHairedGirl (talk) • (contribs) 02:37, 29 May 2022 (UTC)
- What happens
- cite to https://www.yachtingworld.com/americas-cup/british-team-confirmed-as-challenger-of-record-for-37th-americas-cup-130697 converted from {{Cite web}} to {{Cite news}}.
- What should happen
- Yachting World is a magazine, so it should use {{Cite magazine}}
- Relevant diffs/links
- https://en.wiki.x.io/w/index.php?title=2024_America%27s_Cup&diff=prev&oldid=1090358386
archiveurl as the result of an apparent typo
- Status
- Fixed
- Reported by
- Trappist the monk (talk) 14:05, 28 May 2022 (UTC)
- What happens
- changed this:
|archive=url=https://web.archive.org/web/20131017213918/http://www.in.com/jim-allister/biography-138707.html
- to this:
|archiveurl=url=https://web.archive.org/web/20131017213918/http://www.in.com/jim-allister/biography-138707.html
- What should happen
- Because
=
and-
are adjacent to each other on a standard qwerty keyboard, the=
was likely a typo so the 'fix' should have been:|archive-url=https://web.archive.org/web/20131017213918/http://www.in.com/jim-allister/biography-138707.html
- The cs1|2 template is still broken because no
|url=
which the bot can get from the value in|archive-url=
(http://www.in.com/jim-allister/biography-138707.html
) - Relevant diffs/links
- diff
Thanks! Good work
This edit: https://en.wiki.x.io/w/index.php?title=Reasonable_accommodation&type=revision&diff=1084655807&oldid=1079082280 is [IMHO] a good example of how the use of this bot can result in a successful edit.
Skimming around in this "Talk:" page, I found several instances of ["alleged"] problems or "issues", in some cases with an apparent "lack of consensus" -- so far! -- about what should be done. I have no objection to a situation in which other readers, editors and/or coders -- [persons who probably know WAY more than I do about the design and operation of this bot, and the related "issues", if any] -- come here to hash out some decisions that might affect some possible future changes to this bot.
I just wanted to chime in with a *** Thank You ***, because I found this edit (see the above link to a "DIFF" listing) to be inspiring.
I probably should also send a "ping" -- such as
{{ping|User:Whoop whoop pull up}}
to User:Whoop whoop pull up. (right?) If so, then ... here goes: @Whoop whoop pull up:
Thanks! --Mike Schwartz (talk) 17:11, 23 May 2022 (UTC)
State of the bot
The command line does work, if you have Oauth Tokens. Extensive code coverage improvements have been implemented and a huge number of new test cases have been added. This has found a couple of small bugs and should prevent new bugs. The code base has been shrunk significantly by merging duplicate code functionality into functions - this should significantly speed-up bug fixes and code development and reduce new bugs. AManWithNoPlan (talk) 01:29, 17 May 2022 (UTC)
- That's great, @AManWithNoPlan. Huge thanks to you for all your hard work keeping the bot running, and continually improving it. BrownHairedGirl (talk) • (contribs) 03:33, 17 May 2022 (UTC)
- The slowest step - by far - in citations is determining that DOI's are broken. That was a surprise. AManWithNoPlan (talk) 15:57, 17 May 2022 (UTC)
- I am now monitoring the error logs on tool forge. Found a bug in gadget mode. My apologies for that being highly unreliable. AManWithNoPlan (talk) 19:05, 17 May 2022 (UTC)
- That's time-consuming work, @AManWithNoPlan, but very valuable in improving efficiency.
- I hesitate to add to your high workload, but here's an if-you-have-time-and-inclination suggestion: does the bot log failures or successes of the zotero to return a page title?
- I know that any such logs would be huge, but if they could be made even for a few days, then they would invaluable for my work in bare URL cleanup. They would help identify websites where bare URLs rarely or never get filled by Citation bot, and that would allow more targeted processing.
- They would be invaluable for @Rlink2's work on BareRefBot. If we could identify websites where Citation bot cannot fill the title, then articles with those URLs could be pre-processed by BareRefBot to cleanup those URLs. I know of some high-profile ones like nytimes.com and cbc.ca, but if we had a longer list then BareRefBot could make a huge dent in the bare URL backlog without burdening Citation bot for a first pass. It might also be feasible to tag such URLs as not-CB-fixable, and thereby exclude those bare URLs from my CB list-making. BrownHairedGirl (talk) • (contribs) 22:22, 17 May 2022 (UTC)
- I have wondered the same thing over a year ago, and they don't. But, I will think about if it is possible to put stuff in the error log, other than crashes. AManWithNoPlan (talk) 23:26, 17 May 2022 (UTC)
- I was thinking that it would need to be a pair of new logs (e.g. ZoteroNoTitle and ZoteroGotTitle), so that they would be free of other data and it would be would simple to compare them to distinguish transient failures from persistent failures. BrownHairedGirl (talk) • (contribs) 23:48, 17 May 2022 (UTC)
- I have wondered the same thing over a year ago, and they don't. But, I will think about if it is possible to put stuff in the error log, other than crashes. AManWithNoPlan (talk) 23:26, 17 May 2022 (UTC)
- I am now monitoring the error logs on tool forge. Found a bug in gadget mode. My apologies for that being highly unreliable. AManWithNoPlan (talk) 19:05, 17 May 2022 (UTC)
- The slowest step - by far - in citations is determining that DOI's are broken. That was a surprise. AManWithNoPlan (talk) 15:57, 17 May 2022 (UTC)
My new best friend will be file_put_contents( $filename, $data, FILE_APPEND); AManWithNoPlan (talk) 00:14, 18 May 2022 (UTC)
- The People Endorse Your Wise Choice Of Friend. By 102% of the votes in a 105% turnout. BrownHairedGirl (talk) • (contribs) 00:26, 18 May 2022 (UTC)
- Also, tracking failed ones would allow the bot to skip them in the future and speed up runs. AManWithNoPlan (talk) 15:57, 18 May 2022 (UTC)
- Tracking is now running. Existing runs will not track. Files are ZoteroWorked & ZoteroFailed. Note that is includes any failure, including page not found etc. AManWithNoPlan (talk) 17:36, 18 May 2022 (UTC)
- Wow! @AManWithNoPlan, you are on the case as fast a fit terrier. This is great work.
- Can you give me any idea of the structure of the data files (or a few sample lines), so that I can begin thinking about how to analayse them? I want to be able to distinguish difft types of failure.
- If this allows CB to skip some sites, that will be a big help to CB's efficiency.
- @Rlink2: this development is very significant for bare URL cleanup. It will be a big help for BareRefBot. If we can identify websites for which Citation bot cannot fill the ref, the there will be no need to run CB on those pages first: BareRefBot can just get to work on any links to those sites.
- If BareRefBot builds its own list of websites for which it can never get a title, then we can intersect that with the CB-cannot-fix set to build a list of bot-unfixable websites. That would allow us to tag those bare refs as bot-unfixable, which would allow them to be excluded from bot list selections and possibly to display a note that they are bot-unfixable. BrownHairedGirl (talk) • (contribs) 20:23, 18 May 2022 (UTC)
- Tracking is now running. Existing runs will not track. Files are ZoteroWorked & ZoteroFailed. Note that is includes any failure, including page not found etc. AManWithNoPlan (talk) 17:36, 18 May 2022 (UTC)
- Also, tracking failed ones would allow the bot to skip them in the future and speed up runs. AManWithNoPlan (talk) 15:57, 18 May 2022 (UTC)
bot adds |lay-format= (deprecated) without also adding |lay-url= (also deprecated) ...
- Status
- Fixed
- Reported by
- Trappist the monk (talk) 15:48, 30 May 2022 (UTC)
- What happens
- Apparently this is a valid Trove url:
https://trove.nla.gov.au/newspaper/article/2556288?searchTerm=Mrs%20Dunlop&searchLimits=l-decade=184|||l-format=Article|||l-year=1842
- From that, the bot created a whole new
{{cite web}}
– should have been{{cite news}}
since the bot knew enough to include|newspaper=
... - What should happen
- Certainly
{{cite news}}
instead of{{cite web}}
; whatever code there is that wants to add any of|lay-date=
,|lay-format=
,|lay-source=
, and|lay-url=
should be disabled or removed. For this particular example, the url could have been truncated to https://trove.nla.gov.au/newspaper/article/2556288?searchTerm=Mrs%20Dunlop - Relevant diffs/links
- Diff
Goofy Google Books meta-data
- Status
- {{fixed}}
- Reported by
- Guliolopez (talk) 00:15, 31 May 2022 (UTC)
- What happens
- The bot has added values to the author name fields, where the values are not author names
- What should happen
- While the bot has added the first and last name values, verbatim, with full trust in the accuracy of the source metadata, this trust would appear to be misplaced. At least in this case. Perhaps a RegExp to identify dates or other errant numerics would be useful? ("-1922" is clearly not a person name. Perhaps if there are no alpha characters (or only numerics/non-alphanumerics) in a person name field, the script/bot might recognise that it isn't a valid person name?)
