Wikipedia:Village pump (policy): Difference between revisions

Browse history interactively

← Previous edit Next edit →

Content deleted Content added

VisualWikitext

Inline

Revision as of 01:19, 3 November 2011

"WP:VP(P)" redirects here. For the proposals page, see Wikipedia:Village pump (proposals).

Table of contents
First discussion
End of page
New post

Shortcut

WP:VPP

The policy section of the village pump is used to discuss proposed policies and guidelines and changes to existing policies and guidelines.
If you want to propose something new that is not a policy or guideline, use the proposals section.
If you have a question about how to apply an existing policy or guideline, try the one of the many Wikipedia:Noticeboards.

Please see this FAQ page for a list of frequent proposals and the responses to them.

« Archives, 85, 86, 87, 88, 89, 90, 91, 92, 93, 94, 95, 96, 97, 98, 99, 100, 101, 102, 103, 104, 105, 106, 107, 108, 109, 110, 111, 112, 113, 114, 115, 116, 117, 118, 119, 120, 121, 122, 123, 124, 125, 126, 127, 128, 129, 130, 131, 132, 133, 134, 135, 136, 137, 138, 139, 140, 141, 142, 143, 144, 145, 146, 147, 148, 149, 150, 151, 152, 153, 154, 155, 156, 157, 158, 159, 160, 161, 162, 163, 164, 165, 166, 167, 168, 169, 170, 171, 172, 173, 174, 175, 176, 177, 178, 179, 180, 181, 182, 183, 184, 185, 186, 187, 188, 189, 190, 191, 192, 193, 194, 195

RFC on the bot-addition of identifier links to citations

NO CONSENSUS

I don't see enough support for a bot doing this, further more, the community outlines below that the process that things would occur is broken. Lets try fixing the process to consensus and then moving forward on looking into a bot. -- DQ (t) (e) 08:53, 23 October 2011 (UTC)[reply]

The following discussion is closed. Please do not modify it. Subsequent comments should be made on the appropriate discussion page. No further edits should be made to this discussion.

There's been some kerfuffle about bots adding of identifier links to citations recently. Specifically, whether links to a topical database (aka things like arXiv preprints, Bibcode links to the Astrophysics Data System, Mathematical Reviews, PMC or PMID links to PubMed, SSRN, Zentralblatt MATH, etc...) should be added regardless of the topic of the Wikipedia article, or if bots should only add "topic-neutral databases" links (aka doi, JSTOR, ISBN, etc.), unless the bot can guarantee that the identifier links added are "topical".

Example with all identifiers
Field	Citation
Astronomy	J. Cami; et al. (2010). "Detection of C₆₀ and C₇₀ in a Young Planetary Nebula". Science. 329 (5996): 1180. Bibcode:2010Sci...329.1180C. doi:10.1126/science.1192035. PMID 20651118. `{{cite journal}}`: Explicit use of et al. in: `\|author=` (help)
Chemistry	N. Sano; et al. (2001). "Synthesis of carbon 'onions' in water". Nature. 414 (6863): 506. Bibcode:2001Natur.414..506S. doi:10.1038/35107141. PMID 11734841. `{{cite journal}}`: Explicit use of et al. in: `\|author=` (help)
Mathematics	G.L. Cohen (1990). "On an integers' infinitary divisors". Mathematics of Computation. 54 (189): 395–411. Bibcode:1990MaCom..54..395C. doi:10.1090/S0025-5718-1990-0993927-5. MR 0993927.
Medicine	F. Barré-Sinoussi; et al. (1983). "Isolation of a T-lymphotropic retrovirus from a patient at risk for acquired immune deficiency syndrome (AIDS)". Science. 220 (4599): 868–871. Bibcode:1983Sci...220..868B. doi:10.1126/science.6189183. PMID 6189183. `{{cite journal}}`: Explicit use of et al. in: `\|author=` (help)
Physics	F. Abe et al. (CDF Collaboration) (1995). "Observation of top quark production in pp collisions with the Collider Detector at Fermilab". Physical Review Letters. 74 (14): 2626–2631. arXiv:hep-ex/9503002. Bibcode:1995PhRvL..74.2626A. doi:10.1103/PhysRevLett.74.2626. PMID 10057978.

Each database cover different things differently, some contain "citation coverage" (aka "who cites this journal article"), and complement each other (citation counts will differ, and listings will not mention the same citing journal articles, as e.g. the bibcode link will cover "citations in physics & astronomy" journals, while e.g. the PMID will cover "citations in the medical field"), some might contain links to preprints, links to free digitized versions, etc... One could certainly argue that these identifier links do a great deal to established the reliability and verifiability of a citation and the Wikipedia article it is supporting in general.

There's also a great deal to be said about letting people choose which database they prefer. If someone familiar with mathematics databases runs across a Wikipedia article on astronomy, and one of the references is indexed in both the astronomy and mathematics databases, why should they be forced to use the astronomy database if they would rather use the mathematics database (despite the astronomy database link being almost certainly better). If someone familiar with medicine journals stumble across a mathematics citation which is indexed in both mathematics and medical database, why should they be forced to use the mathematics database if they would rather see what PubMed has to say about it? If someone from a physics background runs across a medicine citation, why should they be forced to use the medical database if they would prefer using something they are familiar with?

On the other hand, other people feel these identifier links do little more than clutter the citations and confuse the reader, and should be omitted (or at the least should not be added by bots) unless the identifier link can be guaranteed to be on a database that matches the topic of the the Wikipedia article, or that it requires human judgment to decide whether or not an arxiv/bibcode/PMC/PMID/MR/SSRN/Zbl link should be added to the citation. This would mean that in the relevant Wikipedia articles, you would see something like

Example with only "topical" identifiers
Field	Citation
Astronomy	J. Cami; et al. (2010). "Detection of C₆₀ and C₇₀ in a Young Planetary Nebula". Science. 329 (5996): 1180. Bibcode:2010Sci...329.1180C. doi:10.1126/science.1192035. `{{cite journal}}`: Explicit use of et al. in: `\|author=` (help)
Chemistry	N. Sano; et al. (2001). "Synthesis of carbon 'onions' in water". Nature. 414 (6863): 506. doi:10.1038/35107141. `{{cite journal}}`: Explicit use of et al. in: `\|author=` (help)
Mathematics	G.L. Cohen (1990). "On an integers' infinitary divisors". Mathematics of Computation. 54 (189): 395–411. doi:10.1090/S0025-5718-1990-0993927-5. MR 0993927.
Medicine	F. Barré-Sinoussi; et al. (1983). "Isolation of a T-lymphotropic retrovirus from a patient at risk for acquired immune deficiency syndrome (AIDS)". Science. 220 (4599): 868–871. doi:10.1126/science.6189183. PMID 6189183. `{{cite journal}}`: Explicit use of et al. in: `\|author=` (help)
Physics	F. Abe et al. (CDF Collaboration) (1995). "Observation of top quark production in pp collisions with the Collider Detector at Fermilab". Physical Review Letters. 74 (14): 2626–2631. arXiv:hep-ex/9503002. doi:10.1103/PhysRevLett.74.2626.

This would have consequences for bots such as Citation Bot (which could not add links to arXiv preprints, bibcodes, PMC links, or PMIDs automatically, like it's been doing for the past few years), Bibcode Bot (which would be restricted to astronomy & physics articles), and any future bot such as the hypothetical "SSRN-Bot" or "Mathematical Review-Bot" (which would be restricted to their topics), or users who run scripts to add identifiers to articles such as Rjwilmsi.

Headbomb {talk / contribs / physics / books} 15:45, 15 August 2011 (UTC)[reply]

Comments

Isn't a DOI enough? The links all point to the same journal article; the only difference seems to be some extra details about the citation counts and whatnot. We're not concerned about that as an encyclopedia, we're just citing the article. I mean, it's not even a big deal, so it's sort of depressing that people have been arguing about this, but to me it seems simplest if we use one standard link (e.g., DOI) as anyone who really cares about the other databases will know how to look for the same article on them. /ƒETCH COMMS / 16:03, 15 August 2011 (UTC)[reply]
There are two main things a reader wants from a reference: (1) access to the abstract, (2) access to the full text. I am opposed to bots adding any links to references unless the bot has the intelligence to determine that (a) one of those two things is missing, and (b) the added link provides the thing that is missing. I do not believe that adding clutter to reflists is harmless: it makes them harder for ordinary readers to use. I have no objection to human editors adding whatever information they feel is helpful; my issue is only to having this done on a massive scale by bots. Looie496 (talk) 16:05, 15 August 2011 (UTC)[reply]
I strongly support adding Bibcodes and Arxiv ids to any and all references which are included in those databases. ADS is an excellent free resource which massively improves the chances of a reader finding a copy of the full text that they can access. Arxiv papers are not the 'final' version, but are always available for free to everyone. This is totally different to the DOI, which in almost all cases redirects to a journal website with an extremely expensive paywall. I'm less familiar with the other identifiers mentioned above, but see little harm in including them. There obviously has to be a line drawn somewhere (no point in have 27 different ids on every reference), but even the most extreme example given above is fine. Modest Genius ^talk 16:16, 15 August 2011 (UTC)[reply]

Oh and I should point out that both ADS and Arxiv cover a huge range of subjects, not just their 'traditional' strengths (in astronomy and particle physics respectively). Modest Genius ^talk 16:21, 15 August 2011 (UTC)[reply]

This RfC should've been presented without the pro-bibcode etc. rhetoric. Anyway, PMID/PMC and doi are all I would want/need to see in any bibliography. There's no need to give excessive alternatives; even without any identifier an average user could find any paper in seconds, but a doi and/or PMID provides handy one-click access without bloating the bibliography Jebus989 ^✰ 16:18, 15 August 2011 (UTC)[reply]

Even on mathematics articles? And physics articles? Or social science articles? That's awfully self-centered of you. Headbomb {talk / contribs / physics / books} 16:20, 15 August 2011 (UTC)[reply]

I don't think there is any 'pro-bibcode rhetoric' in the above. It does slant in favour of more IDs, but barely mentions Bibcodes more than any of the other IDs discussed. Modest Genius ^talk 16:25, 15 August 2011 (UTC)[reply]

You misquoted me, "pro-bibcode etc." was what I said. I mean the RfC is written primarily as an argument in support of adding a wide variety of accessions, and I hope it doesn't take me picking out sentences for you to realise that. As a non-mathematician, I know of no pubmed equivalent, but if I were to read such an article, and found only the doi, I would definitely try work up the gumption to click it, rather than perusing through 5 alternative links to the same article and choosing my favourite. I'll ignore the personal attack, but I remind you this is a request for comment, not an "agree with me or I will argue with you until you do" Jebus989 ^✰ 19:14, 15 August 2011 (UTC)[reply]

