Jump to content

Wikipedia:Village pump (proposals)/Archive 218

Add links
From Wikipedia, the free encyclopedia
Village pump (proposals) archive  

This page contains discussions that have been archived from Village pump (proposals). Please do not edit the contents of this page. If you wish to revive any of these discussions, either start a new thread or use the talk page associated with that topic.

< Older discussions · Archives: A, B, C, D, E, F, G, H, I, J, K, L, M, N, O, P, Q, R, S, T, U, V, W, X, Y, Z, AA, AB, AC, AD, AE, AF, AG, AH, AI, AJ, AK, AL, AM, AN, AO, AP, AQ, AR · 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, 20, 21, 22, 23, 24, 25, 26, 27, 28, 29, 30, 31, 32, 33, 34, 35, 36, 37, 38, 39, 40, 41, 42, 43, 44, 45, 46, 47, 48, 49, 50, 51, 52, 53, 54, 55, 56, 57, 58, 59, 60, 61, 62, 63, 64, 65, 66, 67, 68, 69, 70, 71, 72, 73, 74, 75, 76, 77, 78, 79, 80, 81, 82, 83, 84, 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, 196, 197, 198, 199, 200, 201, 202, 203, 204, 205, 206, 207, 208, 209, 210, 211, 212, 213, 214, 215, 216, 217, 218

RfC to determine how reflinks are linked or not in the |work= field as done by bot. -- GreenC 20:05, 3 April 2025 (UTC)

User:GreenC bot (WaybackMedic) fixes broken URLs semi-manually per request at WP:URLREQ on a per domain basis. The bot is uniquely programmed for a single domain.

One of the features is incidentally adding reflinks in the |work= field for example converting |work=theguardian.com --> |work=The Guardian. This is done per the MOS WP:REFLINK which states

"Citations stand alone in their usage, so there is no problem with repeating the same link in many citations within an article".

This is done mostly cosmetically, when making other changes within the citation or article. It is not the bot's primary purpose, but since I am making bespoke code specific to a domain, I can easily do this at the same time.

An editor recently requested this feature be disabled. So that I may continue fixing dead links, I complied and set to feature set 2A below. However I would like to see if there is preference from other editors, and to offer a set of features available.

There are 2 choices (bot or nobot), and if bot, 4 choices how:

1. Don't mess by bot
2. Acceptable by this bot, within certain rules.
A) Convert domain names to work name but don't wikilink - template documentation requires name of the work: |work=theguardian.com --> |work=The Guardian
B) Convert domain name to work name and wikilink it: |work=theguardian.com --> |work=The Guardian -- default for the past 5 years it is low volume
C) Wikilink existing work names: |work=The Guardian --> |work=The Guardian - can be high volume
D) Both B and C - recently done for thetimes.co.uk only, that triggered the complaint due to the high volume
E) No opinion
3. Other suggestion. I can not guarantee other suggestions could be programmed. Thus, please include one of the above in addition to any custom suggestions. Custom suggestions without one of the above will default to #2.E the closer will sort it out.

Note: |work= could also be: |website=, |magazine=, |newspaper=, |publisher=

The complainant User:SchroCat at User_talk:GreenC_bot#Stop_linking_newspapers. Others who may be interested based on their knowledge of this tool and CS1|2: @Οἶδα, MrLinkinPark333, Pppery, Chew, Sariel Xilo, Lyndaship, Nemo bis, Kailash29792, Random fixer upper, Headbomb, Trappist the monk, Redrose64, Izno, ActivelyDisinterested, and Lewisguile:

A !vote might look like Option 1 or Option 2B etc.. -- GreenC 20:05, 3 April 2025 (UTC)

