Wikipedia:Bots/Requests for approval/RjwilmsiBot 8


 * The following discussion is an archived debate. Please do not modify it. To request review of this BRFA, please start a new section at WT:BRFA. The result of the discussion was Symbol delete vote.svg Denied

RjwilmsiBot 8
Operator:

Time filed: 18:19, Wednesday May 30, 2012 (UTC)

Automatic, Supervised, or Manual: Automatic

Programming language(s): AWB, custom module

Source code available: AWB

Function overview: Format endashes & diacritics in wikilinks by selected bypassing of redirects to correctly format wiki text.

Links to relevant discussions (where appropriate): Standard styling per WP:MOS

Edit period(s): On release of database dump

Estimated number of pages affected: Thousands, exact number not yet known.

Exclusion compliant (Yes/No): Yes

Already has a bot flag (Yes/No): Yes

Function details: Following this Dash fixing task request, this bot task is to format wikilinks that are to redirects either for the hyphen form of a title with an endash, or for the non-diacritic form of a title with diacritics. The purpose is to improve the consistency and formatting of the wikilinks. Changes will only be made where the rendered wikilink changes, so piped wikilinks will not be changed. The wikilinks/redirects that will be changed are those that redirect to a page where the title is the same apart from either endashes instead of hyphens, or diacritics instead of Latin characters. The logic will not cover other redirects, so for example a redirect with a hyphen->endash change and a capitalisation change will not be picked up or changed.

Examples of changes:
 * Wikilink blood-brain barrier (hyphen redirect) will be changed to wikilink blood–brain barrier (actual page title with endash).
 * Wikilink Blood-brain barrier (hyphen redirect) will be changed to wikilink Blood–brain barrier (actual page title with endash), keeping first letter case.
 * Wikilink Bjork (no diacritic) will be changed to wikilink Björk (actual page title with diacritic).

Examples of no changes:
 * Wikilink BBB (piped link to hyphen redirect blood-brain barrier) will be not be changed, page rendering would not change due to piped link.
 * Wikilink Blood-Brain Barrier (hyphen and capitalisation redirect) will not be changed, more than a hyphen/endash format change.

Task will apply on mainspace and template namespace.

Discussion
Why not the portal namespace as well? -- John of Reading (talk) 19:34, 30 May 2012 (UTC)

Strikes me as a possible cosmetic bot - who'll notice the changes and it redirects anyway.  Rcsprinter  (yak)  16:22, 1 June 2012 (UTC)
 * In regards to bots, doesn't "cosmetic" mean you only see the difference in the edit window? I support the visible text in links agreeing with target articles' titles, which is what this bot appears to do. I think those redirects are great for getting search requests to the right places, but we should be using the correct presentation in articles. — Bility (talk) 18:20, 1 June 2012 (UTC)

This seems like an extremely cosmetic task. I move to deny. -- Chris 03:10, 2 June 2012 (UTC)


 * Is there a definition of "cosmetic" that we need to consider? If it means only affecting how the source looks, then this is not cosmetic.  These are the sorts of changes that I do a lot of, because articles with blood-brain barrier just look wrong to me.  Many editors respect and appreciate the information that correct typography conveys, but not many care enough to seek out and fix such errors.  This bot would do it nicely.  For those who don't notice the difference, I suppose it will be considered cosmetic, but there's no reason they would be bothered by the changes.  Dicklyon (talk) 18:43, 2 June 2012 (UTC)
 * It just strikes me as a hell of a lot of edits, for little to no benefit. -- Chris 09:32, 8 June 2012 (UTC)
 * Getting hyphens and en-dashes right is part of the accuracy people have a right to expect from an encyclopaedia: it's not just cosmetic. I do a lot of this kind of editing in the course of doing other stuff: a bot to do it would save a lot of fiddly manual work and would be reaping the benefit of all the work other editors have put into creating and categorising redirects. Colonies Chris (talk) 10:18, 9 June 2012 (UTC)
 * "in the course of doing other stuff" is crucial. Fixing those dashes as a sole task will spam watchlists and cover up vandalism. I oppose. Such tasks must be combined with more substantial ones. Materialscientist (talk) 23:27, 9 June 2012 (UTC)
 * This argument that small fixes will 'spam watchlists' comes up occasionally. The edit summary left by a bot will be clear about what it's doing, and watchers can choose to exclude bot edits from their view if they want, so they needn't be troubled by it at all. The propoosed bot has the potential to enormously improve the quality of the encyclopaedia overall by many small increments. Colonies Chris (talk) 08:57, 11 June 2012 (UTC)
 * Edit summaries are not an issue. There are two problems: (i) I do not want to exclude other bots from my watchlists at any time; (ii) bot edits cover up other edits. Materialscientist (talk) 09:20, 11 June 2012 (UTC)


 * This looks like a very good idea. Professional standards of typography – as insisted on my all of the major style guides in English, and our own in-house style guides, brings with it added layers of meaning and greater readability. The correct use of hyphens and en dashes is central. I'm pleased that this bot will fix countless inconsisencies in this respect; I support the application. Tony   (talk)  07:34, 20 June 2012 (UTC)
 * One word of caution with respect to diacritics, though. I believe there is a local agreement within North American ice hockey articles that players' names are given without diacritics (in conformity with the usual practice in North America), even though the player's own article's title may carry the diacritics. So probably any articles in Category:Ice hockey in Canada and Category:Ice hockey in the United States‎ should be excluded. Colonies Chris (talk) 15:27, 20 June 2012 (UTC)