- Relevant diffs/links
- https://en.wiki.x.io/w/index.php?title=Ballyconnell&diff=1090715246&oldid=1086133547
- We can't proceed until
- Feedback from maintainers
Bot imports markup with title
- Status
- {{fixed}}
- Reported by
- BrownHairedGirl (talk) • (contribs) 15:20, 4 June 2022 (UTC)
- What happens
- bot filled a ref from DOI, but wrapped the title in
<strong> ... </strong>
tags, which I presume were in the source - What should happen
- bot should strip HTML tags from the title before adding it
- Relevant diffs/links
- https://en.wiki.x.io/w/index.php?title=Hypsioma_carioca&diff=1091475810&oldid=1091475719
- We can't proceed until
- Feedback from maintainers
I took a quick look at the HTML source for the URL. The issue isn't on ResearchGate's end, the meta tags are pretty clean. The DOI resolves to this article on Biotaxa, which has the following metadata and HTML fields for title:
<title>A new genus and two new species of Apomecynini, a new species of Desmiphorini, and new records in Lamiinae and Disteniidae (Coleoptera) | Zootaxa</title>
- Lines 6-9<meta name="citation_title" content="<p><strong>A new genus and two new species of Apomecynini, a new species of Desmiphorini, and new records in Lamiinae and Disteniidae (Coleoptera)</strong></p>"/>
- Line 23<meta name="DC.Title" content="<p><strong>A new genus and two new species of Apomecynini, a new species of Desmiphorini, and new records in Lamiinae and Disteniidae (Coleoptera)</strong></p>"/>
- Line 63<h1 class="page_title"><p><strong>A new genus and two new species of Apomecynini, a new species of Desmiphorini, and new records in Lamiinae and Disteniidae (Coleoptera)</strong></p></h1>
- Line 221-223
I'm not sure which of the four fields the bot is scraping, but three of them do contain the <strong>...</strong>
tag in some form. Sideswipe9th (talk) 16:23, 4 June 2022 (UTC)
- Oh and the meta tags do really have the entities encoded that way in the content fields. Just in case ya'all thought it might be a issue with the way {{code}} rendered it onto this page. Sideswipe9th (talk) 16:26, 4 June 2022 (UTC)
- Thanks, @Sideswipe9th. That is some mighty ugly HTML.
- P inside H1! Eek.
- HTML inside meta tags? Aaarrgh. BrownHairedGirl (talk) • (contribs) 17:18, 4 June 2022 (UTC)
- the bot does not parse the HTML, crossref has the bad data also https://search.crossref.org/?from_ui=&q=10.11646%2Fzootaxa.4691.5.8 the bot strips a lot of html. i will look at this example AManWithNoPlan (talk) 18:41, 4 June 2022 (UTC)
Caps: BioMed Research International
Bot broke URL
- Status
- Not a bug
- Reported by
- BrownHairedGirl (talk) • (contribs) 23:08, 3 June 2022 (UTC)
- What happens
- bot rewrote URL from (valid) https://www.researchgate.net/publication/228163323_Spectrum_Rights_in_the_Telecosm_to_Come → (invalid) https://www.researchgate.net/publication/228163323
Background: In this edit[14], I added the LCCN which I got from https://digital.sandiego.edu/sdlr/vol41/iss1/16/ ... and then invoked Citation bot. The bot added nothing, but broke the URL - What should happen
- nothing
- Relevant diffs/links
- https://en.wiki.x.io/w/index.php?title=Orthogonal_frequency-division_multiplexing&diff=1091378687&oldid=1091378283
- Both those link resolve to the exact same page. Headbomb {t · c · p · b} 23:27, 3 June 2022 (UTC)
- (edit conflict)
- Both ResearchGate urls work for me. What the bot did not do was supply a value for
|title=
(that particular LCCN isn't going to give that information). Since the cited source is the author's copy that they uploaded to ResearchGate,{{cite web}}
is correct. Because|lccn=n79122466
refers to 'University of San Diego. School of Law' (LCCN n791-22466) that identifier is misleading and doesn't help the reader locate the cited ResearchGate source so should be omitted. If you want to cite the source as a San Diego Law Review article, then write a{{cite journal}}
template and use the SDLR url that you found (also free-to-read and, unlike the ResearchGate copy, is the article-of-record). - —Trappist the monk (talk) 23:38, 3 June 2022 (UTC)
- The LCCN, is the LCCN for the publisher University of San Diego. School of Law. The Bot does not use LCCN's to expand anything anyway, since they are often the wrong thing. AManWithNoPlan (talk) 23:58, 3 June 2022 (UTC)
Thanks, all ... and sorry for the false alarm. Kinda weird: when I first tried it, https://www.researchgate.net/publication/228163323 gave me an error page. Now when I tried it again, it works. Maybe I screwed up copying the new URL from the diff.
Thanks for the explainer on LCCN. After this failed, I did then rewrite the ref[15] to use the SDLR URL.
This arose from trying to systematically fill ResearchGate bare URLs. About half of the URLs are to pages which do supply a DOI, so I wrote a script to scrape them from the page's headers, which are well structured: e.g. <meta property="citation_doi" content="10.1145/2465554.2465559">
. That worked nicely for about 20 URLs, until the site started giving me HTTP 429 errors ("too many requests"). Setting delays of over ten minutes between requests allows me to make some more progress before it's back to the 429s, but only if I do manual requests simultaneously.
I am beginning to dislike ResearchGate. Damn these commercial outfits grabbing part of the academic space, and then protecting their turf. --BrownHairedGirl (talk) • (contribs) 00:53, 4 June 2022 (UTC)
- I asked RG about allow the Bot IP address, and they told me to kiss off basically. The Bot ignores RG URLs for that reason. AManWithNoPlan (talk) 12:57, 4 June 2022 (UTC)
- @AManWithNoPlan: wot bar stewards. I thought of making my own approach, but now I see no point.
- I have adapted my method to use v intermittent script alongside manual work. That combination is keeping me clear of their filters, so I am making some progress: down from 639 to under 500 in the last few days. BrownHairedGirl (talk) • (contribs) 15:17, 4 June 2022 (UTC)
Do not change editor-first1 to editor1-first
- Status
- Fixed
- Reported by
- Hawkeye7 (discuss) 23:36, 31 May 2022 (UTC)
- What happens
- editor-firstn parameter replaced by editorn-first, and editor-lastn parameter replaced by editorn-last. This is not the preferred form. Classic case of WP:COSMETICBOT.
- Relevant diffs/links
- [16] Was reported fixed in 2018 but back again.
Evidence to support your This is not the preferred form
claim? If I believe these search results (some cirrus searches time out so they aren't definitive):
then, |editorn-last=
is used more often than |editor-lastn=
.
Similarly:
Were it up to me, there would be only one enumerated form for these and their author, contributor, interviewer, translator counterparts. Alas, it is not up to me, both forms are allowed and may be used. In the best of all possible worlds, the bot will use the form predominant in the article that it is editing.
—Trappist the monk (talk) 00:36, 1 June 2022 (UTC)
- The citation template documentation is not consistent. That could use some cleanup. I have deleted the code that normalized the editors. AManWithNoPlan (talk) 01:37, 1 June 2022 (UTC)
- There is no problem with normalizing editors as part of a substantive edit. Izno (talk) 02:46, 1 June 2022 (UTC)
- @Trappist the monk: I meant only that the documentation does not favour one form over the other, and both are equally valid. Personally, I prefer the number last form, as it is easier to edit by changing the last character than by altering one in the middle. It's also consistent with the more common firstn form. Hence my Watchlist being cluttered with these changes. Hawkeye7 (discuss) 03:24, 1 June 2022 (UTC)
- There is no problem with normalizing editors as part of a substantive edit. Izno (talk) 02:46, 1 June 2022 (UTC)
- I prefer number last also. AManWithNoPlan (talk) 11:16, 1 June 2022 (UTC)
- Moved your comment. I would like to know where you see the cs1|2 documentation being inconsistent.