  • I don't see why we should stop linking newspapers, etc. Linking is extremely useful. Headbomb {t · c · p · b} 20:08, 3 April 2025 (UTC)
    I !vote 2B, fwiw. Headbomb {t · c · p · b} 16:06, 4 April 2025 (UTC)
  • Option 2B, second choice 2D. Like Headbomb, I am a passionate supporter of linking reference works. In the current information environment, a top question every literate reader should be asking when they look at a reference is "Is this source reliable?" They should not have to blindly trust that it is, and a link to the article about it provides an easy way for them to investigate further. And there is extremely little downside, since references are out of the way at the bottom, and external links are marked as external with the icon, so the source links aren't distracting anyone (thus the guidance at MOS:REFLINK; WP:REFLINK goes elsewhere).
    That said, as much as I urge all editors to make linking the default in their articles, it is something where we allow variation per WP:CITEVAR, and linking only one/some source(s) could create discrepancy. For the situation in B, if an article has not had enough care put into its references to specify the publication name rather than just the URL, I see no issue with updating it to our best practice. (I make a similar call in the AWB task where I correct e.g. |work=New York Times → |work=The New York Times.) But I'm slightly more hesitant to do so for the situation in C. Sdkb talk 20:32, 3 April 2025 (UTC)
  • Option 2B - If it ain't broke, don't fix it. But when it is, may as well kill two birds with one stone. Adding wiki-links to citations is entirely unnecessary unless you are already fixing the citation. I would, at the very least, want it to change from the URL to the name, and if we're already updating it, why not also add a wiki-link? But forcing it to be linked without changing the contents (what is suggested in 2C) doesn't feel super necessary, unless you are already updating the citations.
    The bigger concern here is seeing what should be in the work param in the publisher param. I would, regardless of how it gets changed, make sure the publisher param is moved to work. E.g. changing |publisher=New York Times to become |work=New York Times . And, of course, you can add the wiki-link to this as well when doing so. This might end up being option 3, if the bot doesn't already do this, but I need to make sure I get this comment out. Chew(V • T • E) 21:07, 3 April 2025 (UTC)
  • Option 2A. (Second choice option 1). It should be a decision by editor discretion at page level whether to link newspapers or not. It should not be decided by a few of people here or a bot. Having inconsistently formatted references, which is what this will lead to, is amateurish and second rate; it also clashes with the consistency requirements required for featured articles. The MOS does not require these links, and bots should not be forcing a change if editors have decided on following the MOS to keep them unlinked. - SchroCat (talk) 21:11, 3 April 2025 (UTC)
  • Option 2D or 2B. One complaint in 5 years does not override the clear and demonstrated usefulness of wikilinking reference names. If volume is a concern for 2D I would support rate limiting it (e.g. a maximum number of otherwise unchanged articles per day) Thryduulf (talk) 21:29, 3 April 2025 (UTC)
  • A - please, per SchroCat, or a lot of editors are in for a lot of work undoing well intentioned bot edits. Gog the Mild (talk) 21:34, 3 April 2025 (UTC)
    Which they should not undo if there is a consensus per this discussion. Izno (talk) 21:50, 3 April 2025 (UTC)
    There is no requirement in the MOS for them to be linked, so when editorial discretion follows the line of the MOS in not linking, people will (rightly) revert something that has forced inconsistency into an article. - SchroCat (talk) 01:44, 4 April 2025 (UTC)
    Reverting a bot performing a job with consensus of a level that might be demonstrated in this discussion is simply disruptive behavior and would be worthy of a block. Izno (talk) 18:04, 4 April 2025 (UTC)
    No it isn’t. It is editing within the confines of the MOS. If you think editing within the MOS is worthy of a block, that’s a little on the extreme side that wouldn’t stand up long at a review. - SchroCat (talk) 18:34, 4 April 2025 (UTC)
    I'm not arguing about the MOS though. If a consensus is established here, you have to abide by this consensus also. Not doing so is what earns you the block. Izno (talk) 18:40, 4 April 2025 (UTC)
    I think you’re misunderstanding what this is about. This discussion is about whether to allow a bot to undertake one single step: it is not a discussion that forces all those aspects of the articles to remain like that forever. If this proposal gets consensus I will not stop the bot from undertaking that task (pressing the stop button to stop it, for example, would be against the consensus, and yes, it would be disruptive and blockable). But I am allowed to edit the article afterwards in my way I wish: this discussion does not change the MOS which will continue to allow flexibility on the point that the linking is based on editorial discretion on individual articles. - SchroCat (talk) 18:47, 4 April 2025 (UTC)
    In any way that you wish is in fact not the truth, because that leads to edit warring with the bot. Good luck with your interpretation if the bot's approach becomes consensus. Izno (talk) 16:09, 12 April 2025 (UTC)
    Editing one part of something a bot does that is still within the MoS will not lead to a block, unless the admin rally wants to be dragged to ANI for overstepping any reasonable grounds of behaviour. There is nothing in the RfC that means titles can be unlinked or that all titles must be linked. The MoS says differently and there is nothing in the RfC that will change that. It’s certainly not edit warring either: it’s entirely within WP:BRD to delink the names. I’m not sure why you’re so keen to block people for editing within the bounds of the MoS and our existing editing guidelines, but I hope you try and look at this more rationally before you act. - SchroCat (talk) 16:49, 12 April 2025 (UTC)