As others have noted, this is entirely cosmetic, and it could result in unnecessarily making edits that contrast with WP:USEENGLISH and the neutral approach of WP:DIACRITICS. Furthermore (and in perspective), you're asking for approval on a bot that would essentially be authorized to convert wikilinks to their final target's article name; it's just incidental that for now it would only be doing it articles linking to pages with dashes and diacritics. Because the technical capabilities of redirects deprecate the necessity of doing such a thing, and because the primary benefit is to OCPD editors (as opposed to readers), we haven't historically approved bots for this purpose; hence WP:COSMETICBOT. -- slakr \ talk / 21:36, 20 June 2012 (UTC)
 * As I and others have noted, this is not a cosmetic bot: endash-hyphen usage is both visible and has significance - there is substantial guidance at MOS:HYPHEN on the proper use of each them. This bot would be bringing articles up to the standard recommended by the MoS. If you think the distinction is not important, you should propose changes to the MoS, not argue it here. As for redirects - the statement that "it's just incidental that for now it would only be doing it articles linking to pages with dashes and diacritics" utterly mistakes the nature of the proposal. The proposal is about improving the text visible to the reader - redirects of certain categories and properties are the means to identify the changes to be made, and bypassing some of those redirects is a byproduct, not an end in itself. It is fundamental to this proposal that it operates only on certain very specific classes of redirects, where human editors have created and categorised redirects to provide for convenience in typing. Nor do any of the valid arguments against bypassing redirects apply to this proposal (it's not being done for spurious efficiency reasons, and it wouldn't bypass redirects with possibilities). This is an exciting proposal that could massively improve the quality of the encyclopaedia by building on the foundation that underlies Wikipedia - the power of many editors working together to create something better than any one person could achieve. Colonies Chris (talk) 09:07, 21 June 2012 (UTC)
 * The problem is that this actually might not taking the entirety of our guidelines into account, and you might be fundamentally overruling them with your own interpretation and enforcement of English style guidelines&mdash;especially portions that defer to recognizeability and naturalness, as well as accessibility of the editing environment (e.g., whether characters even appear on a standard keyboard). One example of this is the point raised by Colonies Chris, above, with regard to hockey names. Furthermore, while the most commonly occurring instances of something may be with the "wrong" dash or the "wrong"/unaccented letters, you're proposing we ignore that so that everything uses what English grammar nazis believe to be the "right" dash or accented character. I have problems with this and especially with using wikilinks as a WP:BATTLEGROUND for "good English" as opposed to the English that readers and editors use&mdash;and I definitely have problems with automating that battle.
 * Take, for instance, "Blood-brain barrier." Google it. Wikipedia is in the underwhelming minority to use the endash version, and some even drop the hyphen altogether. Google scholar does the same thing, with basically no consensus as to the de facto "right" version. You're proposing a bot that would establish the endash version above the hyphen version even in wikilinks, which, while it might be the "right" dash in your philosophy, doesn't seem to have clear consensus in the wild.
 * Long story short, I fail to see how the editing experience of users could be anything but diminished by standardizing wikilinks to use dashes and characters that don't even appear on a standard English keyboard. As such, I can't imagine how this bot would be anything other than cosmetic until the ndash, mdash, and the various diacritics are standardized on English keyboards and are in wide acceptance and use by the people who most frequently visit and, most importantly, edit the encyclopedia&mdash;that is, people who didn't major in English or foreign languages, and people who aren't professional printers.
 * -- slakr \ talk / 02:34, 26 June 2012 (UTC)

I am gauging consensus and closing this request today as a contributor who, personally, would prefer seeing higher standards for typography in English Wikipedia articles. However, it's objectively clear that consensus doesn't currently exist for a bot to change all wikilinks in the article namespace to meet such high standards. There are legitimate concerns that these cosmetic changes (and "cosmetic changes" does not mean "invisible changes") do not meet the threshold of usefulness required for bot approval. There are also concerns (that rate more highly in my mind) that these changes may impair searching and linking in the future. Indeed, this is the reason Manual of Style recommends creating redirects sans diacritics and with hyphens in the first place. It's also important to remember that the Manual of Style is a guideline; it acknowledges that more than one style is acceptable and its existence does not compel changing articles' styles to a style with which its contributors may be unfamiliar. If in the future consensus were to be reached elsewhere (e.g., at WP:VP/P or WT:MOS) to make such a sweeping change throughout the article namespace then this bot request could be reconsidered, but for now it cannot be approved. Thank you, &mdash; madman 04:21, 26 July 2012 (UTC)
 * Assuming the bot will function as described under 'function details', I wholeheartedly support this task. This is obviously not a cosmetic task as the page output will be altered. As for those who are opposing because they don't like dashes or they don't like diacritics, this is clearly not the right place to discuss your opposition to such things. Jenks24 (talk) 12:26, 12 July 2012 (UTC)
 * The above discussion is preserved as an archive of the debate. Please do not modify it. To request review of this BRFA, please start a new section at WT:BRFA.