Wikivoyage talk:Stubs needing attention

Do we still need this page?

Do we really need this page anymore? I mean you can just go to Project:Stub articles and click on "What links here". Which, now that I test it, seems an awfully short list. Is it broken, or is there some weird limit? -- (WT-en) Nils 03:07, 16 Apr 2004 (EDT)

Transclusion with the {{msg:stub}} thing doesn't make a backlink connection. So, yeah, I think we still need this. --(WT-en) Evan 09:18, 16 Apr 2004 (EDT)

Actually, going to Template:Stub and clicking what links here? pretty much gets you where you want to be.... except it excludes stuff that says {{msg:stub}} instead of {{stub}}. But I'm thinking that this hierarchical display of stuff needing work is useful anyway since I could quickly find stuff near me that I can help with. -- (WT-en) Colin 01:30, 31 Mar 2005 (EST)

Yes! We absolutely need this page. This page lists all the articles people think should be stubs. It guards against the possibility that someone removes the {{stub}} message from a page. If we did not have this page then you would need to troll through recent changes to find articles that had recently had their {{stub}} notices removed. Or through short pages, oldest pages, dead-end pages, because you cannot find stubs that do not have {{stub}} notices any other way. The geographical organisation also quickly indicate where work is needed or where people are active. -- (WT-en) Huttite 04:54, 31 Mar 2005 (EST)

How long does an article need to be to lose 'stub' status?

Moved from Project:Travellers' pub.

Self-explanatory, really. I was thinking this because articles that are really long (such as Prague, Bratislava and Boston) have stub messages on them. Stub messages, surely, would be more suitable if the article was just 'Whatever is a city in Wherever.'

Do stub messages stay until all the template headings have been filled? It seems a bit odd to have a stub message on a 30k+ article. Would a message along the lines of 'This article is in need of development' be more suitable? (WT-en) Professorbiscuit 13:19, 11 Aug 2004 (EDT)

It's not a matter of length, but completeness. A destination guide is a stub until it would make a useful travel guide. Until then, it's incomplete, i.e., a stub. A useful travel guide should at least have Get in, Get around, Eat, and Sleep stuff available. See Project:Stub articles for more info. If you disagree with the definition of stubs that's there, which we've used for more than a year, please comment on Project:Stub articles. --(WT-en) Evan 16:43, 11 Aug 2004 (EDT)

So why is for example The Marina no stub anymore? Would the Marina being incomplete make San Francisco overall be incomplete and hence a stub (As the Marina is part of SanFran and it being incomplete makes the overall SanFran site incomplete...)?

Evan's comment may have been right at the time, but currently Article_status defines Outline in a way that would cover such cases, has a template and some info but is still incomplete. (WT-en) Pashley 21:09, 23 June 2006 (EDT)

Programmers gather and hear the horrible news

Everytime I type {{msg:stub}} a stub message appears, but the article name does not appear in the "Stubs needing attention" article? Can the programmers of the wikiworld fix this? Many thanks from those who overlook your wisdom. (WT-en) Sapphire 00:46, 19 May 2005 (EDT)

Yes, this really needs to be automated. Manual maintenance is obviously failing. I just removed Istanbul from the list; it is currently at Guide status but was still listed here. Shiraz was at Usable, but here. There are probably lots more, also many stubs not listed. I'd say the page is inaccurate enough to be almost useless, and a fair bit of work to maintai. If it cannot be automated, scrap it. (WT-en) Pashley 20:55, 23 June 2006 (EDT)

Yes, I stopped trying to keep that list updated because it was too much work that should really be automated. — (WT-en) Ravikiran 23:32, 23 June 2006 (EDT)

Isn't the only working approach to look at (for example) Template:Stub and see what links to it? I'd be fine with getting rid of all of the "Stub/Outline/Whatever needing attention" pages provided editorial comments (if any) about what actually needs attention are first copied to the article talk page. I'm not sure that these pages are really of any practical use anymore. -- (WT-en) Ryan 04:02, 24 June 2006 (EDT)

Should we just delete all of these lists? - (WT-en) Andrew Haggard (Sapphire) 06:55, 24 June 2006 (EDT)

That headline really gets to me! I wonder if I added it or did someone change it? Geez, that was a good laugh. - (WT-en) Andrew Haggard (Sapphire) 06:59, 24 June 2006 (EDT)

Delete the Stub/Outline/Etc Articles Needing Attention?