For reference, I'm for the addition of any and all links on any and all articles. I've edited physics & astronomy articles for as long as I could remember and the various PMIDs and MRs links have never bothered me. In fact I found them to be extremely useful in fixing citations or verifying that the references did support the text (or that they were reliable). It would be utterly catastrophic for Wikipedia to disallow bots to add these links by default. Medicine people would lose Citation bot's ability to add the PMID/PMC automatically. Astronomy (and related topics) people would lose Citation bot's ability to add bibcodes. Physics (and related topics) would lose Citation bot's ability to add arxiv preprints. Everyone lose, no one wins.
For anyone these links truly bother, that could easily be "fixed" with a skin tweak (monobook.js/vector.js). Headbomb {talk / contribs / physics / books} 16:30, 15 August 2011 (UTC)[reply]
Putting every single identifier known to man (as well is wikilinking what they are) does seem to produce a vast, confusing sea of messy links which will be meaningless (and possibly distracting) to the average non-academic reader. Idealy there would be a way of hiding these away like we do with ISBN numbers and special:booksources -where clicking on one link would bring up all appropriate identifiers.Nigel Ish (talk) 17:01, 15 August 2011 (UTC)[reply]
I appreciate seeing the two versions of the references side-by-side, which is quite helpful. I think that the default state should be the shorter, topic-specific format. The other way does, indeed, look overly cluttered. I can imagine a reader from the general reading public, not someone who is an aficionado of databases, but just someone wanting to read up on a subject, looking at the lengthier version and having a case of "too much information". I recommend making the topic-specific format the default, with the proviso that any editor may always add more manually, and any WikiProject, by consensus, may request that a bot add more to pages within their project. I also like Nigel Ish's idea of having more but hiding them. --Tryptofish (talk) 17:09, 15 August 2011 (UTC)[reply]
Interesting issue. I agree with Looie496 on what users want and that in the short run humans should make these decisions rather than have a proliferation of alternatives added by bots which 99% of readers wouldn't use. Yet the added content is definitely useful. Perhaps in the short run the links at the end of the biblio entry could be compacted, showing just (ArXiv)(PMID)(doi)(JSTOR), each linked to the appropriate source for that biblio entry. This avoids showing the full details of the code to every reader, and sticks to showing human-useful/readable content. Those details could be in HTML comments. In the long run I think this problem should not be solved right in the article, where it takes up vertical space and mind space, perhaps increasingly over time as the number of such outside sites grows. Instead it would be good if all those links the bot would have offered can be offered on the doi page or some wikimedia-specific intermediate page made available when the user clicks on the biblio entry. Might be created on the fly, or static. If static, the bots could add lavishly to that intermediate page and the bibliography user-interface would be uncluttered and easy to use. Not trivial to implement unfortunately. Wikisource might be friendly to it but their agenda is not directing-to-sources but rather offering the sources. (I see that Nigel Ish suggested this too during an edit-conflict. Hot topic!) -- Econterms (talk) 17:12, 15 August 2011 (UTC)[reply]
Let me be the third person to endorse Nigel Ish's observation that ideally the links should be there (because they really are useful to someone who knows what they mean) but hidden (because they are confusing clutter to people who don't know what they mean). --Joel B. Lewis (talk) 17:32, 15 August 2011 (UTC)[reply]
i support adding as many identifiers as exist. different people will have access to different databases. there might be some reasonable limits similar to books that make sense here also that might be simple enough to get broad support, but i'd rather have more identifier links that are possibly redundant than restrictive rules. generally for books with isbn, oclc and asin are not also listed. pmc and arxiv are really good and should always be listed if available. doi usually doesn't go to a full free-access copy, but many people have institutional access. jstor only shows the first page. pmid and bibcode show bibliographic info with an abstract. this is the first time i've seen the mr database and it doesn't look all that useful from this example. it looks like a link to a doi that goes to jstor. —Chris Capoccia ^T⁄_C 17:36, 15 August 2011 (UTC)[reply]
I strongly oppose bots automatically adding any information to Wikipedia articles. Bots can never replace a living person. Bots are very useful tools that editors can use to research information (including citation information). However, a thinking person needs to review the bot results before the information is actually placed into an article. Even something as simple as a spell or grammar checker bot can cause all sorts of unexpected errors and problems. Blueboar (talk) 17:46, 15 August 2011 (UTC)[reply]
- WP:BOTPOL obviously disagrees with you here. Headbomb {talk / contribs / physics / books} 18:05, 15 August 2011 (UTC)[reply]
- So you want to ban all interwiki bots? Funny idea. Bulwersator (talk) 19:46, 15 August 2011 (UTC)[reply]
  - Often enough they amplify problems rather than resolve them; see meta:IS. Have mörser, will travel (talk) 11:25, 11 October 2011 (UTC)[reply]
I think that Errant's comment on ANI and Nigel Ish's comment here have got to the bottom of this: we should focus on working out how we can configure a default display of the citation that is not crowded, while also allowing the option of displaying the full identifier information. That ought to give the simplified display desired by casual readers and (maybe at one click) the full information and options desired by more knowledgeable readers or researchers. I think this RFC should be to discuss how to display the information, not on whether it should be there in the first place. Rjwilmsi 17:53, 15 August 2011 (UTC)[reply]
Support adding a variety of identifiers/sources, but I want a way to reduce the "clutter". Since that seems to be a dominant issue here, something like this would be sufficient for me (trade-off seeing actual identifier values for space and broader choice):

F. Abe et al. (CDF Collaboration) (1995). "Observation of top quark production in pp collisions with the Collider Detector at Fermilab". Physical Review Letters 74 (14): 2626–2631. arXiv:hep-ex/9503002. Bibcode 1995PhRvL..74.2626A. doi:10.1103/PhysRevLett.74.2626. PMID 10057978.

becoming

F. Abe et al. (CDF Collaboration) (1995). "Observation of top quark production in pp collisions with the Collider Detector at Fermilab". Physical Review Letters 74 (14): 2626–2631. arXiv. Bibcode. doi. PMID.

I also understand a certain identifier may be more useful than others based on the topic/journal, so something like |primaryidentifier=doi could be employed:

F. Abe et al. (CDF Collaboration) (1995). "Observation of top quark production in pp collisions with the Collider Detector at Fermilab". Physical Review Letters 74 (14): 2626–2631. doi:10.1103/PhysRevLett.74.2626. arXiv. Bibcode. PMID.

I realize it's impossible to please everyone, but I also don't want "clutter" to come in the way of presenting our readers with a wide selection of database links. — HELLKNOWZ ▎TALK 18:06, 15 August 2011 (UTC)[reply]

As a note to the "let's tweak the information", that's fine and all, but you have to remember that Wikipedia should be for all people. "Compact links" such as [ [arXiv] [bibcode] [doi] [ISBN] [JSTOR] [pmid] ] will be horrible for people with screenreaders, and will be horrible when the page is printed. The only way to "tweak" it without affecting accessibility and print versions is to do it via skin tweaks (monobook.js/vector.js/etc...).Headbomb {talk / contribs / physics / books} 18:07, 15 August 2011 (UTC)[reply]
We should consider not only what readers see, but also what editors see. When the citation becomes of excessive size, the article becomes difficult to edit. The idea solution would be a link to a sort of automated source ombudsman, where a reader could specify which libraries are nearby and/or which databases the reader has paid access to, and the automated ombudsman would display which sources provide the full text at no incremental cost to that reader. Jc3s5h (talk) 18:12, 15 August 2011 (UTC)[reply]
- That's impossible to do, as there is no "one true identifier" for everything. DOI is the most popular, but not all databases are queryable via DOI. Headbomb {talk / contribs / physics / books} 18:15, 15 August 2011 (UTC)[reply]
Given that different users have different databases they can access, it seems obvious that "anyone can edit" implies that whichever such resources the user has available should be linked. For the fortunate few with institutional access to such databases that does not always extend to all databases or even all of a particular database. My access to JSTOR, for example, is only fulltext for a portion of the serials it contains. Providing diversity of linkages gives editors and readers the best possible chance of finding the source in a repository which is freely accessible (for them). When we choose to omit these links we effectively inhibit people who could otherwise read the source from doing so. That cannot be a constructive practice. A little blue on the screen is a small price to pay. I have no objection to hiding it in hovertext or some such technical approach so long as the linkage is easily available to users that want it. The issue of trusting humans more than bots to make the call misses the fact that many of our human editors are very weak at citations: we're still fighting naked urls! While it should be simple for a human editor to remove links that don't work, and bots should respect such decisions, humans should not delete them simply because a link didn't work for them. Wikipedia:WikiProject Resource Exchange/Resource Request‎ shows us everyday the utility of pooling access. LeadSongDog come howl! 18:14, 15 August 2011 (UTC)[reply]
Comment on fancy display technology: this is a red herring and can't help us. Wikipedia pages need to work also as PDFs (onscreen with simple hyperlinks or printed out on paper); on a variety of mobile devices which may not support the concept of "hover"; on browsers where JavaScript is disabled, with screen reader technology, and where the Wikipedia page is being mirrored by another website. Any option that requires an account (such as a user-specific skin tweak) is also a non-starter as nearly all of our readers are not logged in. It is an important attribute that one can select the display text and copy/paste it: something that popup bits would make impossible. Colin°^Talk 18:54, 15 August 2011 (UTC)[reply]
- If JavaScript is unavailable we could display the minimum, i.e. the doi. If it is available, the citation can be expanded on mouse-over to include other potentially useful links. Such a strategy could work well on PDFs and mobile devices where users probably don't want clutter taking up space. eug (talk) 13:12, 4 October 2011 (UTC)[reply]
Comment—Perhaps we could wrap the identifiers with an inline expandable display? See here for example. This way the text would normally be hidden unless somebody wants to look up the reference. Regards, RJH (talk) 19:19, 15 August 2011 (UTC)[reply]
- Don't know if that could be done, but the default should be the expanded start, otherwise this would create a drastic clash between "manual" citations and template citations. Headbomb {talk / contribs / physics / books} 19:41, 15 August 2011 (UTC)[reply]
  - Leaving it default expanded would make it pointless to implement; no viewer is ever going to go through the list and contract the views. But "manual" citations could probably implement this using a separate template for inline identifiers. RJH (talk) 20:37, 15 August 2011 (UTC)[reply]
    - Couldn't be done with manual citations, as each identifier template would "condensed" individually, rather than globally. You could place them in a wrapper, but that is an extremely convoluted option. Headbomb {talk / contribs / physics / books} 20:40, 15 August 2011 (UTC)[reply]
      - I see such a wrapper template for use in a manual citations as a variant of the citation template, sans the non-identifier information. Shrug, no matter. RJH (talk)
I oppose bots adding these links to articles until the clutter is reduced by appropriate template magic/user preferences/CSS. Once such a solution is implemented, I will change by !vote to support. CRGreathouse (t | c) 19:21, 15 August 2011 (UTC)[reply]
Something similar to what RJHall and Nigel Ish said would be the best course to take here: give all citation databases and let the reader pick which one to use. Edge cases such as the ones Colin identified can still see the current format. That said, getting to the crux of the issue, I don't see a point in removing existing bibcodes/PMIDs because of a few editors' perceptions of the utility of the citation databases. Give the readers more credit. Titoxd^{(?!? - cool stuff)} 19:33, 15 August 2011 (UTC)[reply]
If we do go down the 'keep them but reduce the clutter' route, could it not be hidden behind a 'show/hide' bit of Javascript, like collapsed navigation boxes? That way it still shows up when printed / on screen readers, but reduces the clutter for everyone else. Modest Genius ^talk 19:34, 15 August 2011 (UTC)[reply]
Show/hide boxes work exactly the same on modern screen readers as they do for other people. I would favour keeping the links expanded as they are now, if only because we shouldn't add more bloated JavaScript to pages if we can help it; the citation links don't cause any problems for screen readers. I don't have any strong opinions about which databases should be used. Graham87 00:30, 16 August 2011 (UTC)[reply]
Support - Information is good. —SW— ^express 20:25, 15 August 2011 (UTC)[reply]
Certain of these identifier can be assumed to be topical: (1) arXiv: if an article is on the arXiv, then the author put it there, (2) MR (aka Mathematical Reviews) & Zentralblatt: only lists math articles, so for any article that has an MR, the MR should be topical. I don't know enough about the other identifiers to say, but I don't think that non-topical identifiers should be automatically listed, except for ISBN and DOI (ISSN, too maybe?); by which I mean, I think it would be better to keep the displayed identifiers restricted to topical ones, ISBN, and DOI, but I'm all for adding other ones to the source code, but have them commented out, or somesuch. Additionally, at least the two identifiers I have listed provide useful information that I believe should be linked to from the citation on wikipedia, (and if a bot can do it instead of me, I'm all for that). For the arXiv, this includes a free copy of the paper (though they are sometimes not the same version as the print copy), and for MR & Zbl, this (almost always) includes a summary of the paper, as well as a link to the papers that cite the paper in question. RobHar (talk) 20:30, 15 August 2011 (UTC)[reply]
The only requirement we have for citations is that they identify the source of the text used to write the article text. A plain text standard printed citation or a raw URL both satisfy this requirement, though each can be improved upon. We are careful not to add external links to articles just because someone thinks they are useful and similarly we should be careful not to turn our References section into an external link farm. We are an encyclopaedia, not a compendium of journal database links. Perhaps folk should put pressure on the various journal database websites to do some cross referencing themselves. It shouldn't be too hard for PubMed to link to ADS, say. After all, indexing journals is their job, not ours. As Looie496 says, the aim of our convenience links should be to give the reader access to the full text and the abstract. Access to the full text is complicated by the fact that nearly all our readers will lack the necessary subscriptions for subscription-only texts, or the text may not be online at the publisher's website. So services like PMC and arXiv provide a backup for the full text. The databases themselves provide useful functionality but this is very much secondary to our purpose. There is a strong consensus for linking to PubMed for bio-medical papers and I dare say the same goes for astronomy/physics papers and the ADS. If the paper is only indexed in one of the databases, then there is probably merit in linking without considering the topic of the paper.
The bot should only perform actions where there is a clear consensus and where the edit is useful. So unless we come to a consensus that linking to PubMed and linking to ADS is desirable always, then the bot needs to work out what kind of paper the citation is for. A bot could use the journal to decide the topic in most cases. For journals like Science and Nature, that cross the fields, then the WP article may give a clue as to whether the paper is a life or physical science topic. I think using only the WP article as a guide is rather crude and only reliable for a subset. So, in summary, PMC/arXiv are probably always useful unless the bot can tell that a link to a free online full text is already present. DOI is always useful. Where both a PMID and bibcode are possible, the bot should consider the journal and possibly additionally the article topic and pick the most appropriate -- and if unsure then don't add. -- Colin°^Talk 20:37, 15 August 2011 (UTC)[reply]
I'd like to see PMID links and ISBNs whenever possible, even for non-medical articles (which might, after all, contain some medicine-related information). I don't care about the others, but I suspect that if I always want to see links to foo, then someone else will always want to see bar, so I'm in favor of listing everything as the default. There should, however, be some sort of opt-out system, like an invisible template that editors can place in an article to say "Bibcodes (or whatever) not wanted in this article, thanks". WhatamIdoing (talk) 20:45, 15 August 2011 (UTC)[reply]
Support If you can get a source for free then its a no brainer to add the (legal) links --Guerillero | My Talk 21:32, 15 August 2011 (UTC)[reply]