- —Trappist the monk (talk) 12:07, 1 June 2022 (UTC)
- "Full parameter set in horizontal format" and "Full parameter set in vertical format" are editorN-last. "editors=" is suggested to be replaced with editor-lastN. "Editors" section suggests " for multiple editors, use editor-last1, editor-first1 through editor-lastN". "This template has custom formatting." area only lists editorN-last style. "Recently removed CS1/CS2 parameters" just adds dashes and suggests both. AManWithNoPlan (talk) 14:07, 1 June 2022 (UTC)
- I guess I gotta wonder if the horizontal and vertical full-parameter-set listings are realy of any value because they don't show all name-holding parameter options. We could, I suppose, add links to an end note that would tell editors that 'this enumerated form is used here but this other enumerated form is also supported...' or some such. Same for the actual editor parameter documentation. 'This template has custom formatting', which I presume you mean the abomination that is TemplateData, is not cs1|2 documentation and should never be assumed to be correct – see Template:Cite arXiv § TemplateData as an example. The recently removed parameters table is more or less correct in that it shows that the unhyphenated parameter names have direct hyphenated replacements;
|editors=
in that table could use some work. I believe that the documentation should use a single enumerated form throughout the documentation for the avoidance of confusion; aliases should be mentioned in footnotes or parenthetically. - —Trappist the monk (talk) 15:04, 1 June 2022 (UTC)
- I guess I gotta wonder if the horizontal and vertical full-parameter-set listings are realy of any value because they don't show all name-holding parameter options. We could, I suppose, add links to an end note that would tell editors that 'this enumerated form is used here but this other enumerated form is also supported...' or some such. Same for the actual editor parameter documentation. 'This template has custom formatting', which I presume you mean the abomination that is TemplateData, is not cs1|2 documentation and should never be assumed to be correct – see Template:Cite arXiv § TemplateData as an example. The recently removed parameters table is more or less correct in that it shows that the unhyphenated parameter names have direct hyphenated replacements;
- "Full parameter set in horizontal format" and "Full parameter set in vertical format" are editorN-last. "editors=" is suggested to be replaced with editor-lastN. "Editors" section suggests " for multiple editors, use editor-last1, editor-first1 through editor-lastN". "This template has custom formatting." area only lists editorN-last style. "Recently removed CS1/CS2 parameters" just adds dashes and suggests both. AManWithNoPlan (talk) 14:07, 1 June 2022 (UTC)
Bot still imports markup with title
- Status
- Fixed
- Reported by
- BrownHairedGirl (talk) • (contribs) 12:30, 6 June 2022 (UTC)
- What happens
- Bot imports HTML into title when filing ref from http://dx.doi.org/10.11646/zootaxa.3962.1.7
added|title=<p class="HeadingRun ''In''"> Cyprinid fishes of the genus ''Neolissochilus ''in Peninsular Malaysia. |year=2015 |last1=Khaironizam
(Followup from User talk:Citation bot/Archive_ 32#Bot_imports_markup_with_title) - Relevant diffs/links
- https://en.wiki.x.io/w/index.php?title=Mahseer&diff=1091791591&oldid=1091787637
Every HTML tag has to be handled explicitly, since the bot has to support math and physics titles. AManWithNoPlan (talk) 13:47, 6 June 2022 (UTC)
Incorrect bot guess
- Status
- Fixed - now do not search for PMID for books
- Reported by
- Invasive Spices (talk) 5 June 2022 (UTC)
- What happens
- Alteration of a book's (Gross et al. 2014) fields with some but not all fields from a journal article.
- What should happen
- No alteration necessary.
- Relevant diffs/links
- [17]
- Replication instructions
- My guess. If a book has same author name fields but different order and same title to a journal article.
- We can't proceed until
- Feedback from maintainers
Bot added "chapter" section to ref with "cite conference" tag
- Status
- {{fixed}} for the most part
- Reported by
- 188.66.35.138 (talk) 04:08, 4 June 2022 (UTC)
- What happens
- Bot added "chapter" section with duplicate of title section content to a ref with "cite conference" tag.
- What should happen
- Bot must respect current syntax for references in the first place (chapter sec is for refs with "cite book" etc.) and never do such edits.
In general, please consider getting rid of "add chapter" piece of code altogether (even for books): a good chapter-finding algo that always works is non-trivial and doesn't really add much value. Imo, effort is much better spent on improving date-finding code – also non-trivial to cover all cases, but very useful.
- Relevant diffs/links
- http://en.m.wiki.x.io/wiki/Special:MobileDiff/1091168664
- We can't proceed until
- Feedback from maintainers
Non-mobile diff https://en.wiki.x.io/w/index.php?diff=1091168664 AManWithNoPlan (talk) 13:49, 4 June 2022 (UTC)
- This resulted from the citation having the wrong title to begin with. AManWithNoPlan (talk) 19:25, 4 June 2022 (UTC)
Multiple issues with conference paper citations – please stop the bot from doing further damage until fixed
- Status
- Everything that is fixable is now {{fixed}}
- Reported by
- 188.66.35.138 (talk) 08:59, 4 June 2022 (UTC)
- What happens
- Flawed correction is again associated with citation of a conference paper, this time "META II a syntax-oriented compiler writing language".
1. Instead of quoting conference title verbatim, the bot adds "on -" at the end resulting in e.g. "Proceedings of the 1964 19th ACM national conference on - " title.
2. The bot does not respect "Do not wikilink "chapter" if "chapter-url" is provided" rule for book citations.
3. The bot uses "cite book" instead of "cite conference".
Quite frankly, I got lucky to catch this bot doing all this stuff (along with the bug I reported yesterday), and since not everyone by far is sharp-eyed or caring enough to report such bugs, I would strongly suggest shutting down the bot immediately to prevent it from spoiling more articles and fix it.
- Relevant diffs/links
- http://en.m.wiki.x.io/wiki/Special:MobileDiff/1091235355
http://en.m.wiki.x.io/wiki/Special:MobileDiff/1091419654
- We can't proceed until
- Feedback from maintainers
Non-mobile diffs
https://en.wiki.x.io/w/index.php?diff=1091419654 https://en.wiki.x.io/w/index.php?diff=1091235355
AManWithNoPlan (talk) 12:52, 4 June 2022 (UTC)
- I will fix (1). Can you give an example of (2), I would like to see that. (3) is not a bug, the {{cite conference}} is really intended for proceedings that are not turned into a book or a special journal issue. AManWithNoPlan (talk) 12:54, 4 June 2022 (UTC)
- (2) is also in one of the linked diffs (please search for the quoted paper title to find it).
- I will fix (1). Can you give an example of (2), I would like to see that. (3) is not a bug, the {{cite conference}} is really intended for proceedings that are not turned into a book or a special journal issue. AManWithNoPlan (talk) 12:54, 4 June 2022 (UTC)
- Regarding (3), I must say I'm surprized to hear that, as "cite conference" template has a number of params specifically for printed proceedings (book-title, publisher, volume, isbn etc.) – could you explain the rationale behind preferred use of "cite book"?
- By the way, thanks for cleaning up the mess left by the bot yesterday (much appreciated), and best of luck fixing these and other bugs. — Preceding unsigned comment added by 188.66.35.138 (talk) 14:12, 4 June 2022 (UTC)
- cite conference parameters are significantly different from the rest of Cs1/2 and editors often get them wrong. it really is impossible for the bot to make a cite conference since ACM uses book or journal formatting for all their meta-data. AManWithNoPlan (talk) 15:06, 4 June 2022 (UTC)
- Yes, the point of cite conference is not to cite things that happen at a conference but to cite the conference proceedings. Izno (talk) 19:00, 4 June 2022 (UTC)
- Agree that the only point of cite conference can be to cite published conference proceedings, because that is the only output from a conference that is citable. As the editor of multiple published conference proceedings, I don't understand what sort of thing you mean by "proceedings that are not turned into a book". It is like you are talking about "bananas that are not fruit". Proceedings are a special type of book. No turning-into is needed to make them a book. They already are one. —David Eppstein (talk) 19:04, 4 June 2022 (UTC)
- By the way, thanks for cleaning up the mess left by the bot yesterday (much appreciated), and best of luck fixing these and other bugs. — Preceding unsigned comment added by 188.66.35.138 (talk) 14:12, 4 June 2022 (UTC)
- https://github.com/ms609/citation-bot/commit/86518d8c928ff9eaec9c05bf55d2ea3f79e1a4d9 Bot will not add a chapter-url, if the chapter is wiki-linked. AManWithNoPlan (talk) 19:33, 4 June 2022 (UTC)
Caps: mLife
Changes cite conference title/conference to chapter/conference/title
- Status
- Fixed by avoiding modifying cite conference unless it is very incomplete. The parameters are just too different from the rest of CS1/CS1
- Reported by
- —David Eppstein (talk) 21:28, 9 June 2022 (UTC)
- What happens
- For a cite conference with title=Title of paper and conference=Title of conference, it changes the title= parameter to chapter= and adds a new title=Title of conference parameter, leaving the conference= parameter unchanged, so that the title of the conference appears twice in the resulting citation
- What should happen
- Not that. The citation worked correctly with the previous combination of parameters, and incorrectly with the new parameters, so it should have been left unchanged. —David Eppstein (talk) 21:28, 9 June 2022 (UTC)
- Relevant diffs/links
- Special:Diff/1092364544
- We can't proceed until
- Feedback from maintainers
|title=
gets title of the paper; |book-title=
gets title of the published proceedings; see Template:Cite conference § Title; |conference=
gets name/date/location of the conference (mostly unnecessary when you have |book-title=
; see Template:Cite conference § Conference (documentation for |conference-url=
is misleading because |conference-url=
links |conference=
). For years I have been saying that this template should be revised but the notion of revision has been met which resounding indifference ...