Continuing the discussion above, the Project:Stubs needing attention, Project:Outline articles needing attention, Project:Usable articles needing attention, Project:Guide articles needing attention, Project:Factbook imports needing attention page are horribly out-of-date and not particularly useful. A more accurate way to figure out what stub articles are out there is to look at (for example) Special:Whatlinkshere/Template:Stub. So in the interest of fixing broken windows, should we just get rid of these pages? They are out-of-date, difficult to maintain, and confusing for new users who stumble on them. Thoughts? -- (WT-en) Ryan 13:43, 30 June 2006 (EDT)

How does that first step go... "We admitted we were powerless to maintain these pages, and that they had become unmanageable"? Something like that. With almost 10 kiloarticles in the system, nearly all of them in some state of "needing attention", these pages have outlived their usefulness. - (WT-en) Todd VerBeek 14:26, 30 June 2006 (EDT)

Thinking about this further, it seems like it would be a Herculean task to get rid of all of the existing links to these pages, so rather than deleting how about simply changing (for example) the content of Project:Stubs needing attention to be #REDIRECT [[Special:Whatlinkshere/Template:Stub]]? That turns the page into something more dynamic for those who want to use it, and saves us the trouble of tracking down and orphaning all of the links to this page. If there aren't any objections raised here I'll propose this on Project:Votes for deletion where it's likely to get more attention. -- (WT-en) Ryan 23:19, 30 June 2006 (EDT)

The vast majority of links to these pages are in the status templates; those links would be easy to remove or direct elsewhere. - (WT-en) Todd VerBeek 11:19, 4 July 2006 (EDT)

I disagree strongly. This is a bot just waiting to happen (to be honest, I'm surprised it hasn't happened already), and when it does get created, the resulting, regularly updated pages will be a fine way for the casual editor to find things to work on. This would be particularly true for the Usable and Guide articles, where recently there has been a trend for people to include language on just what needs to be changed to advance their status. -- (WT-en) Bill-on-the-Hill 18:30, 2 July 2006 (EDT)

I don't understand what kind of bot you imagine doing this, or what exactly it's going to do. Would it detect that an article no longer needs listings edited for MOS compliance or that it no longer needs proofreading, or that it now has a map, and take if off the list? Would it add articles to these lists (because almost no one is doing that now)? If not, then these pages will still be fundamentally broken and useless. - (WT-en) Todd VerBeek 19:28, 2 July 2006 (EDT)

No! The advantage of a page like this over the Whatlinkshere page is that this page is organized by geographical hierarchy while the other page is not. I can periodically go through all the stubs under "India", check out which ones aren't stubs any more, and update the status. Or I could find a page in usable status and learn that a bit of googling and MoSing will help me make it a guide. So all that the bot needs to do is to periodically go through the article status, read the article's isIn, and keep these pages updated. Additionally, we could add a parameter for comments in the template tags, e.g. {{outline|Needs "sleep" listings}} and the bot could read those comments and put them on these pages next to the article. — (WT-en) Ravikiran 03:27, 3 July 2006 (EDT)