Support. More pointers is better than none or a few useless ones, which can still be deleted manually if they are a splinter in an editor's mind, right? Peter S Strempel | Talk 22:55, 15 August 2011 (UTC)[reply]
Oppose Bots should not be doing controversial edits. The important stuff here is the citation; once that is given, all other links are convenience. Adding a lot of them makes the citation itself, which can be used in whatever tool the reader has available, harder to find; if, as some arguments here would suggest, we add links to every conceivable citation system, our articles will be buried in kilobytes of linkcruft. Septentrionalis PMAnderson 01:53, 16 August 2011 (UTC)[reply]

I generally support some modest quantity of redundant IDs, because they sometimes can be helpful in tracking down an article (for example when a DOI stops resolving, which is a moderately frequent occurrence). I guess I'm a weak oppose on the subject of having bots add them, however. Having a bot add a PMC has rarely seemed like an intrusion (even if there is already a free link of some kind), but kerfuffles of this sort seem to be par for the course when bots are involved and I'm not sure how a bot would know how many IDs is enough. Perhaps there is a middle ground, like an approved list of IDs which are bot-addable. But if we can't resolve it that way, falling back to human editors doesn't strike me as a horrifying thought. Kingdon (talk) 02:24, 16 August 2011 (UTC)[reply]

I support adding additional info to citations. As others have pointed out, these are useful for readers who have access to different databases. And many are topical, such as the Math Reviews number or the PubMed number, and will only be available for articles in a certain discipline. — Carl (CBM · talk) 02:33, 16 August 2011 (UTC)[reply]
Support - Anything that helps our readers get to the original sources used for our articles is a benefit to all. - Hydroxonium (T•C•V) 03:05, 16 August 2011 (UTC)[reply]
Comment: The proposed solution does not scale in the long term. I would suggest a solution similar to that for either Books or Map Coordinates; in each case a link leads to a master list where the identifier is auto-integrated with resource links. However, the central-linking solution would need to be refined somewhat for the present solution, as the consistency of ISBN or lat-lon is not present for scholarly articles yet (DOI approaches, but has not reached 100% penetrance ... anyone have stats on that?). What might need to be done is a two step process - create links via bot on a central resource page, then review links via bot for ability to resolve to a target, followed by either link culling or link annotation (verified, unverified, unavailable type flags). I've not read through the comments, so I don't know whether this is a new idea or just a rehash of an old one oft added above. Thanks for considering this, nonetheless. --User:Ceyockey (talk to me) 03:22, 16 August 2011 (UTC)[reply]
Fully, strongly support. Note, I am not a Wikipedia:WikiProject Medicine person, nor a Wikipedia:WikiProject Astronomy person (actually, I am a Wikipedia:WikiProject Chemistry person).

1) PMIDs are useful, they are available for a lot of articles, they do provide extra information over a DOI (DOIs directly link to the article, generally the search capabilities of the site where that is hosted is limited to the site of the journal - the page on PubMed is generally cross-site). And that is not only true for articles in the medical corner, it is not only true for statements with a medical implication on articles outside the medical corner. When a PMID is available, it should be added, it does lead you to more info.

1 (too)) BibCodes are equally useful, they are available for another set of articles, they do provide extra information over a DOI (vide supra, comment on usefulness of PMIDs). That is not only true for the astronomy corner, it is not only true for statements with an astronomy implication on articles outside the medical corner. When a BibCode is available, it should be added, it does lead you to more info.

2) the number of journals that is overlapping is quite minimal, they are mainly the general science journals (Nature, Science and such), those articles benefit of identifiers all over Wikipedia. I would be disappointed if CitationBot would not add a PMID to an article in chemistry (as it does already), I would be equally disappointed if Bibcode bot would not add a BibCode to an article in chemistry. Or about whichever subject.

3) it is not up to us to decide whether an identifier is useful for others. I can bring up many examples where a PMID is useless clutter for me, I can bring up just as many examples where a BibCode is useless clutter for me. I can even find examples where both are useless clutter for me. The problem in both statements is 'for me'. Sure, I can guess that in most of these cases will also be true for most other readers of the text, but then we get to 'most' - it may be useful for someone, and that should be enough. (Yes, in most cases, I am interested in the article, not in what links to the article, which articles are cited by the article and how often those articles are cited, or articles that cite the article, who wrote it, where it is published, a direct link to the article will do, DOI is enough, per Fetchcomms, thank you).

4) I am active in subject A, and for references we have a database linked to subject B and a database linked to subject C. Now, subject B nor subject C are topical for subject A. But both the database of subject B and the database of subject A do give more info than the DOI only would, they both provide extra info, extra search capabilities. Unbiased addition of both is then leading to more info, as obviously no choice can be made whether B or C will be better.

5) Regarding 'clutter' - these identifiers are in references, not in prose. Reading references is like reading a telephone book: ~~no-one~~hardly anybody does it for fun. When reading a reference you already have to go through a whole set of 'clutter', journal codecs, year of publishing, volume, issue, pagenumbers, &c. &c. (and all presented in different ways .. per convention in the local subject). One extra code does not 'clutter up the reference'. When you are used to code XXXX:1234, then when code YYYY is not there, you scan for the 'XXXX:' and click on the code next to it, when code YYYY:4321, ZZZZ:5678, CCCC:9876 and PPPP:4578 are there, you scan for the 'XXXX:' and click on the code next to it. It does not add clutter, at all.

6) Now, say, we have a medical article with a medical statement with a medical reference with, obviously a PMID. But that medical article turns out to be also in a astronomy database, and the astronomy database gets added as identifier. When that is not a general journal (Nature, Science) but something specific, then I might wonder 'why is this article in an astronomy database, the article does not have any astronomical content at all?' .. it may turn out, that some info from the article is missing, since there actually turns out to be an astronomical side to the content. A great incentive to look further into that aspect.

7) (unlikely scenario) - say database XXXX goes down (temporarily, or is locally blocked behind a firewall, something I could imagine in places/countries where information is restricted), then you have at least access via database YYYY. If XXXX and YYYY provide equal enhanced information over ZZZZ, then even if the article is in the subject of XXXX, then YYYY may be useful when XXXX for some reason goes down (or whatever).

All in all, I see no reason to not add all identifiers everywhere. They may be useful to some, and that should be enough. --Dirk Beetstra ^{T C} 08:04, 16 August 2011 (UTC)[reply]

P.S. If you really think that the references contain too much clutter, then I am sure that it is possible to put something in your user stylesheet or javascript that reduces references to just the info you want. E.g. which filters out BibCodes from the parsed text, or another identifier, or even reduces a full-text reference to just 'XXXX' when identifier XXXX is available in that reference. --Dirk Beetstra ^{T C} 08:11, 16 August 2011 (UTC)[reply]

P.P.S. I know that cross-namespace links are being frowned upon. But what if citationbot does something smart with the transcluded templates (the {{cite DOI/123456}}-type references)? --Dirk Beetstra ^{T C} 08:02, 18 August 2011 (UTC)[reply]

P.P.S.-Note: See #'Fancy display tech'. --Dirk Beetstra ^{T C} 08:24, 18 August 2011 (UTC)[reply]

I strongly oppose
- cluttering references with anything beyond a doi; once there is a doi, users can trivially retrieve whatever bibliographic information in whatever database;
- automatizing controversial edits;
- doing so while there is an ongoing discussion. -- Marie Poise (talk) 14:25, 16 August 2011 (UTC)[reply]

Trivial? Really? I for one have no idea how to find an arXiv preprint just from the DOI. Similarly for ADS. And I use both of those databases every day. And remember that getting any information whatsoever out of a DOI depends upon the journal website providing it (without being hidden behind a paywall). Journals generally do NOT link to citation databases for the simple reason that it reduces the chances of anyone actually paying for the article. Modest Genius ^talk 17:50, 16 August 2011 (UTC)[reply]

I strongly oppose bot-induced clutter of footnotes. The advocates of this mechanized mess have their opinions, but it is clear from the comments above that there is dissent and they should NOT be assuming consensus here. Carrite (talk) 15:29, 16 August 2011 (UTC)[reply]
Comment not all links have similar value to all readers - so I would oppose limiting links to only one standard (unless links to the others are directly available through that one standard). For example, PMIDs are the most valuable to me, because my library affiliation provides full-text access for a huge range of pubs when linking via PubMed - any other link is far less useful. I would imagine similar considerations apply for other users (and database links other than PubMed - so I'm not simply arguing for PMIDs). If it's possible to make this a reader-level customization for display then that would be great (but might be a resource hog); absent such customization, I would favor inclusive linking. -- Scray (talk) 18:06, 16 August 2011 (UTC)[reply]
Strongly support adding links - Not all educational institutes buy access to every single database. Thus I think we should add links to every database the journal's article appears in to help ensure that people reading the Wikipedia article will be able to find the full text without having to go look up the article in the database they have access to. Reaper Eternal (talk) 18:27, 16 August 2011 (UTC)[reply]
Comment How about DOI plus a single link to a page that can have as many links as you want, like Special:BookSources? --Kkmurray (talk) 19:37, 16 August 2011 (UTC)[reply]
Comment Although I support adding more information to citations, I am troubled by some automated edits. An article gets an automatic edit, it tickles many watch lists, and then many people check the edit. The bots also seem to add a lot of text to the reference that many users will ignore. Bots may pull a citation away from the intended citation to a different one; each fixing something that it perceives is wrong, and the next bot "improving" the citation further. A different approach might be better. The cite/citation templates could have a check for sources link; that link would include some unique information, but it would go to a wikipedia page that would run a (possibly precomputed) query for other sources. The bots could work on fixing the info in the second database without disturbing the article page. They could flag or correct basic info in the article (e.g., adding dates, authors, etc.), but they would not be edit the article merely to include another opaque identifier for some data source. If an article starts getting a lot of click throughs for a source, then the article could be edited to include a direct link for that source. Updates could be limited to so many per month per article. Glrx (talk) 00:46, 17 August 2011 (UTC)[reply]
Too many identifiers. Bots should only add topical ids, like pmid for medical articles, arxiv for physics, etc. A bot shouldn't be cluttering refs with multiple redundants ids. I would support a less ambitious bot that added only a minimal number of ids. I also support bots that add ids to references where there is no id, that tag non-existing ids, that tag possibly wrong ids, etc. --Enric Naval (talk) 10:58, 17 August 2011 (UTC)[reply]
Comment That different editors have different preferred citation display styles is perfectly reasonable. The current citation setup holds the raw data to support it, but doesn't offer much flexibility in rendering. There's been a lot of debate trying to get display preferences out by changing the raw data going in, but that solution can't work. The way forward must be to implement editor display preferences to differentiate between the source fields available and the fields shown to the editor in the rendered page. Of course is must be compatible with screen readers, mobile browsers, print versions, usability etc. and not break existing data, but we have some very knowledgeable technical editors and template writers, I'm sure a neat and effective solution can be found. I really think we must focus our effort on collaborating on neat and sensible editor display preferences. Rjwilmsi 22:16, 17 August 2011 (UTC)[reply]