    As I said, good luck with your interpretation. Izno (talk) 19:59, 12 April 2025 (UTC)
    My “interpretation” of the MoS and the project norms of editing are the commonly accepted ones. I’m not the one trying to stretch a possible consensus on one bot’s actions to cover an entirely different part of the MoS. - SchroCat (talk) 20:28, 12 April 2025 (UTC)
  • At least 2A. I'm pretty ambivalent about whether something is linked in the work field and have personally disagreed with the practice in the past, mostly because people must eventually figure out what the Guardian is. Izno (talk) 21:54, 3 April 2025 (UTC)
  • Option 2B, or 2D; per Sdkb and Thryduulf. Also per Chew, I have also personally not witnessed the bot adding reflinks ONLY. It is an addition made alongside a different function the bot is already performing, for example the migration of URLs from thetimes.co.uk to thetimes.com. If the bot were "fixing" every article without reflinks then that would absolutely be excessive and would have been complained about already. Instead it is merely performing a useful addition, one which is supported by MOS:REFLINK, and one within and edit that is already being performed. This had certainly already been discussed, but I fail to see how reflinks are not helpful. Look at a recent page creation like If You Only Knew (Acetone album). Not a single reflink, and most sources being ones I am unfamiliar with. I agree with Sdkb. Every reader should be wondering, "Where is this information supported?", and upon hovering/clicking on an inline citation and seeing no reflink (made even worse in the absence of URLs) readers are not helped. On the aforementioned article, I would have to copy and paste 20 work titles to even somewhat determine that these are reliable sources. I also believe most references are now being auto-generated from URLs with tools like the one in visual editor, which does not add reflinks. Οἶδα (talk) 22:25, 3 April 2025 (UTC)
  • 1 or 2A. MOS:REFLINK allows repetition of wikilinks within references, but does not require them. Until that changes (which would be a different discussion), linking or not is discretionary, and consistently not linking (or other consistent approaches, like linking only first appearance) shouldn't be changed without discussion. On top of that, changing it as this bot does - on a per domain basis - would introduce inconsistency in most articles, unless one happens to cite only sources from the domains the bot is working on. Nikkimaria (talk) 00:03, 4 April 2025 (UTC)
  • Option 2B or 2D; the links are useful, especially for sources that users may not know about. For instance, most people who actually read the sources will have some idea of what The Guardian is, but I've edited NZ-focused articles that link to The Post which readers may not be familiar with. I personally only add links to the first mention of a work in citations, but IMO a bot adding redundant links is better than there being no links because humans have better things to do than add them to all articles.  novov talk edits 01:08, 4 April 2025 (UTC)
  • I don't oppose links in work fields, although I haven't been using them myself much recently, but it does feel close to a cosmetic change. It may be preferable for 2C to happen only alongside other changes. CMD (talk) 02:50, 4 April 2025 (UTC)
    I would add as well that I do not find the overlinking arguments convincing. Each reference should be fully functional as a standalone item, as that is how it will be interacted with. MOS designed for article prose will not apply in the same way. CMD (talk) 14:56, 22 April 2025 (UTC)
  • 2A. I don't think it's a big consistency problem if some instances of The Guardian in the citations are wikilinked and some are not, or if The Guardian is wikilinked but The New York Times is not, but I don't think a bot should be making that call. 2A is a useful clean up, though. I would be OK with 2B if an editor specifically invoked the bot for a given article, if there's any way to do that. An edit that just adds a wikilink is not cosmetic, but has the potential to flood watchlists so I would prefer not to see 2C. Mike Christie (talk - contribs - library) 04:06, 4 April 2025 (UTC)
  • 2A I don't understand why we would wikilink the Newspaper if there is a URL linking the actual article that is being referenced, which shows where it comes from. It's also not part of MOS either.Davidstewartharvey (talk) 05:21, 4 April 2025 (UTC)
    @Davidstewartharvey, as I argued in my !vote, the main reason is to help readers more easily evaluate the trustworthiness of a source. Why is an external link insufficient for that? Well, I'm sure The Daily Mail describes itself as a reliable source. Sdkb talk 14:32, 22 April 2025 (UTC)
  • 2A, for the convincing reasons above. Tim riley talk 07:29, 4 April 2025 (UTC)
  • 2B I tend to wiki-link from the work/newspaper/magazine/etc. field when I'm constructing citations, as I think it helps to establish reliability, as well as providing sometimes much-needed disambiguation when titles are similar, if not the same, for several publications. That The New York Times is often referred to as The Times is a good example of why such disambiguation is needed. Dhtwiki (talk) 10:05, 4 April 2025 (UTC)
    But it you sctuslly link the article by URL, it takes you directly, so why would you need a wikilink to show who the works is? Davidstewartharvey (talk) 10:17, 4 April 2025 (UTC)