In this example, the template should have been written:
{{Cite conference |last1=Brodal |first1=G. S. L. |last2=Lagogiannis |first2=G. |last3=Tarjan |first3=R. E. |title=Strict Fibonacci Heaps |book-title=STOC’12: Proceedings of the 2012 ACM Symposium on Theory of Computing |location=New York |publisher=The Association for Computing Machinery |pages=1177–1184 |year=2012 |doi=10.1145/2213977.2214082 |isbn=978-1-4503-1245-5 |url=http://www.cs.au.dk/~gerth/papers/stoc12.pdf |via=Aarhus University}}
- Brodal, G. S. L.; Lagogiannis, G.; Tarjan, R. E. (2012). "Strict Fibonacci Heaps" (PDF). STOC’12: Proceedings of the 2012 ACM Symposium on Theory of Computing. New York: The Association for Computing Machinery. pp. 1177–1184. doi:10.1145/2213977.2214082. ISBN 978-1-4503-1245-5 – via Aarhus University.
Pubmed url / pmc cleanup
Further pubmed/pmc cleanup
- What should happen
- [21]
- We can't proceed until
- Feedback from maintainers
Better cleanup of issues
Use for cosmetic edits
It's unfortunate that Citation bot is making edits like this that are entirely cosmetic. {{u|Sdkb}} talk 17:28, 12 June 2022 (UTC)
- A bug clearly. Headbomb {t · c · p · b} 17:55, 12 June 2022 (UTC)
- That is very odd. AManWithNoPlan (talk) 17:58, 12 June 2022 (UTC)
- Fixed. Very odd code case. AManWithNoPlan (talk) 18:13, 12 June 2022 (UTC)
- Thanks! {{u|Sdkb}} talk 18:14, 12 June 2022 (UTC)
- Fixed. Very odd code case. AManWithNoPlan (talk) 18:13, 12 June 2022 (UTC)
- That is very odd. AManWithNoPlan (talk) 17:58, 12 June 2022 (UTC)
User sandboxes
Can user sandboxes be excluded from "fixing" by the bot? I've had plain URL references in my sandbox changed several times over the past week into messy citations that I have to keep reverting. SounderBruce 21:55, 11 June 2022 (UTC)
- That's what {{nobots}} is there for. However, you often don't understand what you're reverting, like this, which removed useless parameters. Headbomb {t · c · p · b} 22:12, 11 June 2022 (UTC)
- I can't see the point of the bot editing sandboxes, esp when they are targeted -- as happened this week, on a big scale.
- Many of the sandboxes edited are stale, so those edits are useless.
- For fresh sandboxes, there will be many cases where an editor has added lots of empty parameters, to be filled as the draft is developed. Having the bot remove those empty params is disruptive.
- The general principle is that editors should be free to do whatever they want with their sandboxes. There are some exceptions to that principle, but I dont see any way that Citation bot's edits fall within those exceptions.
- I don't see why editors should have to add {{nobots}}. The bot should respect WP:USERPAGE and leave all userpages alone, without having to be specifically excluded from every individual page. The only situation in which the bot should edit a userpage is if the page's owner asks Citation bot to clean it up. BrownHairedGirl (talk) • (contribs) 09:48, 12 June 2022 (UTC)
- They should certainly be excluded from mass/category runs, but the bot shouldn't be hard-blocked from being run on a sandbox. Headbomb {t · c · p · b} 10:51, 12 June 2022 (UTC)
- There was a run fixing parameter errors (such as dead-url), that run is now done. Something went wrong in generating the list of pages and new sandboxes gor edited also. AManWithNoPlan (talk) 11:27, 12 June 2022 (UTC)
- @AManWithNoPlan: I still see no need to edit the sandboxes.
- If an old sandbox page has a template error, so what? It's old and stale.
- If a fresh sandbox page has a template error, why does the bot intervene without a request?
- I agree with @Headbomb that user sandboxes
should certainly be excluded from mass/category runs
, just like drafts. BrownHairedGirl (talk) • (contribs) 16:45, 12 June 2022 (UTC)
- There was a run fixing parameter errors (such as dead-url), that run is now done. Something went wrong in generating the list of pages and new sandboxes gor edited also. AManWithNoPlan (talk) 11:27, 12 June 2022 (UTC)
- They should certainly be excluded from mass/category runs, but the bot shouldn't be hard-blocked from being run on a sandbox. Headbomb {t · c · p · b} 10:51, 12 June 2022 (UTC)
Incorrect URL encoding of |
- Status
- {{fixed}}
- Reported by
- SnorlaxMonster 14:42, 18 June 2022 (UTC)
- What happens
- When changing bare references that include
|
, they are replaced by%7B
(which is the URL encoded version of{
), instead of%7C
(the correct URL encoded version of|
). - Relevant diffs/links
- Special:Diff/1093060081
- We can't proceed until
- Feedback from maintainers
Incorrect "date" for Spotify releases
- Status
- {{fixed}}
- Reported by
- Eurohunter (talk) 21:46, 18 June 2022 (UTC)
- What happens
- Adds release dates as publication dates in parameter "date" for Spotify and possibly other music stores and music streaming services.
- Relevant diffs/links
- https://en.wiki.x.io/w/index.php?title=Fest_i_hela_huset&diff=1093793274&oldid=1092908604
- We can't proceed until
- Feedback from maintainers
Release date of single in Spotify ≠ date of page publication of this singe in Spotify (the page may have been published before the release date as pre-order. A situation where the bot added wrong dates previously occured with iTunes and Apple Music pages. Was it fixed? Eurohunter (talk) 21:46, 18 June 2022 (UTC)
Caps: Journal of e-Learning and Knowledge Society
- What should happen
- [23]
- We can't proceed until
- Feedback from maintainers
website gives start date for date range
- Status
- {{fixed}} by flagging this as no date website
- Reported by
- DavidMCEddy (talk) 22:56, 19 June 2022 (UTC)
- We can't proceed until
- Feedback from maintainers
Giving citation date of 1947 for documentation of something in 2011, etc.:
https://en.wiki.x.io/w/index.php?title=Economy_of_the_United_States&diff=next&oldid=1093846318
I will undo this, but it is doubtless messing things up elsewhere.
URL cleanup: #!divAbstract
- What should happen
- [24]
- We can't proceed until
- Feedback from maintainers
Adds obviously-wrong website date (adds first date of data range)
- Status
- new bug
- Reported by
- —David Eppstein (talk) 07:35, 18 June 2022 (UTC)
- What happens
- In Special:Diff/1093654512 it adds a 2012 date to a website used as a source for a claim about something that happened in 2021. The website has no visible publication date in its text. Its metadata includes <meta property="article:published_time" content="2012-06-20T00:00+00:00" /> <meta property="article:modified_time" content="2022-02-18T23:10+00:00" />. Obviously, the modified_time is the more appropriate choice of what date to use, if a date from the metadata must be used.
- What should happen
- My preference would be to leave websites that do not display a visible date undated. But if they are to be dated, it is absolutely incorrect to give the date of a version of the site that does not include the sourced information. The only safe choice is the modified time, but that will also be incorrect in many other cases. Better just not to date at all.
