Jump to content

Wikipedia talk:Special:UnwatchedPages/Archive 1

Page contents not supported in other languages.
From Wikipedia, the free encyclopedia
Archive 1

Initial posts

This only goes up through 1 alphabetically. There must be more. — 0918BRIAN • 2005-12-21 16:42

I think it stops at a thousand pages, I got up to 2002 in Iraq or so.--HereToHelp 01:39, 5 April 2006 (UTC)

Purpose

What's the point of this page? Apparently, it was installed at Jimbo's behest... Dan100 (Talk) 16:50, 21 December 2005 (UTC)

It's a great way to make sure that every page is on someone's watchlist. Vandalism or bad edits often survive Recent Changes but are caught by someone with the article on their watchlist. There's a number of enhancements that can be made to this tool, but I think it's an excellent start. Carbonite | Talk 16:55, 21 December 2005 (UTC)
I've found 3 copyvios and 3 vandalisms so far. — 0918BRIAN • 2005-12-21 17:19
I was quite impressed by the guy who titled an article ''like this'', which made it great fun to try and link to... I've picked up at least one probable dodgy article, too.
On a more prosaic note, could this list be regenerated every few days? It might let us start making a dent in most of it - I've added maybe 150 of the pages here, but there's only a thousand up. Shimgray | talk | 00:30, 22 December 2005 (UTC)

Does anybody else think most of the top level domains (for ex: .kz) do not need their own article and instead should be in a list if at all? - Taxman Talk 17:24, 21 December 2005 (UTC)

  • No, I think they're mostly worth dealing with eventually. A few - the ones which are essentually dead - could probably be listed individually, but if nothing else each one is likely to need a couple of individual external links, and enough of them are detailed enough that we can expand them over time... Shimgray | talk | 00:28, 22 December 2005 (UTC)

About a year ago I thought a good way of dealing with vandals would be for all articles to be on someone's watchlist. But now with less than 1,000 daily users (at a guess) and over 800,000 articles, this is clearly impossible.

Central vandal patrolling, enhanced by the usage of stuff like the IRC channels and CDVF, is the only realistic way to control vandalism (imo). We just need better tools (which are coming) and more people. And maybe IP editing turned off... Dan100 (Talk) 10:35, 22 December 2005 (UTC)

  • Actually, if there are really 1,000 daily users, that would make for only 800-1000 articles on each watchlist, which is easily manageable, since only a minority of those articles are changed daily. — 0918BRIAN • 2005-12-22 13:15

Erm, well, agree to disagree on that I think... Good luck on you quest though, if it works out will undoubtedly enhance Wikipedia. Dan100 (Talk) 23:10, 23 December 2005 (UTC)

Live changes

Apparently, the page will soon be changed to create a new list of unwatched pages every time the page is loaded. Try to keep the number of refreshes on the page low. — 0918BRIAN • 2005-12-22 02:44

  • Cool. That way we won't be stuck watching the same pages at the front of the alphabet while vandals create hoaxes/autobiographies with names like "Zzzack...". I suppose once, if ever, we get all the pages watched, we ought to go back to cached mode for database concerns. — FREAK OF NURxTURE (TALK) 09:16, Dec. 22, 2005
    • Nope, I don't think it's going to happen. Tim Starling was for it, but Avar is against it. So, for now, it will be refreshed twice a week, but Tim Starling can make it more frequent. — 0918BRIAN • 2005-12-22 22:41

Maybe this information should be public

This is very useful information. I know right now it's restricted so it's not a honey pot for vandals, but we might want to consider going the other direction: what if each page, at the top, included a number that stated how many admins had the page on their watchlist? It would be sort of like PageRank -- the higher the number, the more you could rely upon the information on that page, and if it had a zero, you shouldn't rely upon it at all (and maybe the page could include a warning in red.) And perhaps we might want to institute a policy where pages are purged after a year if they remain at "0". (Forgive me if this has already been proposed elsewhere.)--Arcadian 15:47, 22 December 2005 (UTC)

This it to prevent vandals from exploiting this sensitive information.--HereToHelp 01:41, 5 April 2006 (UTC)