OK, maybe you could go through this page and remove stubs from India on a regular basis, but how often do you? I don't see much of anyone else updating this page either, mostly just sporadic edits by people who think maybe they're expected to. (Addendum: I just looked at Project:Outline_articles_needing_attention, and it tells me that India and a few western United States are the only regions with more than a handful of outline articles in them (!!!), so apparently you and Bill are maintaining those sections. But quite obviously no one else is.) More importantly, I don't see the kinds of edits that indicate to me that people are using these indexes... just a couple regulars doing some maintenance to reflect status changes that probably happened because someone came across an article and improved it, rather than because it was listed here. Furthermore, if someone actually built an artificial intelligence engine that would compile hierarchical lists of non-star articles, the results would be huge. There are over 6000 "outline articles in need of attention" on this site; as an editor I'd find that list more off-putting than helpful. Your suggestion that we add comments to the status tags is just a different form of manual maintenance; I don't see that happening either (and all those comments would make the indexes even more huge). Before we talk about how to build a bot to keep the Titanic here afloat (or more accurately, build a whole new Titanic, because this one's sunk), let's first answer the question of why? Does anybody really use this page like you describe, or are we wasting human (and potentially CPU) resources to maintain it for its own sake? After all, if I want to see what articles need attention in India, I just look at India. That hierarchical list is already being maintained (and is actively being used); why build another? - (WT-en) Todd VerBeek 09:24, 4 July 2006 (EDT)

You've misunderstood me. I don't think that humans can keep these pages updated. Only a bot can, by reading the statuses and the isIns on pages, i.e. if someone upgrades an article from "stub" to "outline", the bot will remove that article from here and put it in Project:Outline_articles_needing_attention. I used to keep the pages updated, but I gave up because it was not worth the effort. I was pointing out how this page can be practically useful. I just want to be a consumer of this page, not a producer. — (WT-en) Ravikiran 10:17, 4 July 2006 (EDT)

Ravikiran, you're idea for a bot sits well with me. Would you be able to write one up to do the job? -- (WT-en) Andrew Haggard (Sapphire) 10:20, 4 July 2006 (EDT)

Nope. not me. I am a techie, but I will have to learn at least one new programming language, something I don't have time for right now. — (WT-en) Ravikiran 03:58, 5 July 2006 (EDT)

Todd, are you in favor of deleting these lists? -- (WT-en) Andrew Haggard (Sapphire) 09:49, 4 July 2006 (EDT)

FYI: I just checked the Whatlinkshere for stubs and there are only about 2,050 stubs. -- (WT-en) Andrew Haggard (Sapphire) 09:52, 4 July 2006 (EDT)

Two points here. First, I don't see a humongous list of Outlines or Stubs as a completely intractable problem. The resulting lists don't have to be 2,050 lines long, just 2,050+ words long (since all entries that are "leaves" under a country- or state-level heading get bundled together). We handle articles that size all the time. And in any event, if the complete lists need to be broken out by region ("Wikivoyage:Stubs needing attention/North America" and such), that's not difficult.

Second, I believe it should be possible to create a bot that handles the initial population of the lists, as well as one that "sweeps" occasionally to do updates. The startup bot simply looks at an article's status; looks at its breadcrumbs; finds the lowest-level ply of the breadcrumb that corresponds to an entry in the "(foo) needing attention" page; and updates that ply on the page. If there is no breadcrumb for the article, put it in a "Needs a home" section of the needing-attention page.

I really feel strongly that these lists should be kept. Yes, a better way of maintaining them is needed. I don't see it as all that difficult -- but then, I'm not a programmer (at least not a wiki programmer). -- (WT-en) Bill-on-the-Hill 10:45, 4 July 2006 (EDT)

My main question remains unanswered: Even aside from the user-unfriendly bulk of complete, auto-generated lists (Ravikiran wants them populated with comments, which means they would be an imposing 2000+ and 6000+ lines long), and the resources to build them: WHY? Two of you seem to be assuming that the reason (almost) no one's using these indexes is because they're not maintained; it seems to me that the reason they're not maintained (including why no one bothered to build a bot to automate that) is because people didn't use them. Not even with a bold, italic link to them on every single page with a status template! The lists were a good thing to try, but they failed. Meanwhile we have a functioning, well-maintained geographical hierarchical index of articles needing attention, and people are using that to find articles to work on. - (WT-en) Todd VerBeek 11:10, 4 July 2006 (EDT)

Just to clarify where I sit on this issue: 1. Todd is right that comments following an article doesn't make much since. 2. no one was using the lists in the first place (that's why they weren't being maintained). 3. Having a 2000+ list of stub articles is a bit ridiculous.