- We can't proceed until
- Feedback from maintainers
- Right. Besides which, the only correct date is the date when the website was inspected to extract the information it is cited for. Neither the creation nor modification date correspond to that. Zerotalk 07:44, 18 June 2022 (UTC)
- Huh? That would be
|access-date=
, not|date=
. --Izno (talk) 14:48, 18 June 2022 (UTC) - Not necessarily the bot's fault. The site's metadata contains:
<meta property="article:published_time" content="2012-06-20T00:00+00:00" />
<meta property="article:modified_time" content="2022-02-18T23:10+00:00" />
- IceWelder [✉] 15:10, 18 June 2022 (UTC)
- Yes, it does contain those times, AS I EXPLAINED CLEARLY IN MY BUG REPORT. It is the bot's fault for misinterpreting those times and using them to falsify a reference. —David Eppstein (talk) 17:55, 18 June 2022 (UTC)
- My bad, must have slipped my gaze. However, I do recall several cases where the datePublished/article:published_time data was correct but (presumably by mistake) not actually shown; GameSpot used to do that a lot. In such cases, I would prefer to bot to add that date. The reference from the bug report is probaly an outlier in that it was created and timestamped years ago and then continuously updated to fit entirely new content, not just fix typos. Difficult to say how the bot could avoid making this mistake without blanketly stopping such date imports. Does Citation bot respect {{cbignore}}? IceWelder [✉] 11:56, 19 June 2022 (UTC)
- The bot can avoid this mistake by noticing the wide range between publication and modification dates, noticing that its subhuman intelligence is not adequate to resolve that wide range, and doing nothing. Relying on other editors to detect these problems individually and tag articles with special tags to force the bot to stop falsifying dates is a non-solution that will not stop the majority of date falsifications by the bot. —David Eppstein (talk) 20:17, 19 June 2022 (UTC)
- My bad, must have slipped my gaze. However, I do recall several cases where the datePublished/article:published_time data was correct but (presumably by mistake) not actually shown; GameSpot used to do that a lot. In such cases, I would prefer to bot to add that date. The reference from the bug report is probaly an outlier in that it was created and timestamped years ago and then continuously updated to fit entirely new content, not just fix typos. Difficult to say how the bot could avoid making this mistake without blanketly stopping such date imports. Does Citation bot respect {{cbignore}}? IceWelder [✉] 11:56, 19 June 2022 (UTC)
- Yes, it does contain those times, AS I EXPLAINED CLEARLY IN MY BUG REPORT. It is the bot's fault for misinterpreting those times and using them to falsify a reference. —David Eppstein (talk) 17:55, 18 June 2022 (UTC)
- Huh? That would be
Now this is interesting - I ask on this particular talk page whether or not it's acceptable to add page publication dates that are stated in the page's metadata but not on the page itself, and just one day later this issue is brought up of Citation Bot adding page publication dates that are stated in the metadata, are not stated on the page itself, and are (shall we say) open to misinterpretation.
I have to say, I find myself agreeing with David - if a page/site does not display a visible publication date (I know I'm bringing up LeatherLicensePlates.com again, but none of that site's pages display visible publication dates), then it is probably better to leave them undated - even if the publication dates stated in the metadata (and, for that matter, the modification dates too) are correct. Klondike53226 (talk) 21:05, 18 June 2022 (UTC)
- I have added simonsfoundation.org to the list of no-date websites. This is one of the few websites that give the starting date for the range instead of the ending date for the range. AManWithNoPlan (talk) 12:56, 20 June 2022 (UTC)
Check year when fixing volume/issue=N/A
Parse title better
- What happens
- Title parsing is iffy, which makes the rest of the bot logic not kick in.
- What should happen
- [27]
Cite patent support
Is it possible to add support for autofilling cite patent from just a country and patent number? I was about to fix a page where someone used cite web for patents and was about to just replace it all with cite patent and just the country and number, and then get citation bot to do the rest when I realise citation bot doesn't do patents.
I think I will take a look at the github and try implementing it myself, but it has been years since I coded anything, so that probably won't work out.Kylesenior (talk) 02:44, 30 May 2022 (UTC)
- If you can give examples of URLs that patents usually have, then the bot can parse them. AManWithNoPlan (talk) 13:22, 30 May 2022 (UTC)
Cite news work
or newspaper
parameter question
Preview-editing the lead section of Presidency of Rodrigo Duterte and using the Citation bot converts work=Washington Post
to newspaper=Washington Post
, while other Philippine online news (like work=Philippine Daily Inquirer
) sites are ignored, which might confuse other future editors because the code is not standardized. Is there a possibility to set the bot to either:
- disable the conversion of
work
tonewspaper
or - allow the bot to recognize these Philippine news sites and convert their
work
parameter tonewspaper
?
—Sanglahi86 (talk) 14:18, 6 June 2022 (UTC)
Fixed for that one newspaper. Others can be added. AManWithNoPlan (talk) 18:54, 22 June 2022 (UTC)
Duplicated journal number
- Status
- Fixed with new PREFERS_VOLUME array and code
- Reported by
- Deor (talk) 21:03, 20 June 2022 (UTC)
- What happens
- Duplicating information unnecessarily and incorrectly. (The journal in question only has volume [or issue] numbers, not both. The bot incorrectly implied that "volume 38, issue 38" is a thing, when it should be either "volume 38" or "issue 38".)
- Relevant diffs/links
- [28], [29]
- We can't proceed until
- Feedback from maintainers
JSTOR calls them "issues", despite them being volumes. I would add to "volumes only" list, but some years have issues. Will look into a specific fix. AManWithNoPlan (talk) 12:39, 21 June 2022 (UTC)
'work' to 'journal'?
Why is this bot changing 'work' to 'journal'? It adds no value in doing so; 'work' is the standard and 'journal' is an alias. The bot isn't even consistent in doing so. This seems like an arbitrary change by somebody. Praemonitus (talk) 03:16, 17 June 2022 (UTC)
- Because journal is clearer than journal, and if you have a cite journal, then journal is the standard parameter. Headbomb {t · c · p · b} 03:37, 17 June 2022 (UTC)
- It's not documented as the "standard parameter" for a journal paper, and hence that's just a personal style preference on your point. I reject that as a justification. Praemonitus (talk) 14:55, 17 June 2022 (UTC)
- {{Cite journal}}. All examples use
|journal=
. Headbomb {t · c · p · b} 19:13, 17 June 2022 (UTC)- My issue is with bot edits to {{citation}}, so your example is beside the point. Praemonitus (talk) 00:56, 18 June 2022 (UTC)
- {{Cite journal}}. All examples use
- It's also untrue that journal is just an alias for work. Some of the templates (especially {{citation}}) use the parameter name to choose among some variations of formatting (journals are formatted differently from magazines or web sites or newspapers). —David Eppstein (talk) 04:10, 17 June 2022 (UTC)
- Okay, thank you for the clarification David. In that case your script is working inconsistently. See, for example, this edit and lines 168, 181, and 202, among others. The {{citation}} documentation also needs to reflect your point for clarity. Praemonitus (talk) 14:55, 17 June 2022 (UTC)
- There is kind of a gray area of whether scholarly-society newsletters and bulletins should count as journals or as magazines. The visible differences involve formatting of volume/issue numbers, which are important for academic journals and usually unimportant for magazines and newspapers. For newsletters and bulletins, they are also important. There is also a semantic difference (actual journal articles typically go through a formal peer-review process but newsletter and bulletin articles do not) but that is not important for formatting purposes because it is not really something that the formatting makes visible to readers. So it is not unreasonable to call them journals for the purpose of citation formatting. —David Eppstein (talk) 07:37, 18 June 2022 (UTC)
- Okay, but this doesn't address the issue of why, for example, "Astronomy and Astrophysics" was changed to journal in some templates but not others. Praemonitus (talk) 13:50, 18 June 2022 (UTC)
- The bibcode actually was the reason. AManWithNoPlan (talk) 20:27, 20 June 2022 (UTC)
- They all have bibcodes, so I'm not clear what you're saying. Praemonitus (talk) 20:47, 20 June 2022 (UTC)
- The problem is multi-faceted. The {{citation}} is difficult to deal with. Secondly, "complete" references do not query the API's so they will not get told "this is a journal". I misread the diff and thought that bibcodes were the problem. AManWithNoPlan (talk) 18:46, 22 June 2022 (UTC)
- {{fixed}} the problem with {{citation}} AManWithNoPlan (talk) 20:13, 23 June 2022 (UTC)
- The problem is multi-faceted. The {{citation}} is difficult to deal with. Secondly, "complete" references do not query the API's so they will not get told "this is a journal". I misread the diff and thought that bibcodes were the problem. AManWithNoPlan (talk) 18:46, 22 June 2022 (UTC)
- They all have bibcodes, so I'm not clear what you're saying. Praemonitus (talk) 20:47, 20 June 2022 (UTC)
- The bibcode actually was the reason. AManWithNoPlan (talk) 20:27, 20 June 2022 (UTC)
- Okay, but this doesn't address the issue of why, for example, "Astronomy and Astrophysics" was changed to journal in some templates but not others. Praemonitus (talk) 13:50, 18 June 2022 (UTC)
- There is kind of a gray area of whether scholarly-society newsletters and bulletins should count as journals or as magazines. The visible differences involve formatting of volume/issue numbers, which are important for academic journals and usually unimportant for magazines and newspapers. For newsletters and bulletins, they are also important. There is also a semantic difference (actual journal articles typically go through a formal peer-review process but newsletter and bulletin articles do not) but that is not important for formatting purposes because it is not really something that the formatting makes visible to readers. So it is not unreasonable to call them journals for the purpose of citation formatting. —David Eppstein (talk) 07:37, 18 June 2022 (UTC)
- Okay, thank you for the clarification David. In that case your script is working inconsistently. See, for example, this edit and lines 168, 181, and 202, among others. The {{citation}} documentation also needs to reflect your point for clarity. Praemonitus (talk) 14:55, 17 June 2022 (UTC)
- It's not documented as the "standard parameter" for a journal paper, and hence that's just a personal style preference on your point. I reject that as a justification. Praemonitus (talk) 14:55, 17 June 2022 (UTC)
Bot converts a citation of a web page to a citation of a book mentioned on that web page
- Status
- that one page has been Fixed with flags eetc.