Updated hourly

Avar has set this to update hourly. The only way we are going to see items further down alphabetically is to watch all of the earlier items. So please, watch them, even if you don't know anything about them, especially if they've only had 1 edit in the last 6 months. You'll never notice them on your watchlist. — 0918BRIAN • 2005-12-23 04:29

Mass-watch tool

For those who would like to help, but don't want to click a million times, you can use this new tool to add X articles to your watchlist at once, using only 2 mouse clicks:

Add this to your monobook.js (e.g. User:ABCD/monobook.js)

/**** handy watch ****/
document.write('<script type="text/javascript"' +
  'src="http://en.wikipedia.org/w/index.php?title=User:R3m0t/' +
  'handywatch.js&action=raw&ctype=text/javascript&dontcountme=s"></script>');

/* This is to keep track of who is using this extension: [[User:R3m0t/handywatch.js]] */

The script will trigger automatically on Special:Unwatchedpages. It will prompt you on whether you want to active the script. r3m0t talk 04:10, 24 December 2005 (UTC)

0918BRIAN • 2005-12-24 04:11

Note: to split up the work, each admin needs to add at least 1500 items to their watchlist. The special page updates hourly, usually showing the latest version at around 5 minutes after the hour. — 0918BRIAN • 2005-12-24 04:16

Ok, maybe I'm just dense, but it doesn't work for me. I get the popup confirm box and I say yes, but no articles get added to my watchlist. And you make it sound like you can control how many articles get added somehow. Is it just a function of how many you have displayed, ie, 20, 100, 500? - Taxman Talk 16:50, 28 December 2005 (UTC)

Divide up the workload

I've done my part and using this page and the above mentioned mass-watch tool, I have added close to 2000 previously unwatched pages to my watchlist. I'll be adding more as they become available (apparently you can only view the first 1000 on the list so if you are waiting to see whats under "S" everything before that needs to be watched first). Perhaps someone should create a page to help divide things up. Maybe a list of names of those currently watching a lot of pages, and somewhere you can put the name of pages you are watching but have no real interest in. - Trevor MacInnis (Talk | Contribs) 12:21, 28 December 2005 (UTC)

Sort

It would be even more useful if we could search or atleast sort it. --PamriTalk 12:50, 28 December 2005 (UTC)

Agreed. The best thing I can imagine would be a search via category. I'd gladly add to my watchlist all unwatched pages from the categories I am interested in (Poland, sociology, etc.).--Piotr Konieczny aka Prokonsul Piotrus Talk 00:34, 28 January 2006 (UTC)
I'd be happy if it could be selected by namesapce. — xaosflux Talk 04:02, 27 July 2006 (UTC)

Only shows first 1000?

Is this supposed to stall out at 1000 pages? android79 20:44, 30 December 2005 (UTC)

  • It's a long-running bug on all such pages. — 0918BRIAN • 2005-12-31 00:16
    • If we can somehow get people to watch these first 1000 we can check out what follows. Any suggestions? - Trevor MacInnis (Talk | Contribs) 01:07, 31 December 2005 (UTC)
      • First 1000? There were thousands previous to this that others (myself included) have been watching just to get to these. Each admin needs to add at least 1500 (and more likely 3-5000) to get this list under control. — 0918BRIAN • 2005-12-31 01:26

I only asked because I thought, since this is only updated twice weekly, we'd get a lot of redundancy... thought I'd start somewhere in the middle of the alphabet. Oh well. When is it updated? android79 03:56, 31 December 2005 (UTC)

Getting around 1,000 limit; alternate offset?

I know you can offset the list by offset=<x> where x is a number. But, is there a way to offset the list to tell it to return unwatched pages beginning with, say, the letter h? --Durin 16:00, 4 February 2006 (UTC)

Clearing up to A

I've taken all the unwatched up to A (7 more just been created). I'm now going through them to check for possible deletions (AfD), moves etc. One thing that I have also been doing is checking the history. If the article was created after January 1 2006 and it looks like the creator is still active I am leaving this on their talk page. With a bit of luck they will turn the feature on and the list will not populate so quickly. I'm going to amend the text asking that they reply if they turn it on. That way I can perhaps drop some from my watchlist. The main problem is that the editing function of the watchlist opens all as one page (6000+). CambridgeBayWeather (Talk) 00:59, 5 February 2006 (UTC)

New refresh?