That said these lists be more useful provided they are maintained and maybe a link to each list on the project page would increase visibility. I'm leaning toward deleting these lists unless someone really wants to write the bot. I don't think we need a guides needing attention list, because a guide must be fairly complete and could with minimal work be a star. -- (WT-en) Andrew Haggard (Sapphire) 11:25, 4 July 2006 (EDT)

I'm trying to follow the arguments being made here, and I think they boil down to:

KEEP:

1. If a bot was written it would provide an automated way of tracking these articles.
2. The geographical hierarchy is useful for someone looking for articles within a region that need work.

DELETE:

1. Even with a bot, most people aren't actually using these lists.
2. The list would be huge (6000+ articles for "outline").
3. New users may find these lists and think they are expected to use them somehow. That's misleading since they are not really maintained.

I obviously agree with Todd that these lists just don't seem that useful. If there was a bot that automatically kept them up-to-date then it probably wouldn't hurt to have them around, but in their current form I feel like they do more harm than good - at best it makes it look like this Wiki is unorganized, with pages that are hopelessly out of date, and at worst it confuses people who wonder why an article's actual status doesn't match what's found on this list. To address the two "keep" arguments above, if/when a bot is written we could obviously recreate these pages. Until then, for those who do want to maintain lists of articles to work on within a geographical hierarchy the appropriate list section could be copied to a talk page, such as Talk:India, where it might be more likely to be updated and seen by those working on articles within the region. Would that be an acceptable way to address the outstanding concerns? Did I miss any arguments for or against keeping these lists? -- (WT-en) Ryan 13:34, 4 July 2006 (EDT)

Sounds good to me. -- (WT-en) Andrew Haggard (Sapphire) 13:39, 4 July 2006 (EDT)

I suppose that would work, but I'm still having problems with the "what problem needs fixing here?" test. I doubt that 99% of WT readers even know there's an issue with these pages that needs "fixing" one way or another, and under the circumstances, why mess with it? The more so since some small number of participants do look at the needing-attention pages and take occasional actions. But yeah, re-creation if/when there's a satisfactory bot would satisfy my interest in having the capability, more or less. -- (WT-en) Bill-on-the-Hill 18:18, 4 July 2006 (EDT)

The "problem that needs fixing" as I see it is that we've got pages that are so out of date and difficult to maintain that they provide no real benefit, but new users who stumble on them think they are important and may waste time trying to update them or (worse) figure out why an article marked as "usable" on the article page is listed as an "outline" here. I realize that I may not be the sharpest tool in the shed, but when I first started editing on Wikivoyage I eventually clicked on the "What links here" link for one of my articles and found these "needing attention" articles. Being a new user I figured that I had "screwed up" by creating new articles without adding them to the proper "needing attention" page, and thereafter I spent a good deal of time trying to do things "properly". It was only much later that I realized that most people don't update these pages, and they therefore aren't all that useful. If the pages aren't useful in their present state, why keep them around and confuse people? I agree with you that if the process was automated then they might have some value as a "to do" list sort of thing, but at the moment I see them only as a point of confusion for new users and a bit of an embarrassment for those of us who work on the site regularly. -- (WT-en) Ryan 19:09, 4 July 2006 (EDT)

I agree that if it can't be automated, these out-of-date lists shouldn't be there. I am okay with your solution, but I am also going to strongly push for these things to be automated. — (WT-en) Ravikiran 03:58, 5 July 2006 (EDT)