- Reported by
- – Arms & Hearts (talk) 21:53, 17 May 2022 (UTC)
- What happens
- The bot converts {{cite web}} citations to catalogue entries for books to {{cite book}}
- What should happen
- The bot should leave alone. If there was a good reason for the book to be cited instead of the web page (and if it was hypothetically possible for the bot to determine that), it should do so in the way books are usually cited, i.e. without the
|website=
or|access-date=
parameters, rather than a muddled halfway approach (as a result of which it seems to tell us, for example, that the page on the Naval Marine Archive website was published in 1961). - Relevant diffs/links
- https://en.wiki.x.io/w/index.php?title=The_Spanish_Civil_War_(book)&curid=61551753&diff=1088226376&oldid=1082664491
- We can't proceed until
- Feedback from maintainers
Web->Book: I don't think that it was right in this case...
- Status
- Fixed on the page.
- Reported by
- Shaav (talk) 19:29, 18 May 2022 (UTC)
- What happens
- web citation converted to book
- What should happen
- shouldn't change
- Relevant diffs/links
- https://en.wiki.x.io/w/index.php?title=Harry_Potter_in_translation&diff=1083930999&oldid=prev
- We can't proceed until
- Feedback from maintainers
I suspect because the citation is for a webpage about a book, that the 'web' citation was replace with a 'book' citation... but it really was intended to be a citation for the website (I created it) because the citation is supporting the existence of the book and it's properties. I'm not sure if there's something that can be included so that it doesn't get converted again in the future. — Preceding unsigned comment added by Shaav (talk • contribs) 19:29, 18 May 2022 (UTC)
- That looks fine to me? The title field still gets transformed into a link in the reflist as before. The only difference to the reader is that the title is now italicised, which makes sense as the citation should be to the book based upon how it's being used in context? Sideswipe9th (talk) 20:23, 18 May 2022 (UTC)
- It is not fine. It was a reference about a translation of a book, sourced to a publisher's web site describing the fact that they have published it. It was replaced by a reference directly to the book itself, with a link that is likely to be removed or replaced by something else (because for book references the link is not the important part). The purpose has been lost. —David Eppstein (talk) 04:25, 19 May 2022 (UTC)
- This is the same issue as the section above, which shows why this sort of thing's a bad idea even if in this case it only resulted in cosmetic changes. More generally, I'd say that while it's perfectly reasonable in some cases for a human editor to conclude that
the citation should be to the book
, that's not a decision we can expect a bot to make with any degree of reliability. – Arms & Hearts (talk) 18:46, 19 May 2022 (UTC)- I have flagged the ref to not be changed. AManWithNoPlan (talk) 15:42, 7 June 2022 (UTC)
Weird ref conversion bug
- What happens
- Doesn't leave the citation alone
- Relevant diffs/links
- [30]
Not sure why it wants to convert things in the first place, but the comment should halt the process. Headbomb {t · c · p · b} 04:42, 30 June 2022 (UTC)
Misleading edit summary
- Status
- Not a bug
- Reported by
- BrownHairedGirl (talk) • (contribs) 13:08, 26 June 2022 (UTC)
- What happens
- edit summary says
Alter: template type. Add: magazine. Removed parameters. Some additions/deletions were parameter name changes
... which is untrue. No parameters were removed - What should happen
- edit summary should be
Alter: template type. Add: magazine. Some additions/deletions were parameter name changes
- Relevant diffs/links
- https://en.wiki.x.io/w/index.php?title=Tom_Hollander&diff=prev&oldid=1095114825
Feature request: populate author-link
Couldn't citebot automatically populate |author-link=
? — Guarapiranga ☎ 06:54, 1 July 2022 (UTC)
- No. Besides it being highly context-sensitive about who is who (many people share the same names), author linking is often completely superfluous and unneeded. Headbomb {t · c · p · b} 12:11, 1 July 2022 (UTC)
- @Headbomb: How about adding an
|author-link=
parameter when the|author=
incorrectly contains a wikilink? For example:|author=[[John Doe]]
→|author=John Doe |author-link=John Doe
|author=[[John Doe|Doe, John]]
→|author=Doe, John |author-link=John Doe
|author2=[[John Doe]]
→|author2=John Doe |author-link2=John Doe
|author3=[[John Doe|Doe, John]]
→|author3=Doe, John |author-link3=John Doe
- Thanks! GoingBatty (talk) 00:46, 2 July 2022 (UTC)
- I think you will find editors who argue that such usages are not incorrect and should not be changed. Certainly, links in the author parameter are necessary if the authors are listed in Vancouver style in a single author parameter. The one that I find annoying and incorrect is when people put a link on part of a name using the last= parameter. —David Eppstein (talk) 00:48, 2 July 2022 (UTC)
- It is not 'wrong' to write any of the examples you gave; cs1|2 knows how to parse those links so both the rendered citation and the metadata are correct.
|author-link=
is required for|last=
/|first=
pairs.|first=[[<name>]]
is an error but because|last=
and|author=
are aliases of each other, assigning linked values to them is allowed. - —Trappist the monk (talk) 01:02, 2 July 2022 (UTC)
- @Trappist the monk: Then should we update Template:Citation Style documentation/author to remove the instruction "Do not wikilink—use author-link instead." so users know that assigning linked values to them is allowed? Thanks! GoingBatty (talk) 01:26, 2 July 2022 (UTC)
- Ask Editor Jonesey95 who added that instruction at this edit.
- —Trappist the monk (talk) 02:18, 2 July 2022 (UTC)
- This is the second time people have wondered about this instruction recently, and the previous time, I'm pretty sure I looked for a discussion about it and was unable to find one. Since putting wikilinks in the author fields appears to work without creating problems, I have removed that bit of instruction. – Jonesey95 (talk) 02:35, 2 July 2022 (UTC)
- @Trappist the monk: Then should we update Template:Citation Style documentation/author to remove the instruction "Do not wikilink—use author-link instead." so users know that assigning linked values to them is allowed? Thanks! GoingBatty (talk) 01:26, 2 July 2022 (UTC)
- @Headbomb: If many notable authors share the same name, the dab page would be linked, leaving it for editors to disambiguate, as usual; otherwise, if the citation is of an RS, its author is most likely to be the one that is notable among the many people sharing his name (and unlikely false positives can be remedied by article editors, who'll be clued in to the referenced authors' homonyms). — Guarapiranga ☎ 03:46, 2 July 2022 (UTC)
- Let's say there is one John Smith notable in the world, and all the other John Smiths are not notable, meaning there are no disambiguation pages at John Smith. What's the guarantee that the John Smith you are citing is the notable John Smith, and not one of the non-notable ones? Headbomb {t · c · p · b} 07:29, 2 July 2022 (UTC)
- {{notabug}} – disabled while the discussion continues
- @Headbomb: How about adding an
- No. Besides it being highly context-sensitive about who is who (many people share the same names), author linking is often completely superfluous and unneeded. Headbomb {t · c · p · b} 12:11, 1 July 2022 (UTC)
Suggestion: Could citebot learn from its mistakes?