I would agree with that. No real reason why citations couldn't be displayed according to certain preferences. Headbomb {talk / contribs / physics / books} 23:35, 17 August 2011 (UTC)[reply]

See my comment above about fancy display tech. Not going to work. Does anyone know the ratio of readers to editors? It is high enough that our individual preferences are pretty irrelevant, certainly so for something unconnected with editing. I bet the "let's fix this with editor preferences" solution is probably one of those perenial suggestions that keep getting shot down. Remember date formatting as an editor preference. That turned out well. Not. Colin°^Talk 07:20, 18 August 2011 (UTC) Plus this suggestion does smell a bit of "Once I can make this problem go away for me then I'll be happy." The real issue is that we've gone away from thinking about what the core requirements of references in an encyclopaedia are, and have started adding "useful" bells and whistles. Our sources don't add these extras to their references sections. And just like our WP:EL policy, we need to draw a line and consider what is essential to our purpose. Colin°^Talk 07:38, 18 August 2011 (UTC)[reply]
The 'fancy display tech' does not necessarily mean that it should only be restricted to logged in users (there is also global CSS and JS ..). And, using CSS and JS is only one way (but indeed, suboptimal), but it may also be possible in other ways (/me puts on thinking cap again). The two main opposing concerns I think there are is that it is useful for certain people (seen that PMIDs are everywhere, and not only added by Citationbot, also by human editors), but that it does clutter our references (which is for another group of editors clearly a big problem). Removing the extra identifiers leaves out info which is used but removes the clutter, having them there clutters, but gives us extra information. Say for yourself, you use the PMID link in the articles where you have them - I do as well, but my 'WP:MED-activity' is just on a thin overlapping border. I would miss them, badly. Removing them all would also mean to remove them from WP:MED, or allowing them in WP:MED means that other projects would also have the possibility to allow them .. and that would just continue the controversy, inevitably WP:MED and WP:Astronomy will overlap, and then the fight comes in how much, and ... do I need to continue? --Dirk Beetstra ^{T C} 07:59, 18 August 2011 (UTC)[reply]

See #'Fancy display tech'. --Dirk Beetstra ^{T C} 08:24, 18 August 2011 (UTC)[reply]

This isn't an RFC on removing or adding these links -- that would be an issue for the MOS -- it is about bot edits. Most people are happy with these links when added by human beings after consideration as to the utility of them. After all, the citations aren't inserted by a bot, so there was a human involved when they first got added. The issue that started this is that a bot is adding these simply because it can and not necessarily because it should. That's a fundamental issue wrt all bot edits. The question of whether it should quite clearly has little consensus one way or the other and the bot owners need to grasp this and start applying Wikipedia:Bot policy rather than endlessly arguing with the rest of to accept their personal opinion. The compromise position of only adding them when the chances of utility are very high (based on the topical relevance of the journal being cited and the article containing the citation) seems likely to upset the least number of people. Bots should tackle the low-hanging fruit. Colin°^Talk 08:33, 18 August 2011 (UTC)[reply]

Well, that is where the issue is .. I think that these links should be added, irregardless of the topic, irregardless of which identifier, irregardless of who adds it, even irregardless if it is by a bot or not. That the bot does it does not make any difference. The underlying issue is not a bot issue, it is a MOS-issue. --Dirk Beetstra ^{T C} 08:49, 18 August 2011 (UTC)[reply]

Well if you think that, then create another RFC and raise it at MOS+world. This RFC is on bot additions. Look at the title. It is a separate issue because what you are demanding is that not only are these links added by bots, but that they should be always added by humans too and never removed. That's imposing a citation requirement above what is necessary to actually cite the source. Considering that citations are an issue where WP typically refuses to impose rules, you don't stand much chance. Remember that even using templates isn't mandatory and is never likely to be. Colin°^Talk 09:24, 18 August 2011 (UTC)[reply]

Well, I do stick to that .. as I do think that the bots should be adding these, as I do think they are useful, everywhere. In that way, I don't need to bother finding them myself. --Dirk Beetstra ^{T C} 09:31, 18 August 2011 (UTC)[reply]

Strongly support adding identifiers that (1) point directly to the publisher's official version of a paper (doi, or jstor); (2) provide unofficial but free versions of papers (arxiv, or sometimes but not always bibcode), or (3) provide third-party reviews of citations rather than just a copy of the abstract (sometimes but not always mr). I don't have a strong opinion about identifiers that don't meet these criteria. And I also prefer not to see both doi and jstor when they end up both going to the same place. —David Eppstein (talk) 03:10, 20 August 2011 (UTC)[reply]
Support keeping all links; I don't see how removing them would improve Wikipedia. I would also support condensing them by making the external link shorter in read mode, but would oppose condensing them by using subpages. It Is Me Here ^{t / c} 10:11, 22 August 2011 (UTC)[reply]
Support adding ONE free full text link. A bot with a mandate of making free full text available is highly valuable. A bot with a mandate of adding a second, third, fourth way to get the same article is much harder to justify. I am willing to accept an occasional redundant link added by bot if it is the only workable way to get a free full text link on a significant fraction of the articles. Just don't make a fetish out of putting every possible link for every possible paper just for the heck of it. Oh, and it should be easy for a user to invoke the bot to provide all possible information for a reference he selects, both to allow him to edit the article to add it and to follow it for his own use. Wnt (talk) 21:55, 10 September 2011 (UTC)[reply]
Support. No longer having a stable affiliation with an institution, I only have prepaid access to the journals I actually subscribe to through JSTOR, not MR, even though my subscriptions are to math journals. Wnt's comment would make sense if there was a free, stable, current link. This is rarely the case. (And I don't know why this RfC still appears to be open.) — Arthur Rubin (talk) 13:39, 30 September 2011 (UTC)[reply]
Support but do try to minimise clutter. In medicine links to doi (original publisher), PMID (free abstract + related articles & citations), PMC (free full text) are all potentially useful. I think it's a good idea to try reduce clutter via client-side javascript, e.g. showing the links only when the mouse if over them. eug (talk) 13:03, 4 October 2011 (UTC)[reply]

The discussion above is closed. Please do not modify it. Subsequent comments should be made on the appropriate discussion page. No further edits should be made to this discussion.

Standard of review for non admin closes

Closing way early per WP:SNOW. Beeblebrox (talk) 16:47, 24 October 2011 (UTC)[reply]

The following discussion is closed. Please do not modify it. Subsequent comments should be made in a new section.

{{rfc}} The policy for non admin closes of deletion discussions is currently located at WP:NACD and states that: "Decisions are subject to review and may be reopened by any administrator." The guideline provides no further guidance on when it is appropriate to reverse a NAC. The guideline should be clarified so it is clear in what circumstances an admin should unilaterally overturn such a non admin close. Monty ₈₄₅ 02:49, 24 October 2011 (UTC)[reply]

Notifications

Proposed change "Decisions are subject to review at WP:DRV and in the case of a clearly erroneous close may be reopened by any administrator." Some administrators have interpreted the authority at WP:NACD to allow the reopening of any non admin close that is at all a close call. This is inconsistent with the fundamental view outlined at WP:NOBIGDEAL that administrators are not super editors, but are instead just regular editors trusted with extra tools. Non admin closes of deletion dicussions should be treated with a similar level of deference as a regular admin close, just as any other non admin close would be outside the deletion context. If challenged, the challenge should normally be made at WP:DRV. Only in cases where the close is so seriously erroneous (for instance a clearly involved editor) that a DRV listing would be a waste of time should an admin unilaterally reverse it. Specifically, that a close was controversial should not be grounds for overturning a NAC that would have been accepted if made by an admin. Monty ₈₄₅ 02:52, 24 October 2011 (UTC)[reply]

Reject change - Non-admins shouldn't be within spitting distance of of 99.9% of XfD closures, IMO. They should be allowed to close blatant hoaxes, i.e. Barack Obama for deletion, or blindingly obvious WP:SNOW cases. If there's been anything resembling a debate with a variety of keep and deletion opinions then they have no standing to make a call in such cases. We appoint admins for a reason; this is one of them. Tarc (talk) 04:39, 24 October 2011 (UTC)[reply]
Reject change pretty much exactly per Tarc. WP:NOBIGDEAL is a big lie, anyway. There are certain things that should only done by those who've been trusted to do so, that is, administrators, and closing controversial deletion discussions is one of those. Heimstern Läufer (talk) 05:44, 24 October 2011 (UTC)[reply]

I disagree with the proposed change. As Tarc says, administrators are particularly elected and authorized by the community, among other things, to close AfDs; and are tested in their RfA on their knowledge of the deletion policy and process. This does not apply to other editors. Consequently other editors should not close any but the most obvious "keep" discussions. It is therefore not helpful to require a DRV consensus to overturn an inappropriate non-admin closure in the same way as it would be required to overturn an admin closure. On the contary, as the current instructions indicate, if a non-admin closure is overturned, then this is by itself an indication that the outcome was not obvious and the non-admin should not have closed the discussion. Sandstein 05:45, 24 October 2011 (UTC).[reply]