    You may be assuming that the cited article remains live on the newspaper's main active site. Older newspaper articles are often to be found only on archive sites such as British Newspapers Online, Newspapers.com or Gale. The article is normally paywalled and most readers can't click through; even if they can, the link won't help them find the newspaper's main website. MichaelMaggs (talk) 13:17, 4 April 2025 (UTC)
    For some publications, especially those that are less well known or are published in foreign languages and using non-Latin scripts, it's not easy to discern whether they are legitimate news sources or something less reliable. Dhtwiki (talk) 14:50, 4 April 2025 (UTC)
    @MichaelMaggs and @Dhtwiki That may be the case, however, automatically linking the work of each reference, which in some cases may have been used more than once (I.e. two or three Times articles), would be overlinking. Davidstewartharvey (talk) 15:03, 4 April 2025 (UTC)
    Not when it’s useful, namely in the references. MichaelMaggs (talk) 15:25, 4 April 2025 (UTC)
    How can overlinking of, for example, The Times, been useful? If an article is linked to three different articles, a bit would link every single ref to it. That is overkill. Which is why it should be down to editors to link the article. Davidstewartharvey (talk) 15:44, 4 April 2025 (UTC)
    To help to avoid overlinking is in large part why I voted for 2B, not 2C. Dhtwiki (talk) 15:58, 4 April 2025 (UTC)
    Reflinks stand alone in their usage with regards to overlinking. Whether the bot should do it is another argument which is already being discussed here. But simply put, how many times a work is reflinked is not an instance of overlinking (MOS:REFLINK). Οἶδα (talk) 20:37, 4 April 2025 (UTC)
    I fail to see how reflinks should ever be suppressed in the absence of URLs. Readers are not further directed anywhere for content or context. But as you alluded to, URLs correct that somewhat. However, in my view, readers are still lacking necessary context, as a URL is only a primary source for information about itself, without providing broader context for readers to discern whether the source they are reading is reliable. Unless a source is deprecated or blacklisted, any URL can be added. Also a lot of cited news sources have generic names (“Gazette”, “Herald”, “Star”, “Record”, “Mirror”), often cited without the added context needed to disambiguate, such as location. I understand the argument here is that the bot should not be making the decision to add reflinks, but this is what I find to be true at least with regards to best informing readers. Οἶδα (talk) 22:23, 4 April 2025 (UTC)
    I also find it interesting that WP:CITEVAR states: "The data provided should be sufficient to uniquely identify the source, allow readers to find it, and allow readers to initially evaluate a source without retrieving it." How does one "evaluate a source without retrieving it" in the absence of further context or content? Οἶδα (talk) 05:18, 5 April 2025 (UTC)
    If it were the case that a link is needed to uniquely identify a source or to evaluate it, then the MOS would already insist on the need for such links. It doesn’t and instead leaves the question down to editor discretion at the level of individual articles. If you wish to claim that this is the only was to identify or evaluate a source, then you’ll need to open an RfC to change the MoS to do just that. - SchroCat (talk) 05:56, 5 April 2025 (UTC)
    I was not claming that MOS:REPEATLINK prescibes reflinks as mandatory. I was merely quoting a guideline whose choice of words I found interesting in the context of what I emphasized above. You are correct though, this would require an RfC. Οἶδα (talk) 08:19, 5 April 2025 (UTC)
  • Option 2B, or second choice 2D, per Headbomb, Sdkb, Thryduulf and Οἶδα. I’m particularly unconvinced by the argument that as MOS:REFLINK permits non-linking, the bot should never be allowed to do anything better, and that if it does “people will (rightly) revert”. The MOS no more says it's right to unlink than it does to link. Convenience links to newspapers and other works are extremely useful, and that utility is in my view far more important than trying to enforce essentially trivial internal consistency within a single article's source formatting. The difference, after all, is merely that the names of some works within sources may appear in blue, and some may not. So what? In the longer term, we'd serve our readers better by gradually moving towards linking all works where possible. MichaelMaggs (talk) 11:10, 4 April 2025 (UTC)
    If you wish to rewrite the MOS to insist on links, then you will need to have the discussion there to change it. At present it does not require links to be linked or unlinked: it is down to the consensus of editors at each individual article. Trying to force the issue by having a bot do it is a form of back-door instruction creep by proxy. - SchroCat (talk) 11:17, 4 April 2025 (UTC)
    Bots are permitted to do whatever the community authorises them to do, in discussions such as this. Their actions must be consistent with the MOS, yes, but few would be able to operate if they could do only what is specifically insisted upon by the MOS. We're not addressing here what individual editors must or can do; only what authorisation the bot should have to do something that is generally permitted by the MOS. MichaelMaggs (talk) 11:47, 4 April 2025 (UTC)
    As I've said, it's a back-door instruction creep by proxy. If this rather disruptive measure passes, I don't look forward to reverting these when I see them, but will do so. - SchroCat (talk) 12:00, 4 April 2025 (UTC)
    You have voiced your opinion and it is well-understood. To be clear though, you reverted the bot wholesale, which included the main edit it was performing (migrating The Times URLs). Such a reversion would be disruptive. Οἶδα (talk) 20:41, 4 April 2025 (UTC)