I don't expect this to be easy, but it'd be great if citebot could learn from manual edits made to the citations it created. I suspect this would involve maintaining a (rather large) table of all auto-gen'd citations, indexed by ISBN, URL, etc, that could be routinely updated as editors correct and change citations. This will probably involve some sort of heuristic to decide what version (or mix of versions) should prevail against various citation instances of the same source. I'm not suggesting overrunning editors' consensus ref formulations, of course; just that in new citebot cites it uses past editors' input. — Guarapiranga ☎ 07:07, 1 July 2022 (UTC)
- This is not an AI bot, so this is impossible. Headbomb {t · c · p · b} 12:12, 1 July 2022 (UTC)
- I don't see how AI would be required, Headbomb; I use the term 'learn' in a loose, informal, sense, not actual machine learning. What I mean by it is that citebot, in its daily rounds, take note of citations it created that have been subsequently edited. A precursor to the (rather large) table of citebot citations already exists at WP:RSP. It seems to me Web2Cit has a manual version of this process. — Guarapiranga ☎ 04:00, 2 July 2022 (UTC)
- {{wontfix}}
bot removes publisher
- Status
- Fixed
- Reported by
- BrownHairedGirl (talk) • (contribs) 15:17, 4 July 2022 (UTC)
- What happens
{{cite web|author=Leeps |url=https://news.google.com/newspapers?nid=1309&dat=19890604&id=tKFUAAAAIBAJ&sjid=NpADAAAAIBAJ&pg=5932,900833&hl=en |title=Rust Busters |publisher=[[New Straits Times]] / [[Google News Archive]] |date=1989-06-04 |access-date=2015-05-03 }}
→{{cite web|author=Leeps |url=https://news.google.com/newspapers?nid=1309&dat=19890604&id=tKFUAAAAIBAJ&sjid=NpADAAAAIBAJ&pg=5932,900833&hl=en |title=Rust Busters |date=1989-06-04 |access-date=2015-05-03 }}
- What should happen
- nothing.
The old "newspaper/archive" format is non-standard, but|publisher=New Straits Times / Google News Archive
is much better than nothing. - Relevant diffs/links
- https://en.wiki.x.io/w/index.php?title=Opel_Rekord_Series_C&diff=prev&oldid=1096449963
- It could also change the reference to
{{cite news|author=Leeps |url=https://news.google.com/newspapers?nid=1309&dat=19890604&id=tKFUAAAAIBAJ&sjid=NpADAAAAIBAJ&pg=5932,900833&hl=en |title=Rust Busters |newspaper=[[New Straits Times]] |via=[[Google News Archive]] |date=1989-06-04 |access-date=2015-05-03 }}
GoingBatty (talk) 22:44, 4 July 2022 (UTC)- @GoingBatty: that would be ideal. It's what I later did manually.[31]
- However, I think it would be hard for the bot to reliably disentangle malformed data like this, so I would be content for the bot to just refrain from removing it. --BrownHairedGirl (talk) • (contribs) 00:11, 5 July 2022 (UTC)
CAPS: ESC Heart Failure
Caps: Current HIV/Aids Reports
- Status
- Fixed
- Reported by
- BrownHairedGirl (talk) • (contribs) 17:07, 6 July 2022 (UTC)
- What happens
- ref to http://hdr.undp.org/en/indicators/137506 already filled with {{Cite web}}. Citation bot:
A) converted from {{Cite web}} to {{Cite journal}}
B) added|last1=Nations |first1=United
- What should happen
- Nothing.
Both changes were unhelpful. The Human Development Report (HDI) does not look to me like a journal, and "United Nations" is not a human name. - Relevant diffs/links
- https://en.wiki.x.io/w/index.php?title=Economy_of_Turkey&diff=prev&oldid=1096788716
Mount Bandra Church
Bot says no changed reqd for Basilica of Our Lady of the Mount, Bandra just inserted a bare url Nolicamaca (talk) 01:24, 9 July 2022 (UTC)
- no idea. worked for me. google books might have been down for a second. AManWithNoPlan (talk) 12:09, 9 July 2022 (UTC)
A momentary bug i suppose thanks for doing it AManWithNoPlan may you can head over here WP:Teahouse Nolicamaca (talk) 14:30, 9 July 2022 (UTC)
Weird Author
- Status
- new bug
- Reported by
- SusanLesch (talk) 00:22, 12 July 2022 (UTC)
- We can't proceed until
- Feedback from maintainers
Greetings. Why do you create a news article out of a web citation in Minneapolis#Health_care? Beaverton isn't mentioned in the source, so where did you get her name? Thank you.
- That person is listed in the meta-data as the author. I have added a comment to make sure no one else adds it back. AManWithNoPlan (talk) 13:47, 12 July 2022 (UTC)
Authorization forgets page title
- Status
- sadly Won't fix
- Reported by
- SWinxy (talk) 01:07, 11 July 2022 (UTC)
- What happens
- Wikimedia Meta redirects back to https://citations.toolforge.org/process_page.php, but the site forgets what page I entered. It also frequently asks for authorization.
!Nothing requested -- OR -- pages got lost during initial authorization !No links to expand found
- What should happen
- The page should begin analyzing when the OAuth authorization is received.
- We can't proceed until
- Feedback from maintainers
The page loss is not fixable. AManWithNoPlan (talk) 13:35, 11 July 2022 (UTC)
- No idea why you would need to keep authorizing stuff. AManWithNoPlan (talk) 15:22, 11 July 2022 (UTC)
- I often face the issue because it often causes me to re-login. Could you elaborate on why page loss isn't fixable? From what I know of OAuth, a solution would be to add a parameter to the website to autofill the page, and have mediawiki redirect to that URL (e.g. https://citations.toolforge.org/?page=ASDF). SWinxy (talk) 20:24, 11 July 2022 (UTC)
- Are you using the webpage, or clicking on the "expand citations" on the left of the page you want to expand? AManWithNoPlan (talk) 13:50, 12 July 2022 (UTC)
- Wait... there's a sidebar action?? omg. I had no idea that was a thing. I've been using the webpage to do those citations all this time... SWinxy (talk) 18:40, 12 July 2022 (UTC)
- The reason I ask is that the BOT uses POST for everything on the webpage, and POST gets lost. GET does not get lost, and that is used for the thing on the side. We only allow GET for single pages for some technical reasons.. AManWithNoPlan (talk) 22:08, 12 July 2022 (UTC)
- Ah I see. Thanks for the clarification. SWinxy (talk) 22:33, 12 July 2022 (UTC)
- The reason I ask is that the BOT uses POST for everything on the webpage, and POST gets lost. GET does not get lost, and that is used for the thing on the side. We only allow GET for single pages for some technical reasons.. AManWithNoPlan (talk) 22:08, 12 July 2022 (UTC)
- Wait... there's a sidebar action?? omg. I had no idea that was a thing. I've been using the webpage to do those citations all this time... SWinxy (talk) 18:40, 12 July 2022 (UTC)
- Are you using the webpage, or clicking on the "expand citations" on the left of the page you want to expand? AManWithNoPlan (talk) 13:50, 12 July 2022 (UTC)
- I often face the issue because it often causes me to re-login. Could you elaborate on why page loss isn't fixable? From what I know of OAuth, a solution would be to add a parameter to the website to autofill the page, and have mediawiki redirect to that URL (e.g. https://citations.toolforge.org/?page=ASDF). SWinxy (talk) 20:24, 11 July 2022 (UTC)
Feature request: populate author-link (cont.)
This is the continuation of a thread from a week ago: |
Couldn't citebot automatically populate |author-link= ? — Guarapiranga ☎ 06:54, 1 July 2022 (UTC)
|
I failed to answer a question Headbomb asked me at the end:
Let's say there is one John Smith notable in the world, and all the other John Smiths are not notable, meaning there are no disambiguation pages at John Smith. What's the guarantee that the John Smith you are citing is the notable John Smith, and not one of the non-notable ones?
None. Citebot would pick the wrong John Smith when generating the citation, and editors would correct it before publishing the edit by simply removing |author-link=
(easier than checking whether the source author has a WP article, and manually adding it in). — Guarapiranga ☎ 04:46, 12 July 2022 (UTC)
- "Citebot would pick the wrong John Smith when generating the citation"
- This is exactly why this is not going to be implemented ever. There is also a WP:CITEVAR issue, issues of WP:OVERLINK, and also of link relevance. Headbomb {t · c · p · b} 05:27, 12 July 2022 (UTC)
inappropriate/wrong last & first
- Status
- Fixed
- Reported by
- Michael Bednarek (talk) 10:02, 14 July 2022 (UTC)
- What happens
- inappropriate/wrong
|last=
and|first=
added - What should happen
- nothing
- Relevant diffs/links
- here and previously in the same article
Adds weird date/volume
Refinement suggestions for Philippine citations
Hello. The following notable Philippine print newspapers have online versions. Please include them in the bot's list of recognized newspapers to process from |work=
to |newspaper=
. Currently, only Philippine Daily Inquirer (https://www.inquirer.net/
) is recognized.
Also, please check if the tabloids below warrant inclusion. Tempo, owned by Manila Bulletin, is arguably the most reputable tabloid in the Philippines. Pilipino Star Ngayon is a sister newspaper of The Philippine STAR, while Bandera is owned by the Philippine Daily Inquirer.