I have to disagree with the proposed change, admins are appointed for a reason, and I say this as an editor that does non admin closes now and then. While I disagree with the notion held by some that an XFD must have 927,482 keeps and no deletes to qualify as a NAC, this is a step too far in the wrong direction. The rules are there to stop editors who don't know what they're doing from closing AFDs. In practice, if a non admin were to close an AFD split 10/2 in favor of keep, with keep arguments being strong and delete arguments not, not many admins would overturn that. It's when people do the wrong thing that their NAC gets overturned, and this new requirement would do worse than good. Things are fine as they are, and while I admittedly would like admins to demonstrate their reasonings a bit more in their AFD closes as opposed to one word closes, thats just how things are. Steven Zhang ^{The clock is ticking....} 06:10, 24 October 2011 (UTC)[reply]
I too disagree; not that admins necessarily close disputed AfDs very well , but they at least have some screening before doing so. I would in fact strengthen the policy to permit NACs only for undisputed, procedural, or absolutely obvious closes; in my opinion, the only reason to permit them at all is to have some basis for seeing that admin candidates recognize at least the obvious. As for dealing with admins who do not explain themselves, as a first step, I would treat them as NACs: any disputed close that does not explain the rationale should be reverted and relisted if any other admin thinks appropriate. DGG ( talk ) 06:23, 24 October 2011 (UTC)[reply]
Just a comment. WP:NOBIGDEAL actually says administratorship is a big deal today: "However, Wikipedia's worldwide cultural impact and visibility grew in the intervening years, and as the community grew with it, the role of administrators evolved. Standards for adminship have risen considerably and the community generally holds administrators to a higher standard of editorial and interpersonal conduct." Have mörser, will travel (talk) 11:06, 24 October 2011 (UTC)[reply]
I also disagree. I make non-admin closures every now and then and I'm of the opinion that if anyone (with the possible exception of SPAs) makes a good faith comment/request that they believe your closure was incorrect, then you should revert it and let an admin close the XfD, even if you still think the consensus was obvious. If the consensus truly is as obvious as you think, then the next admin to come along will close the XfD the exact same way. That said, I think some of the above comments are taking things a little too far – consensus can be obvious without being unanimous. Jenks24 (talk) 12:51, 24 October 2011 (UTC)[reply]
Also disagree. People have definitely failed RfA specifically because of concerns about how they would close XfDs. This implies that the community (well, at least the "RfA regulars") only wants certain people to have the ability to close them. Wikipedia adminship stopped being "no big deal" quite a while ago because Wikipedia itself stopped being no big deal. What articles stay can have an actual impact on the real world (if they didn't we wouldn't have press agencies trying to get their clients non-notable companies an article), because of the very high prominence of Wikipedia in most search engine results. Qwyrxian (talk) 13:54, 24 October 2011 (UTC)[reply]
The proposed change does nothing to resolve a perceived issue. The phrase "clearly erroneous" is subjective and no better than "decisions are subject to review". If you have a problem with a particular admin making improper unilateral reverts of closures, then bring it up at AN/I or an RfC/U. Don't pick a worse wording. /ƒETCH COMMS / 14:21, 24 October 2011 (UTC)[reply]
Disagree with proposed change. First, any deletion discussion is subject to review by an uninvolved admin, and any deletion is also subject to WP:DRV proceedings. Second, WP:NACD is interpreted as allowing non-admins to close XfD discussions only when use of admin tools is not required, e.g. blatant hoax nominations, WP:SPEEDY or WP:SNOW keeps, and the like. For once, let's keep the bureaucracy from expanding to meet the needs of the expanding bureaucracy. --Alan the Roving Ambassador (User:N5iln) (talk) 14:39, 24 October 2011 (UTC)[reply]
Disagree with proposed change. NAC's should exist to simply relieve the burden on sysops by taking on the uncontroversial closes. If it is controversial, or if anyone objects to it for any reason (which ipso facto makes it controversial), it should then get sysop review. Regards, TransporterMan (TALK) 15:01, 24 October 2011 (UTC)[reply]
Disagree Questionable non-admin closes should be reviewable by an admin. Yes, I know that admins are just folks with extra tools, but there is no other group with extra vetting. (maybe we need that) Otherwise a questionable close by just anyone would be considered (as) final (as a close gets). North8000 (talk) 15:18, 24 October 2011 (UTC)[reply]

The discussion above is closed. Please do not modify it. Subsequent comments should be made on the appropriate discussion page. No further edits should be made to this discussion.

Non-AfD NACs

Please consider joining the feedback request service.

An editor has requested comments from other editors for this discussion. This page has been added to the following list:

Wikipedia policies and guidelines

When discussion has ended, remove this tag and it will be removed from the list. If this page is on additional lists, they will be noted below.

Is there any interest in discussing NACs of non-AfD discussions, such as WP:Requests for comment, WP:Requested moves, and WP:Proposed mergers? {{Request close}} and Category:Requests for Close were proposed at WP:Village pump (proposals)/Archive 78#Template to request a discussion be closed, where there was a little discussion of admin versus non-admin closers. Flatscan (talk) 04:24, 25 October 2011 (UTC)[reply]

Questions to consider:

Should non-admins close divided or controversial non-AfD discussions?
Can non-admin closes be summarily reverted by any uninvolved admin? (Non-admin AfD closes can be summarily reverted per Wikipedia:Deletion process#Non-administrators closing discussions.)
Should a guideline regarding non-admin closes of non-AfDs by created? Cunard (talk) 06:01, 25 October 2011 (UTC)[reply]

Notifications

Wikipedia:Requests for comment/Wikipedia policies and guidelines (tag) (list)
Individual notifications of 18 users by User:Cunard
Template:Centralized discussion (list)

I have notified the participants of the above discussion and Wikipedia:Village pump (proposals)/Archive 78#Template to request a discussion be closed about this new discussion. Cunard (talk) 06:14, 25 October 2011 (UTC)[reply]

Could we not let non-admins mark permission requests as already done? For instance when a user requests confirmation but is already auto confirmed? Jamietw (talk) 07:38, 25 October 2011 (UTC)[reply]

I wouldn't be opposed to any one, and I preform some NAC closures at MfD. No new guideline please. ~~Ebe123~~ (+) ^talk
_Contribs 10:27, 25 October 2011 (UTC)[reply]

Number 2 skips past an even more basic question. Right now anybody can undo a NAC. North8000 (talk) 10:49, 25 October 2011 (UTC)[reply]

Comment If a non-admin is going to close a discussion, first and foremost, they should be uninvolved in that discussion, as well as any editing episode that engendered that discussion. That's the same standard which admins are expected to abide by, and whether or not we believe adminship is WP:NOBIGDEAL (which is a concept which appears deprecated), I see no reason not to apply the WP:INVOLVED standard equally between admins and non-admins. As to Jamietw's comment regarding marking permission requests as already done, I see that as more a matter of clerking, as it simply notifies other editors of a previously-taken action, rather than being a unilateral or arbitrary closure of a discussion. Finally, as to creation of a new guideline, I don't see a need, since in my experience, those guidelines already exist...at least, they do in the two areas where I, myself, am most likely to consider closing a discussion: WP:AFD and WP:RFA. In both cases, non-admins are already able to close discussions in certain clearly-defined circumstances, and since one of those cases has already been discussed here, I won't re-open the can. --Alan the Roving Ambassador (User:N5iln) (talk) 13:12, 25 October 2011 (UTC)[reply]

Regarding my opinions on the answers to the three numbered questions:

1) Yes. Admins have tools, and if the action does not require the use of tools, anyone should be able to do it. Any sufficiently experienced user who is uninvolved in a discussion should be allowed to close any discussion and have their good faith summary thereof respected.

2) No. If a discussion is improperly closed, the first thing is to make a private request of the closer on their user talk page to re-open the discussion, and then if that is unsatisfactory to start a new discussion. This is the same procedure if an admin closes a discussion improperly. Admins do not carry special weight with regards to the "power" of their actions.

3) I'm iffy on this. I would rather general guidelines for closures and summations of discussions (such as RFCs, ANI discussions, etc, even AFDs) be created, and the guidelines make explicit that admins are not required to close them. If a discussion requires the use of tools, it may be best practice (but not necessarily required) for admins to close and use their tools if the conclusion is they are needed. For example, if a discussion decides that a user needs to be blocked/banned. It would be best practice (but not required) that an admin close and summarize the discussion and block the user. However, if a non-admin closed the discussion, they could also just ask an admin to follow up and block the user per the conclusions of the discussion; that should also be allowed. Admins role is purely the ability to use their tools, there is nothing about being an admin that should give them more powers that non-admins don't have excepting the use of the three tools. Take another example: For a ban discussion, where the user in question already has all of their accounts indef blocked (this does happen frequently). The ban discussion can be closed without blocking anyone, the enacting of the ban is a purely academic exercise, a non-admin should be allowed to do this, notify the account of the ban, place the ban on the ban list page, NONE of that requires any blocking, deleting, or protecting. I don't see why an admin would be needed for any of that.

So, in conclusion, for me it is pretty cut and dry: Admins only abilities are the actual use of their tools. The only thing non-admins are not able to do is use those tools, and that is not a social restriction, it is a technical one. There should be absolutely and totally no social restrictions on the actions of non-admins with regards to closing and summarize and enacting the results of discussions. In all such aspects of Wikipedia, admins and non-admins are equal. When an administrator closes a discussion, they do so merely as an experienced editor, and any other similarly experienced editor should have the same rights (and be afforded the same level of respect) with regard to closing discussions. The only thing admins are actually needed for is "flipping the switch" on blocks, protections, deletions, adding permissions, etc. But the ability to flip the switch does not give any extra social "powers" over any other user. --Jayron32 13:47, 25 October 2011 (UTC)[reply]

comment / questions A solution has bee presented, but what problem(s) is this trying to address? Are there too many non-admin closures? Too few? Too many have been overruled by admins without reasons for that? Too few? There are too few admins closing discussions? Too many? What? For myself I can think of some problems in decision making in WP, but it would help a lot if we first settled on what is the problem, and then, if needed, figured out a solution. - Nabla (talk) 14:12, 25 October 2011 (UTC)[reply]

For better or for worse (and probably for worse) "admin" means several different intended and unintended things in Wikipedia, two of them being:

Someone given the admin tools, and entrusted with using them properly
The only large group of editors vetted and tagged as being experienced and with no major wiki-problems.

We really need to create another group which is just #2. Until then, certain situations (not involving the tools) may need to call for an admin. North8000 (talk) 14:17, 25 October 2011 (UTC)[reply]

Just as a minor point, we have a group for #2: It is called "editors not currently blocked or banned". --Jayron32 20:33, 25 October 2011 (UTC)[reply]

I think realistically group (2) is just a subset of that. And the subset varies depending on the task at hand. See WP:BAG, the FA[C/R] directors and their aids, etc. You don't want some nationalist POV pusher to stay uninvolved in a particular discussion where he has on obvious bone and then swoop in to close it. It can be very hard to actually ban people here if they're not WP:GIANTDICKs. Uʔ (talk) 05:18, 26 October 2011 (UTC)[reply]

Interesting discussion. Speaking from experience, more leeway is given to NACs at RM than AfD because of the perpetual backlog we have there. I'd be interested in what others think of my NAC at Wikipedia talk:WikiProject Ireland Collaboration#Requested move: Republic of Ireland → Ireland (republic), a clearly contentious topic, but one where I thought the consensus was clear enough for a non-admin to close. Jenks24 (talk) 15:42, 25 October 2011 (UTC)[reply]
{{request close}} has an adminonly parameter, which should be used when appropriate; by default, it mentions nothing either way about admins. This seems appropriate enough. As far as I can recall from the discussion leading to the template's creation, there was no overall intention to prompt a review of admin/non-admin closures. A new tool for submitting requests needn't change the core policies or practices surrounding those requests. – Luna Santin (talk) 20:35, 25 October 2011 (UTC)[reply]

(1) Should non-admins close divided or controversial non-AfD discussions?

No, generally no. Admins are well vetted with ability to read and decide rough consensus. Participants in these processes expect some level of formality in the close. Exceptions may include: an editor with particular expertise in the area; a very experienced wikipedian with a history of good closes with very few contested. Any non-admin editor with a history of challenges to NACs should not close divided or controversial discussions. --SmokeyJoe (talk) 22:17, 25 October 2011 (UTC)[reply]

(2) Can non-admin closes be summarily reverted by any uninvolved admin? (Non-admin AfD closes can be summarily reverted per Wikipedia:Deletion process#Non-administrators closing discussions.)

Within reason. The reason for reversion should include a substantial rebuttal to the close. --SmokeyJoe (talk) 22:17, 25 October 2011 (UTC)[reply]

(3) Should a guideline regarding non-admin closes of non-AfDs by created?