    MichaelMaggs, you say that REFLINK doesn't imply that bots should not link; I'd like to ask you more about that. If one of two options is allowed by the MoS, but the community authorizes a bot to always apply one of those two options, that clearly doesn't contradict the MoS, but doesn't it effectively implement one of those two options to the exclusion of the other? I think the issue here is whether the assertion in the MoS that something is up to editor discretion implies that it should not be changed globally (that is, it should be decided at the individual article level). Do you see this differently? Mike Christie (talk - contribs - library) 14:54, 4 April 2025 (UTC)
    Yes, if what the bot is doing is authorised globally. The point of this discussion is to determine that. MichaelMaggs (talk) 15:31, 4 April 2025 (UTC)
  • Option 2A (second choice option 1). Linking works should not be automatic -- that is antithetical to MOS:REPEATLINK. If the work is referred to repeatedly in the article, it will create unhelpful overlinking. Instead, linking should be a decision made by the editors at each page. -- Ssilvers (talk) 14:11, 4 April 2025 (UTC)
    References are not the article. If I click on reference 3, I want a link in reference 3. If I click a link on reference 49, I want a link in reference 49. That it's linked in reference 3 is irrelevant. Headbomb {t · c · p · b} 15:11, 4 April 2025 (UTC)
    One could make the same strawman argument about wikilinks in general, but we don't link everything, everywhere. - SchroCat (talk) 15:36, 4 April 2025 (UTC)
    Wholly concur with SchroCat. A modicum of common sense is wanted here. Tim riley talk 18:02, 4 April 2025 (UTC)
    @Ssilvers: To be clear, what you said is not correct. As I stated above, and which is actually quoted in the OP, reflinks stand alone in their usage with regards to overlinking. Whether the bot should do it is another argument which is already being discussed here. But it is not correct to suggest this is antithetical to MOS:REPEATLINK. What you are referring to is the guideline for links within article sections, not for citations. Οἶδα (talk) 21:31, 4 April 2025 (UTC)
    To be clear, what you are saying is not correct. It is not necessary or helpful for refs to link the names of works again and again, no matter how many times you repeat that you like it. -- Ssilvers (talk) 00:31, 5 April 2025 (UTC)
    Okay I'm not sure what we're doing here. Yes, I have voiced support for reflinks in this discussion. It appears I misunderstood what you wrote here and for that I apologise. You stated rather forthrightly that repeat links constitute overlinking, but then stated that it is up to consensus. I understood "overlinking" not as a general reference to an article's "citation style", so I again apologise for misunderstanding. No need for snark. Οἶδα (talk) 04:19, 5 April 2025 (UTC)
  • Option 2A (second choice option 1). Linking publications/publishers should not be automatic. We have WP:CITEVAR for a reason; it's disruptive to force a certain citation style using a bot. I had a situation with an article several months ago where there was some controversy over the linking of publishers for book citations; I see no reason why |work= can be thought to be exempt from such differences in opinions. With citation formatting, it is much better to allow human flexibility than to force-format things a certain way with a bot. We should be deferring to human judgment here. This proposal honestly feels like a backdoor attempt to force a certain citation style across a wide range of articles, contrary to common sense. Hog Farm talk 20:11, 4 April 2025 (UTC)
    It is clear the issue is boiling down solely to WP:CITEVAR. So I must ask, to what extent are reflinks actually considered part of a specific "citation style", meaning that they can become part an article's established and consistent stylistic choice, one that must be deferred to and adhered to, with any addition/removal seen as an undue disruption warranting reversion? When I think of WP:CITEVAR, I think of everything outlined at Wikipedia:Citing sources: full citation, short citation (Harvard, MLA), general references, templates, no templates, citation order, etc. Not the "variation" of whether there are reflinks or not. Could an editor also be reverted for adding an author link to a citation because it is not "consistent" in the article or because the most significant contributor of the page decides it goes against their personal/established preference? This seems like a possibly misguided cross-application being that it is not unambiguously supported by WP:CITEVAR or consensus elsewhere. Otherwise, if they are considered a component of "citation style" or the ambiguity skews toward that interpretation, then I suppose 2A really would have to be the way to go. At least until the bot can account for an article's prevailing practice. Οἶδα (talk) 05:26, 5 April 2025 (UTC)
    The broader MOS:VAR indicates: "Sometimes the MoS provides more than one acceptable style or gives no specific guidance. When either of two styles is acceptable it is generally considered inappropriate for a Wikipedia editor to change from one style to another unless there is some substantial reason for the change...Unjustified changes from one acceptable, consistently applied style in an article to a different style may generally be reverted. Seek opportunities for commonality to avoid disputes over style. If you believe an alternative style would be more appropriate for a particular article, seek consensus by discussing this at the article's talk page or – if it raises an issue of more general application or with the MoS itself – at Wikipedia talk:Manual of Style." With regards to the matter of wikilinking works within references, MOS allows but does not require this be done, bringing VAR into play. Nikkimaria (talk) 13:58, 5 April 2025 (UTC)