Also, for citations whose URL parameters start with |url=https://www.pna.gov.ph/
, please set the bot to use |work=Philippine News Agency
instead of |work=www.pna.gov.ph
, |agency=Philippine News Agency
, or |publisher=Philippine News Agency
. The citation should look like below:
{{cite news |title=San Juanico view deck gives tourists safer area to take selfies |url=https://www.pna.gov.ph/articles/1178743 |work=Philippine News Agency |language=en}}
The same goes for citations whose URL parameters start with |url=https://pia.gov.ph/
, please set the bot to use |work=Philippine Information Agency
instead of |work=PIA
, |agency=Philippine Information Agency
, or |publisher=Philippine Information Agency
. The citation should look like below:
{{cite news |title=DILG to continue “war on drugs” |url=https://pia.gov.ph/news/2022/07/05/dilg-to-continue-war-on-drugs |work=Philippine Information Agency}}
Thank you. Sanglahi86 (talk) 11:33, 12 July 2022 (UTC)
- Should all be done now. AManWithNoPlan (talk) 15:35, 14 July 2022 (UTC)
Adds weird volume
Caps: Medscape
Weird title
- Status
- new bug
- Reported by
- Medusahead (talk) 08:34, 18 July 2022 (UTC)
- What happens
- instead of inserting a correct title the bot put some longish text with typos ("festa" instead of "feast" which iMHO even doesn't appear in the given source as an alleged title of the source
- Relevant diffs/links
- https://en.wiki.x.io/w/index.php?title=Our_Lady_of_Mount_Carmel&type=revision&diff=1098349944&oldid=1098121150
- We can't proceed until
- Feedback from maintainers
Oddly it is actually the title listed on the website for these pages. Very sloppy work by the web editor. AManWithNoPlan (talk) 14:19, 18 July 2022 (UTC)
www.academia.edu/download/
- Discussion moved (back) to Wikipedia:Link_rot/URL_change_requests#www.academia.edu/download/
Remove invisible character
There's an invisible character between = and The, which the bot should remove. Headbomb {t · c · p · b} 15:46, 18 July 2022 (UTC)
Automatic cite magazine conversions
- Status
- new bug
- Reported by
- adamstom97 (talk) 03:35, 1 May 2022 (UTC)
- What happens
- Quite often this bot converts {{cite web}} into {{cite magazine}} just because the website that is being cited is associated with a magazine. I believe this is incorrect, {{cite magazine}} should only be used if an actual magazine with physical pages is being cited. I'm not sure in what circumstances a bot would be able to determine that.
- Relevant diffs/links
- I have been ignoring or reverting these changes for a long time so there are plenty of examples out there, here is one recent one: diff.
- We can't proceed until
- Feedback from maintainers
long discussion
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
BreakHoly jeez is this thing still going on. I'll transclude the CS1 documentation here.
Note in particular: {{cite web}}: [for] web sources not covered by the above. Covered by the above is {{cite magazine}}. EW is a magazine, so use cite magazine (and again, online magazines are magazines). End of story. Headbomb {t · c · p · b} 04:18, 28 May 2022 (UTC)
I'd like to point out as well, that though some editors are saying "online magazines = magazine", as far as I can tell, no where in the documentation of either {{Cite magazine}} or {{Cite web}} is this defined. All Cite magazine says on the matter for anything online is two instances of how to
RfC Sample Question workshopRfC Question: Citation Bot has a feature that automatically converts {{Cite web}} citations to {{Cite news}} and {{Cite magazine}}. For the purposes of this feature, are articles that are published exclusively on the websites of hybrid-print/digital publications considered to be published in a newspaper or magazine?
Alrighty, making this a bit more structured. InfiniteNexus and others, you've said that there are a number of edits that the bot has made that you've reverted. Could you please provide diffs to a selection of those edits, so that we can provide some actual examples of this in action? Sideswipe9th (talk) 16:44, 29 May 2022 (UTC)
The initial question for the RFC still needs to be brief, much like the one I proposed earlier today. And then give examples, and a "Background" section as Rlink2 said to cover all the information editors would need on the matter. - Favre1fan93 (talk) 02:14, 31 May 2022 (UTC)
Why do we need an RFC?In the section above, huge amounts of work are being put into drafting an RFC. If there is an RFC, lots more editor time will be put into the responses, and then into weighing a close. But why? What is the problem to be resolved? Some editors disagree with some instances of the conversion of {{cite web}} to {{cite magazine}} or {{cite news}}. They make a distinction between the publication and its website. I don't share that view, and think that any distinction is pointless ... but clearly, some want to maintain a distinction. But after all this debate, I don't see why they want to maintain a distinction, and why they object to {{cite magazine}} being used in e.g. some articles on the EW website. Sorry if I have missed some statement of this, but it would help to have a clear answer to the question: What is the harm done by these conversions? For example, does the change cause an unwanted alteration in the display of the citation? Do the objectors dislike the five extra character used by "cite magazine"? Or is this just about the name of the template? Or is it something else? Pinging @InfiniteNexus, who is one of the objectors. --BrownHairedGirl (talk) • (contribs) 20:27, 31 May 2022 (UTC)
Because more data is always good, I've pulled together a table of the same analysis for all of Citation bot's edits over the seven day period 6-12 June. Collapsing it because it's a big table, but I want to include it here so you all can check my working.
In summary, Citation bot made approximately 36,415 edits over this period. 6,784 edits contained "Alter: template type" in the edit summary. As before, "Alter: template type" only indicates a conversion from one CS1 template to another. 88 total edits were reverted over the period: 35 were to test accounts, 53 were to non test accounts, and 5 were part of this dispute. After doing this analysis, I'm very heavily leaning towards what BrownHairedGirl has said;
We're not making any progress if we're just sitting around and neither side is willing to concede. @Sideswipe9th: is the RfC moving forward or what? InfiniteNexus (talk) 21:13, 27 June 2022 (UTC)
Links to the RFCI hope that this proposed RFC will not happen, because I think it's a waste of time. However, if the RFC goes ahead, it would be helpful if while the RFC is open, Citation bot linked to the RFC in relevant edit summaries. For eaxmple, instead of "Alter: template type" in the edit summary, there could be "Alter: template type (see [[WP:somepage#CBALTERTEMPLATE|RFC]])". @AManWithNoPlan, would that be doable without too much work? --BrownHairedGirl (talk) • (contribs) 16:59, 16 June 2022 (UTC)
RfC Question second draft and formatIf an article is published on a website associated with a magazine, or newspaper, but does not appear on the print edition of the publication, should you use {{Cite web}}, or {{Cite magazine}}, or {{Cite news}}?
So format is based on the Pro and con example format for RfCs. The first sentence is the brief, neutral question per WP:RFCBRIEF. The table after is, or will be the summary of the past discussion covering the main points of each side of the discussion. While I still think we should include examples, I've collapsed them here because it kinda becomes unwieldy. I've also bold texted the differences between the two templates for each example. We could add an extra row for each example, to show what the wikitext output of the citations looks like, though it will be identical for both templates. Sideswipe9th (talk) 20:33, 1 June 2022 (UTC)
None of the pro or con reasons address whether the choice makes any visible difference at all to readers. If it does not make a visible difference, why are we arguing about the very important choice we must all make correctly or the world will end about how many angels can dance on the head of a pin? Just let the bot rationalize the citations and don't worry about legislating the way citations must be identically coded by everyone who uses them. Or, alternatively, argue by WP:COSMETICBOT that it should only change citation type when that will make a visible difference, rather than by how very angry you are that the bot is not letting you use the wrong citation type. —David Eppstein (talk) 05:06, 3 June 2022 (UTC)
I want to make a more fundamental challenge to this draft RFC, because it is framing the question to an unduly narrow perspective. The real question to ask is whether CB should replace a generic {{cite web}} with any of the more specific CS1 templates. If the argument is good for {{cite news}} and {{cite magazine}} then it is equally good for {{cite book}}, {{cite journal}}, {{cite press release}} – indeed any document that has a web presence. There is no reasonable basis on which to pick magazines as the hill to die on except to distract from the broad principle of whether specific templates are more useful than the generic one. --John Maynard Friedman (talk) 19:46, 13 June 2022 (UTC)
the first statement of the argument to use cite magazine or cite news is a fallacy. There is no way to confirm that articles on a website are also in print, a list of facts or top 100 sci fi films is unlikely to feature in print for instance. As we move ever onwards to a digital world and print continues the slow death initially prophesized by professor Spengler, how will cite magazine apply when the magazine no longer exists? Do we then go alter all those cite magazines? And how is it a valid use to cite a magazine when clicking the title will take the user to a website? It is an unreasonable statement arguing for the use of cite magazine and cite news. Cite web is not a generic template either, it's heavily used because we are not in 1982. There's a reason cite web is used nearly 5 million times and cite magazine less than 200K. Darkwarriorblake / SEXY ACTION TALK PAGE! 21:45, 28 June 2022 (UTC)
|