Wikipedia:Non-admin closure is a quite decent guideline, despite the tag. Note that we’ve had a shift in the meaning of “guideline” and “policy”, with wikipedia-guideline now meaning real-word policy, and wikipedia-policy equivalent to real-world law. --SmokeyJoe (talk) 22:17, 25 October 2011 (UTC)[reply]

If a discussion is divided or controversial then it seems that, by definition, there is no consensus. Problems will then arise if a consensus is claimed and this seems likely no matter who closes it. Closing such a discussion as no consensus seems a reasonable action for any editor to take per WP:BOLD. Whether such a close sticks or not is a matter of practical politics, not a hard or fast rule. Note that it is our policy that Wikipedia is not a bureaucracy. Warden (talk) 12:17, 26 October 2011 (UTC)[reply]

Yes, as long as the non-admin is not involved in the discussion. Any experienced editor in good standing should be allowed to use good judgment, and make closures.
No, absolutely not. Per Jayron32. Administrators are presumed to be trusted by the community. However, their job is to perform maintenance tasks. They do not have any special authority in closing discussions, or in any of their actions. Questionable closures should be discussed with the user on their talk page. Clearly disruptive closures can be reverted by any uninvolved editor. It has worked in the past, no reason to change it now.
No, there shouldn't be a separate guideline for non-admins. As stated above, admins do not have any special "power" in closure actions. Creating a separate guideline would be instruction creep. The current closure instructions we have will be fine. Alpha_Quadrant ^(talk) 02:42, 27 October 2011 (UTC)[reply]

# No. Non admins should close only uncontested or obvious deletions. The key role of an admin is being trusted to delete or block, I wouldn't say all admins do better than all non-admins, but at least they've been screened somewhat. Too many of the non-arbs repeated closing discussions have been doing them wrong. Frankly, when I was a non-admin it would never even have occurred to me to do such a close & I found it amazing from the startthat it was considered acceptable.

Yes, if it were contentious enough for an admin to revert, they shouldn't have been closing them. It;s the only protections short of Del Rev, which has enough to do dealing with the bad closures by admins.

Yes, the guideline is simple: unanimous, obvious, uncontroversial. It's not instruction creep, its the same policy as at AfD, and it should be basic generally. DGG ( talk ) 21:36, 27 October 2011 (UTC)[reply]

Um, this was about non-AfD non-admin closures, wasn't it? SamBC(talk) 23:01, 27 October 2011 (UTC)[reply]

stupidly, I thought it was about xfds other than AfDs. I'll answer the real question later; I have a very different answer. DGG ( talk ) 04:03, 28 October 2011 (UTC)[reply]

Response by Monty ₈₄₅

Yes, being an admin should grant extra maintenance tools, not a license to determine deletion outcomes. Limiting AFD-NACs to outcomes that do not require the admin toolset is fine for practical reasons, and we don't need to encourage non admin closes in disputable situations, but they should be allowed, and if made reasonably, they should be respected to the same extent an admin close would be. Though it seems very clear that consensus does not agree with me on this point.
Yes, but only if the close is clearly wrong, not just disputable, but so out of line that if an admin had made the same close, people would question their competence. Less obviously wrong closes should be discussed and not unilaterally overturned.
Sort of, we should have one guideline on closes that applies to everyone, and it should be clear that non-admin closes are to be given nearly the same level of respect as an admin close. Monty ₈₄₅ 22:47, 27 October 2011 (UTC)[reply]

I have no problem with non-admin Jenks24 (talk · contribs)'s close of Wikipedia talk:WikiProject Ireland Collaboration#Requested move: Republic of Ireland → Ireland (republic). Jenks24 explained their closure in detail and had clearly read the discussion and neutrally analyzed the arguments.
Another non-admin close of a contentious RfC debate was Talk:Fox News Channel controversies#RFC and responses (permanent link). Cerejota (talk · contribs)'s close indicated that they had read the discussion and dispassionately considered the arguments.
When non-admin closes are well thought out, they should be accorded the same respect had an admin closed the discussion with the same level of thought. However, I've recently seen non-admin closes of discussions that show no indication that the non-admin had read the discussions. When pressed to explain themselves, the user lacked the competence to do so. In such cases, the non-admin closes should not be respected and can, in my opinion, be summarily reverted by an uninvolved admin. I agree with DGG (talk · contribs) here regarding what should be done with any admin closures of the same poor caliber. Cunard (talk) 23:43, 27 October 2011 (UTC)[reply]

Yes, if not otherwise disqualified (i.e. if uninvolved and in good standing).
Not exactly. Anyone should be able to summarily revert a close if it's egregious; this could hypothetically include a non-admin reverting an administrator's close if it's bad enough to justify that. Of course, if it's not egregious, then a summary reversion is not appropriate and we're into the realms of talk page discussion, AN/I threads and other dramah.
No, I see no need for yet another rule.

I want to add that the wisdom of Solomon is not one of the tools administrators receive on passing RFA, and nowhere does it say that administrators have a monopoly on judging consensus. They do not: anyone can judge a consensus. The reason some AfD closes are restricted to admins is because they require use of the "delete" button, not because administrators have any kind of authority to do this. Our admin corps is generally well-meaning, but has widely varying levels of competence. There are known children and known drug users among our admins, and a small percentage of them are complete fools. A rather larger percentage wouldn't have a snowball's chance of passing a modern RFA, and there are some whose tools I expect to see confiscated soon after we finally get a functioning community deadminship mechanism. There are admins who shouldn't be allowed within a hundred miles of closing a controversial RFC, and there are also non-admins who could do it perfectly competently.—S Marshall T/C 01:40, 28 October 2011 (UTC)[reply]

All of the same policies, guidelines and other norms apply equally to admins and non-admins. Therefore 1 Yes (in accord w/ usual 'rules' and one would hope common sense), 2 Absolutely not summarily reverted, no; any such reversion should be appropriately explained, 3 No.

I agree totally w/ S Marshall above. If anything, we need to move away from the notion that admins have special Godlike abilities - other than some techie buttons, they do not. Chzz ► 08:09, 29 October 2011 (UTC)[reply]

I agree that careful reading, understanding, and evaluating of arguments is not limited to admins. RfA doesn't always test these, but the common AfD-related questions tend to demonstrate the candidate's thought processes. For a closer, I would prefer an admin of 1–2 years (recent RfA, but some experience) to a random non-admin. I have been involved in a few deletion RfCs where I requested an admin close at WP:Administrators' noticeboard, and I would have opposed a non-admin close (starting with a polite note, but escalating to AN/I if necessary), had one been attempted. On the other hand, the two NACs linked by Cunard look fine, and I'm pleasantly surprised that I didn't see any tendentious objections based on the closers being non-admins. Flatscan (talk) 05:26, 2 November 2011 (UTC)[reply]

"Verifiability, not truth" RfC

There is an RfC here about whether to remove from the lead of Wikipedia:Verifiability that "the threshold for inclusion in Wikipedia is verifiability, not truth—whether readers can check that material in Wikipedia has already been published by a reliable source, not whether editors think it is true."

[insert begins here]

Note: the link above should read that the RfC can be found at Wikipedia_talk:Verifiability#RFC - Compromise proposal re first sentence Unscintillating (talk) 19:08, 30 October 2011 (UTC)[reply]

[insert ends here]

The RfC is likely to close in a few days so if you want to comment, please do so soonish. Many thanks, SlimVirgin ^{TALK|CONTRIBS} 23:45, 28 October 2011 (UTC)[reply]

This does not accurately present the situation or the change. Please see wp:ver and the wp:ver talk page. North8000 (talk) 00:47, 29 October 2011 (UTC)[reply]

Yes, the proposed change is a lot more complex than SV makes out... but you can read the proposal and the rational for it at the RFC. The RFC was announced on this page back on October 6th... but its good to have a reminder for any one who missed it. Blueboar (talk) 02:46, 29 October 2011 (UTC)[reply]

More specifically, of the portion which Slim claims that the proposal is removing from the lead, the actual proposal RETAINS all but two words of it in the lead, and moves the two words ("not truth") into a following section. North8000 (talk) 13:25, 29 October 2011 (UTC)[reply]

It's even more complex than that... the proposal does change the wording of the sentence SV is concerned about, but it seeks to retain the concept behind that sentence, and tries to explain that concept more clearly by expanding on it to an entirely new section of the policy. Please just go to the RfC, read the proposal and the rational that accompanies it, and make your own decision. Blueboar (talk) 14:52, 29 October 2011 (UTC)[reply]

That's an improvement, but there still needs to be made a change to Verifiability + Veracity. Carrite (talk) 00:56, 1 November 2011 (UTC)[reply]

Time for rollback edits to be unambiguously identified as such

All rollbacker rollbacks should be clearly labeled in the edit summary with [[WP:Rollback|RB]] along with the action (undo edit by xxx, tttt dddd), if no edit reason will ever be offered.

The generic rollback edit summary is necessary, but not sufficient. It indicates action(undo), but not reason(vand, blanking) or source(rollback).
Unidentified as a "rollback" edit, it is indistinguishable from an "undo" with no reason given (specifically discouraged in WP:Edit summaries).
It allows masquerading: An editor can say "oh, that wasn't rollbacking, it was just an undo with no reason", or vice versa.
Accountability (labeling in edit summaries) is required of all other semi-automated actions by bots and scripts (Twinkle, others). Rollbacker rollbacks should be no exception.
It is anti-convenient for every other editor faced with a reasonless edit summary, forced to diff. WP:Wikipedia is a volunteer service - action by one editor (in this case a rollbacker) should not force more work on other editors: one click, but 5 - 20 seconds of boring unproductive waiting.
Breeds discontent (see 5); it gets worse when an editor (unknown to be a rollbacker) is encouraged to included a reason in edit summaries. The replies are priceless. Point: tools should not lead to more work, or surprise, due to their unrevealed operation.
Per WP:Edit summaries (linked from WP:Editing policy#Be helpful: explain, Help: Editing, and WP:Etiquette) every editor is encouraged to summarize the action and its reason, such as "rv v" as an aid to other editors and in content disputes. I see no reason for any editor to be excused from that by a coding implementation, especially when the tool can be coded to mark it automatically.
Argument against: "The reason for the edit it already clear" - false. It's only clear when looking at diffs or at old pages, and in the rollbacker's User Contributions, but not in watchlists, where many, many editors live. (How many? Does anyone know?)
Argument against: "We shouldn't change anything - I don't like it." Well, that's not a reason. Mute, reasonless edit summaries go against the open nature of the editing process. They fail to inform following editors. They are yet an inequity, a privilege which lacks direct accountability.

I'm starting here at (policy), because it all starts with policy: all edits should have action and reason, and/or tool used, unambiguously stated. --Lexein (talk) 05:54, 29 October 2011 (UTC)[reply]

Um...I thought the whole point of rollback was that editors with its 'power' are trusted to only use it appropriately. It's almost always obvious why someone used it -- when it's not, they probably shouldn't have in the first place. ♫ Melodia Chaconne ♫ (talk) 06:14, 29 October 2011 (UTC)[reply]

No, it's not obvious without looking at the diff. Or viewing the two page versions. Or looking at the rollbacker's Contributions. Not very obvious from the edit summary, which only states that there was an undo. Please note that I'm asking for two letters to be added to the edit summary to identify it as a rollback. All other bot and script tool edits are required to be identified. That's it. --Lexein (talk) 06:48, 29 October 2011 (UTC)[reply]

OK, so, currently if you use actual Rollback, and don't have any fancy scripts, then the edit summary is e.g. Reverted edits by Example (talk · contribs) to last version by Chzz (talk · contribs).

So - you're suggesting we add a link to 'rollback' on the letters 'RB'? Can you clarify, because at the top you wrote ...along with the action (undo edit by xxx - and when using actual 'standard' Rollback, it wouldn't have any additional info - unless the user has something like User:Gracenotes/rollback.js (which allows use of Rollback with custom edit summaries).

If it's just adding RB and a link to the standard Rollback edit summary, then yes, OK, fair enough, I support that.