    @Οἶδα: Try running an article with inconsistent linkage through FAC and you'll probably see where this gets sticky. Hog Farm talk 17:00, 5 April 2025 (UTC)
    Then an RfC would be appropriate. I understand that the MOS leaves the door open to variety, but this does not appear to be a very common or contentious phenomenon on Wikipedia and as such this exact point does not seem to have been deliberated much before. If such a discussion exists, I cannot find it. In the absence of such discussions, it's hard to not bring these aspects up because CITEVAR is being cited as if a community consensus was determined to remand the issue of reflinks in citations to individual consensus. Rather than a general application of MOS:VAR. Was there ever a discussion to determine consensus for a large-scale disruption of established citation styles by the addition/removal of reflinks? I don't believe so. Again, it's not a very common or contentious phenomenon, nor have I seen bots performing these additions which would trigger such discussions until now. If this discussion indicates anything it is that the community would like a consensus on reflinks, and apparently we are not going to have it through a decision about this bot. There is enough ambiguity with WP:CITEVAR as it makes no prescriptions for reflinks (literally no mention whatsoever) nor does it confirm that reflinks are an established component of an article's "citation style", which is what I was referring to above. Οἶδα (talk) 23:18, 10 April 2025 (UTC)
  • 1 or 2a: per above, and WP:CONLEVELS -- a discussion as to how to programme a bot shouldn't override the MoS, which is to leave this up to individual discretion. We already see good-natured but time-consuming bot edits from various bots which are, by their nature, unable to understand the citation practices established in an article (WP:CITEVAR), and end up acting in ways (such as repeatedly editing an article to change its established citation style) which would see a human editor criticised or sanctioned. 2B, 2C and 2D would all make this problem worse. UndercoverClassicist T·C 20:21, 4 April 2025 (UTC)
    So then if a bot was hypothetically able to determine the "citation practices established in an article" by assessing whether the citations fully or at least consistently (greater than 50%) included reflinks, you would endorse it? Οἶδα (talk) 22:47, 4 April 2025 (UTC)
    It would be easy for the bot to determine a prevailing practice of reflinks: parse all the templates and count how many have links. A bot could be more aware and consistent of prevailing practice than humans. BTW in all my years making millions of edits, not a single editor has ever complained of an edit war, it's not that kind of bot constantly running unattended across 6 million pages. It's targeted based on requests for certain domains only, and I don't usually repeat the same domain. -- GreenC 00:32, 5 April 2025 (UTC)
    I suspect that's because 2B, as you note at the top, includes the conversion of the domain name to the work name, which is a useful thing to do, and because it's relatively low volume. If I had seen one of those edits on an article for which the link contradicted an established consensus, I would not have reverted; I'd have just unlinked. Mike Christie (talk - contribs - library) 09:30, 5 April 2025 (UTC)
  • Option 2B or 2D: as other editors above have highlighted, linking within sources can be useful; I don't think whether or not something is linked is really an citation style issue so much as many editors use automated tools for creating citations to save time which may or may not include a link for them leading to inconsistency. The fundamentals of the citation format doesn't change (ie. WP:LDR vs in-line with the visual editor) by improving existing citations with links. I've mostly seen requests for consistency (ie. either all sources link or all sources don't link) in good/featured article reviews so having an option for the bot to convert one direction or the other on demand would be useful. Sariel Xilo (talk) 21:20, 4 April 2025 (UTC)
  • GreenC, would it be possible for the bot to include an option for the requesting user to ask for all links, no links, or as you suggested above to follow the prevailing practice? That would ensure that the decision is always left to editor discretion rather than being a bot default. MichaelMaggs (talk) 04:01, 5 April 2025 (UTC)
    eh? It wouldn’t be editor discretion, would it? That would be a single editor’s personal preference enforced across several hundred or thousand articles at any one time, regardless of the local consensus at each individual page. - SchroCat (talk) 04:11, 5 April 2025 (UTC)
    With such an option, the editor has discretion to instruct a globally approved bot to follow existing prevailing practice on every page, in perfect compliance with the MOS. Though I’m not expecting that even that will be enough to change your mind, it does at least dispose of all the arguments you have enunciated thus far. MichaelMaggs (talk) 08:26, 5 April 2025 (UTC)
    Sorry, but that's nonsense, and it disposes of absolutely nothing. If there is "an option for the requesting user to ask for all links", it enables a single editor to overrule the status quo on hundreds or thousands of individual pages. That's ridiculous. It would be akin to an editor ordering a bot to add (or remove) every serial comma to their own preference, or change language parameters - not just to one article but to thousands. And that's before you even think about what happens when Editor A asks for links to be added and Editor B comes along with the next request and exercises "an option for the requesting user to ask for ... no links" - ie, asking for them to be removed? This isn't a question that can be determined by this RfC - it would need a more fundamental change of the MOS before it even comes close to this. - SchroCat (talk) 08:37, 5 April 2025 (UTC)