I think they should be automated too. I'm thinking something like Special:Allpages, but with article status in a dropdown rather than namespace. It might also be nice to query by geographical area (so you can see which articles need work under Montana or Uganda or Mongolia), and even by article type (destination guides, travel topics, itineraries, phrasebooks) and destination type (city, region, country). I think that we'd need to put a little RDF into the status templates (see Project:RDF Expedition/Article status for an example), and I have to do some architectural work to make all-site searches possible. But I think it's doable, and I think it would help out a lot. --(WT-en) Evan 18:14, 8 July 2006 (EDT)

Can we at least edit Template:stub, Template:outline, and the various usable* and guide* templates so they no longer direct people to these "needing attention" lists? (Those first two are used on thousands of articles each, so they should be done at separate times – preferably when the site isn't being used heavily – to avoid too big a hit on server performance.) - (WT-en) Todd VerBeek 10:28, 8 July 2006 (EDT)

It sounds like there is some agreement that these lists aren't valuable in their present state, so yeah, it makes sense to start getting rid of some of the links to the pages. I don't think the cache automatically clears for templates changes (quickbar template didn't cause cache clears) so you can probably just make the change. -- (WT-en) Ryan 14:57, 8 July 2006 (EDT)

The cache for articles that link to a template are supposed to change when the template changes, but... my guess is that multiline templates like the quickbar screw up my query. (By multiline, I mean that the template's use stretches across multiple lines, not that it contains multiple lines.) It's definitely supposed to work like that, though. With the article status templates, it totally should. If it doesn't, I can purge the cache by hand if needed. --(WT-en) Evan 18:17, 8 July 2006 (EDT)

I just changed Template:Stub, and viewing pages that include that template it doesn't appear that the cache was cleared. Not a big deal, but I figured you might want to know. -- (WT-en) Ryan 22:30, 8 July 2006 (EDT)

This is a long discussion, so in an effort to sum up where we are and what's next:

• Bill and Ravikiran seem to be OK with getting rid of these lists provided that some automated way of creating them is put on the agenda for the future. I've added an item added to Project:Feature requests that includes Evan's proposal for automation.
• There is still an open issue over whether to delete the pages or redirect them to (for example) Special:Whatlinkshere/Template:Stub. Personally I think simply redirecting might be easier since it may be tough to fully de-link all articles pointing to these lists, but I'm OK with either solution - does anyone else have a preference?
• It should be OK to begin removing links to these pages, including from templates.
• Anyone interested in keeping any of the data from these pages should (for now) copy it into the appropriate article talk page, for example Talk:India.

Does that sound about right? If everyone is OK then we can begin de-linking the articles, and in a few days/weeks either convert these pages to link to the "what links here" page, or else delete them entirely. -- (WT-en) Ryan 18:44, 8 July 2006 (EDT)

Great summary. For deletion versus redirection: I suggest moving the articles somewhere else for archival purposes -- maybe Project:Collaboration of the week/Guide, etc., since they might be OK candidates for CotW? Although these pages aren't comprehensive, there has been a lot of work put into them, and it'd be good to archive that for data mining.

I'd think after the move, make a redirect to the Whatlinkshere. --(WT-en) Evan 21:53, 8 July 2006 (EDT)

Yeah, thanks for the recap Ryan! The conversation really took off... I'm happy with the consensus. (WT-en) Majnoona 22:45, 8 July 2006 (EDT)

The pages have been archived under the Project:Stub articles, outline, etc pages, and the old pages turned into redirects to the "what links here" page for the appropriate template. Rest in peace. -- (WT-en) Ryan 00:41, 14 July 2006 (EDT)

• Wikivoyage_talk:Stub_articles/Stubs_needing_attention
• Wikivoyage_talk:Outline_articles/Outline_articles_needing_attention
• Wikivoyage_talk:Usable_articles/Usable_articles_needing_attention
• Wikivoyage_talk:Guide_articles/Guide_articles_needing_attention