However, if you are suggesting that Rollback edits need any other edit summary - not just a default one - then I oppose, because the whole idea behind Rollback is that it's a quick way of removing something that so obviously needs removing that no additional details are required - and if that is abused, the userrights can be revoked - Wikipedia:ROLLBACK is the applicable guideline, whereas Help:Edit summary is just an info page. Chzz ► 07:51, 29 October 2011 (UTC)[reply]

You understood me well: the default, adding RB (and/or a wikilink) is exactly what I meant. Based on your info, any of these would be helpful to editors, because the tool used is disclosed, even if the edit reason is still not:

(Reverted edits by 23.34.45.56 (talk) to last version by 12.34.56.78) -- links "last version"
(Reverted edits by 23.34.45.56 (talk) to last version by 12.34.56.78 using RB)
(Rolled-back edits by 23.34.45.56 (talk) to last version by 12.34.56.78 )

--Lexein (talk) 08:31, 29 October 2011 (UTC)[reply]

Well if THAT is all you want, why didn't you just say so? What was the whole point of all that text about undo and suggested edit summaries and having to check diffs and so on? But really, rolling back has its own unique format that looks nothing like undo or anything else, so I'm still missing why you're even making a big deal. ♫ Melodia Chaconne ♫ (talk) 13:56, 29 October 2011 (UTC)[reply]

Fair enough. It's a process; the prior discussions were pretty gnarly, and I wanted to address that and existing policy, while proposing this change. Chzz's point helped focus my thinking about the options available. I never saw rollback summaries as looking unique, really; my bad, I guess. But still, looking unique isn't the same as being explicitly identified (named). Yes, it's a small change, so hopefully, easily adopted. --Lexein (talk) 03:37, 30 October 2011 (UTC)[reply]

I support providing a link to that guideline in the automated edit-summary. It's less confusing to the newbies with a link, and I don't see any downsides to providing it. ASCIIn2Bme (talk) 20:48, 2 November 2011 (UTC)[reply]

Categorization and other markings related to educational institutions

Greetings! I have been a Wikipedian since 2007 and am an English teacher at ITESM- Campus Ciudad de México. I have worked for some time to get my school to take contributing to Wikipedia seriously and have made headway. Not the least of the reasons for this is that they see the public relations possibilities.... very important in Mexico and to a private institutions dependent on tuition money!

Anyway, in the Commons, there are ways to categorize and otherwise mark contributions as coming from an institution to encourage institutional donations. I dont see much in the way of this here on en.wiki and Ill assume that is the case as well in es.wiki. I have managed to create this category Category:Projects related to Instituto Tecnólogico y Estudios Superiores de Monterrey, with a subcategory for some of the instruction and student pages related to when I had students write for WP at the Toluca campus back in 2007.

My question is this. Would it be possible to create and place categories such as "Articles created or significantly expanded by faculty at ITESM-CCM" on say articles themselves? I could put such on the 300+ articles I have created or expanded about Mexico. Related question, could a "tag" (for lack of better term) be put on the talk page of these articles announcing same?

Im sure both are technically possible, but the real issue is whether or not the community would accept something like this. Thanks! Thelmadatter (talk) 12:51, 29 October 2011 (UTC)[reply]

I am sure there will be no objection to a template on the talkpage like {{WikiProject Articles for creation}}, which will automatically categorize the article talkpage. Categorization in the actual article? Absolutely not. Yoenit (talk) 13:11, 29 October 2011 (UTC)[reply]

You could create a portal or project and add a template "Part of a series on ..." to those articles. It's been done to promote special interest groups which could hardly be called NPOV, so yours would be one of the more neutral ones. And you get to put a "This article is within the scope of ****, a collaborative effort to improve ****** on Wikipedia." notice on every talk page. See Wikipedia:WikiProject Council/Guide/WikiProject for more DS Belgium (talk) 14:28, 29 October 2011 (UTC)[reply]

Part of a series on... or portals have been use to promote special interest groups? Do you happen to have an example of that. Yoenit (talk) 23:24, 29 October 2011 (UTC)[reply]

Id be interested in seeing an example too. Thank you!Thelmadatter (talk) 00:58, 30 October 2011 (UTC)[reply]

You could perhaps use a talk-page template similar to {{WAP assignment}} (shown here) Chzz ► 10:41, 2 November 2011 (UTC) Template:WAP assignment[reply]

I suggest that Thelmadatter or ITESM contact the Foundation (here) about participating in the Global Education Program (also known as the Ambassador Program) before using that particular template. Please contact me for information about the program and/or assistance in joining it. -- Donald Albury 11:31, 2 November 2011 (UTC)[reply]

It has been suggested... (drawn-out merge proposals/tags)

If you look at XpressMusic, there is a notice advising readers that numerous specific articles have been marked for merging.....for over 18 months. What policies exist to monitor merge suggestions, and what policy exists which deal with long time editors like myself who feel it's better to just remove the notice and get on with our lives? Summary - are "Is has been suggested that..." templates ignored and what can be done about it? doktorb _words^deeds 13:25, 29 October 2011 (UTC)[reply]

The WP:Editing policy is that someone (you) eventually looks at the relevant talk pages, determines whether there is any agreement to do the merge, and either boldly merges the pages him- or herself, or boldly declares the merge proposal to have failed. You should feel free to do this for any page whose merge proposal is more than a few weeks old. WhatamIdoing (talk) 17:27, 29 October 2011 (UTC)[reply]

You may also (hopefully) bring it to the attention of a larger audience at Wikipedia:Requested moves - Nabla (talk) 23:46, 29 October 2011 (UTC)[reply]

I've seen some merge tags in place for over a year myself. Perhaps some sort of centralized place where to list these would be good. ASCIIn2Bme (talk) 22:09, 2 November 2011 (UTC)[reply]

Since you asked..

Part of a series on... or portals have been use to promote special interest groups? Do you happen to have an example of that. Yoenit (talk) 23:24, 29 October 2011 (UTC)

Id be interested in seeing an example too. Thank you!Thelmadatter (talk) 00:58, 30 October 2011 (UTC)

I'll put it in it's own section, easier to remove..

Take for example the "Part of a series on Feminism" in the Template:Feminism sidebar; and all the articles related to feminism

Amazon feminism ; Cultural feminism ; Postcolonial feminism ; Chicana feminism ; French structuralist feminism

Analytical Feminism ; Psychoanalytic feminism ; Difference feminism ; Equality feminism ; Standpoint feminism ; Feminist existentialism ; Material feminism ; Lesbian feminism ; Equity and gender feminism ; New feminism ; Feminist theory ; Waves of feminism ; First-wave feminism ; Second-wave feminism ; Third-wave feminism ; Postfeminism ; Postmodern feminism ; Post-structural feminism

Pro-life feminism ; Atheist feminism ; Christian feminism ; Feminist theology ; Jewish feminism ; Orthodox Jewish feminism

Lipstick feminism ; Ecofeminism ; Fat feminism

Feminism and modern architecture ; Feminist literary criticism ; Feminist science fiction ; Feminism in culture

Feminism and the Oedipus complex ; Men and feminism ; Feminist effects on society

Liberal feminism ; Marxist feminism ; Socialist feminism ; Anarcha-feminism ; State feminism ; Transnational feminism ; Feminism in international relations ; Global feminism ; Feminism and Nationalism in the Third World

Patriarchy ; Political lesbianism ; Radical feminism ; Separatist feminism

Sex-positive feminism ; Transfeminism

Feminism in 1950s Britain

Just a small sample, wouldn't want to take up the whole page.

Considering the amount of books about every subject and it's relation to feminism, like Feminist archaeology, Gender archaeology, Feminist anthropology, I can understand that WP:VERIFY would not pose a problem (independent sources on the other hand...). Whether the result is worthy of inclusion in an encyclopedia, or is even understandable, is a different question. Take for example Cyborg theory

Haraway’s cyborg called for a non-essentialized, material-semiotic metaphor capable of uniting diffuse political coalitions along the lines of affinity rather than identity. Following Lacanian feminists such as Luce Irigaray, Haraway’s work addresses the chasm between feminist discourses and the dominant language of Western patriarchy. As Haraway explains, “grammar is politics by other means,” and effective politics require speaking in the language of domination. [4] To counteract the essentializing, and anachronistic, rhetoric of spiritual ecofeminists who were fighting patriarchy with modernist constructions of female-as-nature and earth goddesses, Haraway employs the cyborg to refigure feminism in cybernetic code. As she details in a chart of the paradigmatic shifts from modern to postmodern epistemology within the Manifesto, the unified human subject of has shifted to the hybridized posthuman of technoscience, from “representation” to “simulation,” “bourgeois novel” to “science fiction,” “reproduction” to “replication,” and “white capitalist patriarchy” to “informatics of domination.” [5] While Haraway’s “ironic dream of a common language” is inspired by Irigaray’s argument for a discourse other than patriarchy, she rejects Irigaray’s essentializing construction of woman-as-not-male to argue for a linguistic community of situated, partial knowledges in which no one is innocent.

The side-effect of this bias is that women who don't owe their fame to feminism are marginalised. Look at the publications in archaeology written by women: 1991 in archaeology "Engendering Archaeology: Women and Prehistory"; 1993 in archaeology "The Archaeology of Korea", but when you go to the authors page: "She has also conducted extensive research in the archaeology of gender" 1996 in archaeology "Gender and Archaeology (University of Pennsylvania Press)." 1997 in archaeology two books from the same author, one called "Gender in Archaeology: Analyzing Power and Prestige"

Anthropology and psychology are even worse, especially when you read some of the studies published by "university professors"; the methodology is about what you'd expect from a second year student around here. They have created "gender" studies so they can talk about themselves, and good women scientists get even less attention than before.

Don't know what is worse, these special interest groups or the new skeptics movement where every fact is weighed against the possible misuse by fringe groups before it can be included in an article. There seems to be no escape from POV-pushers on wikipedia, apart from maybe mathematics. </rant> DS Belgium (talk) 02:37, 30 October 2011 (UTC)[reply]

Not all women's articles are confined to feminist topics, I don't know why you make that claim. For example, there's a "Women in Society" sidebar/bottom menu template, which is more general. Most articles that happens to be about a woman, or women, are of interest to more than one WikiProject and get another sidebar attached to it (like WikiProject Islam or WikiProject Buddhism, lots of projects have topical templates).
WikiProject Military has 14 Infobox templates, plus a Navigation box for military campaigns. WikiProject Chemistry has several templates. If the Marie Curie article doesn't have any, maybe one could be added. WikiProject Medicine has several templates too. Women in medicine could have those added to them.
Your logic that a WikiProject Feminism sidebar results in articles about women scientists getting "less attention than before" doesn't follow, neither does your statement "women who don't owe their fame to feminism are marginalized". I realize you could possibly be trolling for an argument over whether feminism or gender studies are even a valid academic subject, that seems to be your point; unfortunately for you, a lot of universities seem to think that it is. So I don't think you'll find this is a useful place to have that debate. I don't think Wikipedia ought to debate whether or not an academic discipline is philosophically justifiable or not if it's widespread, historic, commonly known, verifiable academic discipline. Feminism and gender studies are areas of academic study. (And are also political influences, and cultural influences, and economic influences, etc.)
It may be true that to certain people who have a visceral aversion to feminism or gender studies, that seeing such a template on an article would perhaps drive them away from the article. I guess if they understood liberal arts theories, like post-structuralism and the interplay between signifier and signified... that might help. OttawaAC (talk) 07:14, 30 October 2011 (UTC)[reply]

That rant is TLDR. I still read most of it, and understood more than half, I think. But I don't understand what it proves. If you reply with something long and hard to understand, I'll probably stop trying. - Peregrine Fisher (talk) 02:54, 30 October 2011 (UTC)[reply]

With a portal you get to put your template on every page you feel is related to that portal; But it looks like I posted part of another post I was preparing as well, about how special interest groups can crate articles that rely on self-published sources without being challenged. DS Belgium (talk) 04:46, 30 October 2011 (UTC)[reply]