    If I understand correctly, you're not objecting to Option 3: the bot is changed so that it always follows existing prevailing practice on every page. MichaelMaggs (talk) 09:14, 5 April 2025 (UTC)
    There is no option three at present. - SchroCat (talk) 09:26, 5 April 2025 (UTC)
  • I propose the discussion of a new Option 3A: that the bot is changed so that it always follows the existing prevailing reflink practice on every page. GreenC has stated above that this would be easy to program, and it avoids the objection of some contibutors that the editor who instructs the bot could potentially be overriding local page consensus, where that exists. MichaelMaggs (talk) 11:11, 5 April 2025 (UTC)
    I think we should have a clear definition of what the set of "existing prevailing reflink practices" are before this can be considered. Also, a problem I see is that if the practice is being consistently followed already at a given page, there's nothing for the bot to do; and if it's not being consistently followed, it can't determine what the prevailing practice is. It would not be OK to then assume there is no prevailing practice, since a recent edit might have rendered the page inconsistent and not yet been reverted. Mike Christie (talk - contribs - library) 11:29, 5 April 2025 (UTC)
    I am a bit confused by the different uses of "prevailing practice" and "consistently followed" here. I believe you're saying local page consensus might not be reflected in the current revision of an article due to a recent edit. Yes, each article can have their own documented consensus for reflinks and the bot needs to account for that. But in reality, they typically do not. It hasn't exactly been a very common or contentious phenomenon from what I can find. An article's current makeup should be enough for the bot to run. If a revert is already needed then the bot can be reverted as well, at which point the bot will not perform that same edit. If anything, I was more wondering if the bot could account for pages where the reflink style is MOS:LINKONCE. Without mistakenly linking twice, for example. Οἶδα (talk) 05:27, 7 April 2025 (UTC)
  • 2D Per Dhtwiki, when the work is wikilinked, I immediately know that the source is notable, and I can read its article to determine reliability before heading off-wiki. When the work is simply a URL, I am potentially left confused between sources with similar names or wary of heading to an unfamiliar site. When MOS:REPEATLINK exists to avoid a sea of blue in the article text, I agree with GreenC that the current guideline that "citations stand alone in their usage" justifies this new wikilinking function. While I can understand requesting the bot to not wikilink in citation templates that do not support it, the appeal to WP:FACRITERIA is maddening. WP:CITE's discussion of consistency in citation styles explicitly refers to the big choices over templates, not whether some works are wikilinked and some are not, even stating that "the data provided should be sufficient to uniquely identify the source, allow readers to find it, and allow readers to initially evaluate a source without retrieving it." Thanks for maintaining this useful bot work, GreenC! ViridianPenguin🐧 (💬) 16:05, 5 April 2025 (UTC)
  • 2B My personal preference would be 2A, as I dislike the sea of blue that overlooking in references can cause. However other editors appear to find such linking useful, and my distaste of it is not enough to impede other editors. The flip side of that though is that high volume edits such as 2C/D also impact editors, so the lower volume of 2B seems like a sensible compromise. -- LCU ActivelyDisinterested «@» °∆t° 20:44, 5 April 2025 (UTC)
  • 2A per comments above. If someone prefers the "link each one" style, go for it, but the default bot-level option should be the safest option. If a reader is curious about a reference, we generally want them to click on a URL of the article itself, not an article about the work it was published in. There are times when adding such links is good, but let humans do that, not bots. SnowFire (talk) 04:11, 7 April 2025 (UTC)
  • 2B In a multicultural English language encyclopedia, linking to the Wikipedia article for the publication is a benefit for users of this Wikipedia. I know it would be for me when I check a citation to an unfamiliar publication. — Neonorange (talk to Phil) (he, they) 14:53, 8 April 2025 (UTC) —
  • 2B Others have said it better, but I think having the publisher linked to its WP article is good for the encyclopedia, even it I personally don't do so consistently when creating citations. - Donald Albury 15:52, 8 April 2025 (UTC)
    There is no option to link to the publisher and no-one has suggested that would be a beneficial step. That is not what this RfC is about at all! - SchroCat (talk) 04:26, 9 April 2025 (UTC)
  • 2B citevar is not an excuse for avoiding doing something that has a clear, tangible benefit to any reader. PARAKANYAA (talk) 02:32, 9 April 2025 (UTC)
  • 2D or 2B. Wikilinks are very useful in giving context to the citation. I've also found that when the values aren't linked, then there are often times typos in his field which are left uncaught because they aren't linked.
Gonnym (talk) 07:20, 9 April 2025 (UTC)
The issue you are complaining about won't be resolved by a bot adding wikilinks to correctly spelt titles. Traumnovelle (talk) 09:26, 11 April 2025 (UTC)
Yes they would. Linked redirects that are tagged with {{R from misspelling}} appear on Wikipedia:Database reports/Linked misspellings, which then can be fixed. Gonnym (talk) 07:57, 13 April 2025 (UTC)
The bot wouldn't add wikilinks to misspelt names unless it was coded to recognise them, in which case it may as well just fix the spelling. Traumnovelle (talk) 23:07, 13 April 2025 (UTC)