I understand better now. I don't get what portals have to do with it. Not exactly, at least. For instance, Atheism has one of those boxes. It has a link to a portal, but it doesn't seem obtrusive. (the thing about self published sources is a widespread/longterm problem that isn't interesting to me right now, since it's so hard to fix). - Peregrine Fisher (talk) 05:20, 30 October 2011 (UTC)[reply]

I dont think these examples relate to my question as all of these portals and marking are related to topic and or ideology. Im looking for a way to give credit to an educational institution which encourages its faculty and staff to contribute by creating or significantly improving articles, not matter what the topic. I think some kind of tag on the talk page is better.Thelmadatter (talk) 01:46, 31 October 2011 (UTC)[reply]

Yes; see above. Chzz ► 10:49, 2 November 2011 (UTC)[reply]

Solving the massive backlogs at RPP and AIV

RPP and AIV often get absolutely enormous backlogs, even though we have more admins than any other Wikimedia wiki (by far). Any ideas on how to solve the backlogs?Jasper Deng (talk) 03:29, 30 October 2011 (UTC)[reply]

Fix RFA? How many admins are regulars at those notice boards? Not that many. Also, how does en. compare in terms of active admins/IP edits per day? Just because the total is higher, doesn't mean the workload per admin is lower. Monty ₈₄₅ 03:36, 30 October 2011 (UTC)[reply]

A first step might be identifying exactly what RPP and AIV are. I have no clue, speaking for myself. How about using full names of obscure acronyms at first mention? Carrite (talk) 05:14, 30 October 2011 (UTC)[reply]

In general, Wikipedia-related capitalized acronyms are found by searching for WP:(thing), for example WP:AIV. It's also good practice for posters to either wikilink or spell out full names once in discussions. ---Lexein (talk) 16:06, 30 October 2011 (UTC)[reply]

Requests for Page Protection and Administrator Intervention against Vandalism, I see after doing research... Here's a suggestion: requiring people to register and sign in to edit. Probably 2/3 of vandalism would be blown away instantly, thus eliminating the demand on scarce janitorial services. Carrite (talk) 05:17, 30 October 2011 (UTC)[reply]

Baby, bathwater. See WP:PEREN#Prohibit anonymous users from editing. Anomie ⚔ 15:37, 30 October 2011 (UTC)[reply]

Jasper - how big of a backlog are you noticing? Last I checked, AIV was impressively promptly handled. Did someone leave to take a restroom break?

If something is urgent and clear-cut, ask at irc://irc.freenode.net/wikipedia-en-help , either !helper or !admin . --Lexein (talk) 16:06, 30 October 2011 (UTC)[reply]

The wait was once fully two hours. I'm a little more concerned about RPP.Jasper Deng (talk) 17:42, 30 October 2011 (UTC)[reply]

Two hours is not a long wait. I would define anything which got done at Wikipedia in two hours as "prompt". --Jayron32 01:54, 31 October 2011 (UTC)[reply]

Not when it's urgent, which is pretty often (Backlogs are OK at AfD and other admin-requiring processes that aren't as urgent).Jasper Deng (talk) 01:59, 31 October 2011 (UTC)[reply]

I've always gone to IRC for anything urgent (rapidfire, defamation, policy blankers), especially if I've noticed 3 or 4 AIV events ahead of me. Perhaps IRC should be more prominently recommended for urgent matters? --Lexein (talk) 02:05, 31 October 2011 (UTC)[reply]

Not everyone knows how to or bothers to use IRC (including admins).Jasper Deng (talk) 02:10, 31 October 2011 (UTC)[reply]

So, in other words, you can't be bothered to do the thing that will fix the problem you are having, so you expect the rest of the world to conform to what you want. I just want to get this straight before we decide how to address the problems you are having... --Jayron32 02:50, 31 October 2011 (UTC)[reply]

(Now now, I took his point as valid, while seeking a longer term solution to the foundation issue ...) --Lexein (talk) 02:55, 31 October 2011 (UTC)[reply]

You'd be surprised how many admins hang out on IRC. But I hear ya; though I'm happy with the IRC turnaround time I've gotten, that's beside the point. How would one improve AIV and RPP response times? --Lexein (talk) 02:55, 31 October 2011 (UTC) (edited) --Lexein (talk) 03:19, 31 October 2011 (UTC)[reply]

One idea I was contemplating is "Report reviewers." They would review all RPP and AIV requests as "endorsed" or "declined", saving admins a bit of time.Jasper Deng (talk) 03:27, 31 October 2011 (UTC)[reply]

It is an interesting idea, but if someone is trusted enough that an admin would block someone based on their endorsement of the report without further review, they really should be made an admin and cut out a step. If they are not that trusted, it doesn't really help does it? Monty ₈₄₅ 03:34, 31 October 2011 (UTC)[reply]

The admin does not necessarily have to agree.Jasper Deng (talk) 03:59, 31 October 2011 (UTC)[reply]

Would these "support"/"oppose" comments be inline, or would this necessitate another pipeline as used in DYK? If they're non-binding, then I'm afraid I don't see the utility. If they're binding, admins will (correctly, IMHO) balk. I think the best approach is more admins. Recruit, nominate, and approve admin candidates even if they express a desire to focus in particular areas. I can think of one editor (not me), talented and fair-minded, nominated for adminship once, who would rock at clearing the very backlogs mentioned. --Lexein (talk) 05:58, 31 October 2011 (UTC)[reply]

Part of the problem is the amount of workload. There are backlogs all over. If you try to adjudicate many of the discussions, you are often taken to task for being wrong. All of this wears on you over time. Would more admins help? Sure. Something like RPP is not a simple drive by and cleanup. It means looking at what is happening and then deciding what is the correct action. However, if everyone would pitch in and clear up stuff that does not need an admin, then some admins could spend more time on admin work. Anyone want to clean out the misclassified stubs in the list of suspect articles? Vegaswikian (talk) 06:28, 31 October 2011 (UTC)[reply]

(WP:Wikignome demurs with a terse mumble, nimbly diving back into the mushroom forest, goes back to Category:Articles needing link rot cleanup, etc.) --Lexein (talk) 07:12, 31 October 2011 (UTC)[reply]

As someone who watches RFPP there aren't that many challenges. Ultimately if you want to reduce the number of times you as an admin get challenged the only way you are going to achieve that goal is to make the relevant policy clearer and more deterministic. If you want to reduce the drama caused by being challenged you need to be prepared to agree to disagree earlier and/or have a more productive escalation process - possibly including some way of getting another admin(s) to give a second opinion. -- Eraserhead1 <talk> 08:10, 31 October 2011 (UTC)[reply]

Policies for access to the OAI-Repository

Disclaimer: If this is the wrong place to ask, please delegate me to where it would be more appropriate :)

Where can I find the policies (if they exist) about how to get access to the Special:OAIRepository (OAI = OAI-PMH = Open Archive Initiative - Protocol for Metadata Harvesting)? The background story is, that there is the DBpedia project, which, in a nutshell, is about extracting structured data from Wikipedia articles and freely publishing that data as downloads and via publicly accessible databases (called triple stores/SPARQL endpoints). As the structured data enables one to ask complex queries on Wikipedia, DBpedia already received quite a lot of attention. At the University of Leipzig we have developed a live-synchronization module which keeps the data in such triple store up to date. However, so far we have done the extraction only for the English Wikipedia, as we do not have resources for maintaining multiple instances. But other research groups have already announced their interest in setting up the DBpedia live extraction for other languages. So we would like to know how they could apply for access or who they should contact. Aklakan (talk) 10:10, 1 November 2011 (UTC)Aklakan[reply]

Translations from other-language Wikipedias and WP:V

This seems like muddy terrain to me. Wikipedias (including the English one) are generally not considered reliable sources for writing articles (except perhaps in very narrow circumstance in writing articles about Wikipedia itself, but I don't want to get into that.) Literally WP:SAYWHEREYOUGOTIT says that if one translates an article from another language Wikipedia, and if the translator does not personally check the references to see if they support the text he/she is translating, then the translated text should actually be cited to another Wikipedia. Of course, in practice it's difficult to prove what the translator did this one way or the other. Thoughts on this? (Yes, I know, WP:AGF, anything else?) ASCIIn2Bme (talk) 20:45, 2 November 2011 (UTC)[reply]

I would say, definitely state where it was translated from, but then the article is subject to the usual rules; citations to reliable sources (not the other language Wikipedia), no original research, etc. If it can't be supported by reliable sources, it may have to be stubbed. -- Donald Albury 22:54, 2 November 2011 (UTC)[reply]

Translations should always be acknowledged on the talk page using {{translated page}}. Translators should also always check the sources in the original article. But often it may be quicker and easier to write the article from scratch in English than to attempt a translation, given that frequently citation standards on other language Wikipedias are more lax than on en:wp and that, where possible, English-language sources are to be preferred. --NSH001 (talk) 00:15, 3 November 2011 (UTC)[reply]

When I translate articles, I do check out the cited sources as far as I can -- usually they are online anyway; the main reason to do this is to see for myself that nothing is misrepresented in the article, and that it's worth translating in the first place. Plus, sometimes, it's helpful to see the cited source for translation purposes if the article isn't clear enough for me to understand -- looking at the cited source itself will help me write the English version with more accuracy. And I always leave a message on the Discussion page of the article indicating that it's a translation, from which language Wikipedia it came from, and if I've added more material to it myself. I try to add English sources, even if only to a Bibliography.
That being said, there are some other issues worth pointing out for translating articles into English from other Wikipedias, which present policy challenges: Some articles on topics that are well-known to another language/culture, but essentially unknown in English, will not have many (or any) English sources. And if the original article cites only books published in the original language, and I don't have access to those books, I can't verify the cited sources myself. If articles on such topics cannot be posted to the English Wikipedia due to the fact that the article translator can't verify the sources, I think that'd be a pity, and a lot of global information will not make it into the English Wikipedia. There can be cultural topics that only have physical documentation in a native language, and the cited sources are limited to a small geographic area, and if English Wikipedians can't get their hands on the source material... What to do? Let the articles disappear down the foreign language memory hole rather than translate them?
There are some ways to try to evaluate the trustworthiness of cited sources that I can't see for myself -- does the ISBN of a book check out, are there a large number of other cited sources given by the article's original editor that do check out, does the article have credibility for other reasons (context of links in the article check out, prior track record of the original editor, etc.). That's more personal discretion and judgment though, rather than complying with a bullet-proof policy.
How much weight should we give to reviewers and editors of the other language Wikipedias? And the people reading English Wikipedia who could well have knowledge of the original language for the article, click on the link to it, look up sources to confirm or challenge the info, etc. OttawaAC (talk) 01:03, 3 November 2011 (UTC)[reply]

RfC for proposal to promote Notability (astronomical objects) to notability guideline

An RfC has been put out on the proposal to promote Notability (astronomical objects) to a notability guideline. Please join the discussion. This is the result of a long discussion at WikiProject Astronomy. Cheers, AstroCog (talk) 01:19, 3 November 2011 (UTC)[reply]

@@ Line 476: / Line 476: @@
 How much weight should we give to reviewers and editors of the other language Wikipedias? And the people reading English Wikipedia who could well have knowledge of the original language for the article, click on the link to it, look up sources to confirm or challenge the info, etc.
 [[User:OttawaAC|OttawaAC]] ([[User talk:OttawaAC|talk]]) 01:03, 3 November 2011 (UTC)
+== RfC for proposal to promote [[Wikipedia:Notability_(astronomical_objects)|Notability (astronomical objects)]] to notability guideline ==
+An RfC has been put out on the proposal to promote [[Wikipedia:Notability_(astronomical_objects)|Notability (astronomical objects)]] to a notability guideline. Please join [[Wikipedia_talk:Notability_(astronomical_objects)|the discussion]]. This is the result of a long discussion at WikiProject Astronomy. Cheers, [[User:Astrocog|AstroCog]] ([[User talk:Astrocog|talk]]) 01:19, 3 November 2011 (UTC)