  • 2A definitely oppose 2C/2D. I've only ever once clicked on a wikilink to a work in a reference and it was an accidental click when I was aiming for the url. I have never found these useful, some readers may find them useful but I do not believe their usefulness justifies the the enormous amount of edits this bot would be undertaking to do so. Traumnovelle (talk) 09:23, 11 April 2025 (UTC)
  • 2B – everything should be wikilinked. As Wikipedians, it should be obvious that wikilinks are useful – I use wikilinked source names all the time, for instance when trying to figure out the ownership and biases of the many Hong Kong newspapers. The easiest, most consistent, and only sustainable solution is to link every reference. Linking the only first instance means that whenever the reference order is changed (a very common occurrence) the link has to be moved, which is busywork that nobody should actually bother doing. Toadspike [Talk] 09:30, 11 April 2025 (UTC)
  • 2D with the exception of not converting |work=[url name] to a wikilink. -- Shmuel (Seymour J.) Metz Username:Chatul (talk) 13:28, 11 April 2025 (UTC)
2B - as a reader, my preference is for the work to be wikilinked whenever it can be. This link leads me to an article where I can learn more about the publication and its reliability, which is a crucial and core service of Wikipedia. Helping readers assess the reliability of a work helps readers assess the reliability of a citation and thus the accuracy of a Wikipedia article.
The URL to the work is not a substitute for information about the publication; more information about the publication will often be available in a Wikipedia article about the publication than on the publication's own website (which will inevitable have a pro-publication bias).
Wikilinking the work field in every reference is important, even though that results in repeated wikilinks in the reference section. To understand why, first remember that the vast majority of readers read on a mobile phone. Next, take out your phone and browse (in default mobile mode, not desktop-on-mobile mode) to any article and click on any reference. Note how it just shows you the one reference you clicked on--it doesn't take you to the references section the way desktop mode does. If that one reference you clicked on doesn't have the work wikilinked, you won't see another wikilink on your screen. Even on desktop mode, in articles that have hundreds of references (which most highly read articles do), if the one you clicked on doesn't have a wikilink to the work, it can be hard to find that link amongst the hundreds of other references listed. For both mobile and desktop users, it's important that the work field be linked in every citation. Levivich (talk) 15:46, 13 April 2025 (UTC)
  • 2A would be preferable to prevent further fighting. I appreciate the attention to cleaning up citations, but it would be jarring to introduce linking inconsistency to an otherwise stable article, all at the hand of a bot without a human to sign off on every edit. SounderBruce 00:10, 14 April 2025 (UTC)
  • 2B I find these links useful, especially for lesser known publications. Helps to click them into new tabs quickly especially when reviewing an article, rather than searching for each name individually. seefooddiet (talk) 03:14, 22 April 2025 (UTC)
  • 2B weakly (as it's generally useful, but not all articles that the link might go to are of equal quality), but I do object to the idea that a reversion of the wikilink added by bot could be seen as disruptive based on this RfC outcome. There is a difference between developing a consensus that says it's fine for a bot to add a link in the reference as it's usually more useful than not, rather than having a consensus that states that references should have a link. We are discussing the former, as the RfC question isn't formulated to address the latter and I also don't see the necessary advertising of the RfC (i.e. notifications to relevant MOS pages). In practical terms, anyone who wants to revert the bot addition would merely need to make a reasonable case that the link doesn't benefit the article (BRD) rather than meeting the bar of being an exception to something that editors should "generally follow" as defined in a community endorsed guideline. Scribolt (talk) 07:57, 23 April 2025 (UTC)
  • 2A to make the references more human readable, but do not indiscriminately add wikilinks by bot without general agreement on citation style. —Kusma (talk) 12:38, 23 April 2025 (UTC)
    It occurs to me that this would change |work=whitehouse.com and |work=whitehouse.gov to the same |work= parameter; they used to be very different. -- Shmuel (Seymour J.) Metz Username:Chatul (talk) 14:56, 23 April 2025 (UTC)
  • 2B: MOS:REFLINK allows it so the bot should do it. It's rote grunt work that human editors might not do out of time, so it's the right job for a bot. The whole point of Wikipedia is to be an encyclopedia of hypertext so we should use as much of it as makes sense. Sure it might run afoul of CITEVAR one way but as long as a bot is fixing bad citations, it's better to go overboard than underperform its task. Dan Leonard (talk • contribs) 19:37, 28 April 2025 (UTC)
  • 1 or 2A, and I am strongly opposed to 2C or 2D. Bots should aim for minimal disruption, wikilinking every periodical title is generally distracting and unhelpful but the links can be left if they are an established convention on some page where the local editors prefer them. Bots should not be otherwise disrupting established styles on other pages where non-linked names are preferred or where the links are added by discretion. –jacobolus (t) 16:34, 8 May 2025 (UTC)
Retrieved from "https://en.wikipedia.org/w/index.php?title=Wikipedia:Village_pump_(proposals)/Archive_218&oldid=1291492604"