Special:Unwatchedpages appears to have been static since the server problems earlier in the week. Is there any chance someone could start it refreshing again, or are there technical issues? Shimgray | talk | 20:40, 18 February 2006 (UTC)

Yeh I've noticed the same thing... enochlau (talk) 14:13, 23 April 2006 (UTC)

RfA regarding Special:Unwatchedpages

There's a Request for adminship currently underway at Wikipedia:Requests for adminship/Genisock2 requesting a second account of admin Geni be granted admin rights so as to create a second watchlist for him of 5000 articles from Special:Unwatchedpages, allowing him to fight vandalism on a subset of these unwatched pages. Those of you following this talk page may be interested to follow the RfA and it's talk page. --Durin 16:02, 4 February 2006 (UTC)

Geni changed his mind, citing Guettarda's excellent suggestion. Er, OK, but what/where is this excellent suggestion of Guettarda's? -- Hoary 12:38, 24 October 2006 (UTC)
Since neither of the two mentioned that solution publicly, it's possible that some WP:BEANS are involved. Tizio, Caio, Sempronio 12:55, 24 October 2006 (UTC)
No, it was actually pretty simple. If you choose to use a sock for this purpose all you need to do is log in with your admin acct, and go to this page, where you will see the unwatched pages. Then open another tab (or browser window) and log out of your admin acct and log into the sock acct (but don't close the tab that's open on this page). Go back to the tab, and open each unwatched page you want in a new tab/window. Then add them to your watchlist. As long as your original page doesn't refresh, you will still see the unwatched list, but since you are now logged in with your sock, you will be adding the pages to that acct and not to your main acct.
One BEANS consideration though is if you have a dedicated sock you may not want to identify it was such to everyone (maybe just to someone with checkuser?) since any page the sock edited would probably be an unwatched (or now, slightly watched) page. Guettarda 05:18, 25 October 2006 (UTC)
Duh, why didn't I think of that. Nice one. Thanks! -- Hoary 07:07, 25 October 2006 (UTC)

why is this feature available for administrators only?

I don't see any reason why this feature should be available to administrators only. I'm sure that there are many users who would be willing to "adopt" some pages. --Ixfd64 00:59, 13 June 2006 (UTC)

That's answered above, and if you think about it it makes a lot of sense. Of course there are non admins that could be trusted to help out with this, but the issue is there hasn't been a technical way to restrict access to trusted users besides restricting it to admins. You could try asking at the technical village pump, but if non admins want to help out with vandalism, it's probably better to use one of the automated tools like Vandalfighter, AWB, etc. - Taxman Talk 16:25, 13 June 2006 (UTC)
Well, you could set it so it is only visible to editors who have x number of edits or have had an account for x number of days. This would prevent most drive-by vandals simply looking for something to mess with. Though really, I'm perfectly fine without seeing the page. Koweja 20:05, 12 December 2006 (UTC)

Updates infrequent

Seems to only update a couple of times a week. Given the enormous number of pages, is there anyway we could make this special page update at will? Is anyone working on this? Cool Hand Luke 01:50, 22 January 2007 (UTC)

We'll its been a few years, is there any progress on making this page actually usefull? - Trevor MacInnis (Contribs) 20:42, 3 November 2007 (UTC)

Suggestion

Would it be possible to have 37 of these Unwatchedpages? Twenty-six alphabetic (one each for A, B, C....Z), then one each for the numbers 0-9, and one for special characters? That way the 1000 article limit would be slightly less odious. Thanks, Ruhrfisch ><>°° 02:07, 16 December 2007 (UTC)

That would be better. We could even have an admin claim each letter and try to reduce the list each time it's updated. However, I get the impression that no one cares about this list. Cool Hand Luke 02:15, 16 December 2007 (UTC)
I am a brand new admin (just got the bit yesterday) - where could we ask about getting this split up / fixed? I don't want to spill the beans at the village pump technical or would WP:AN work? Ruhrfisch ><>°° 04:09, 16 December 2007 (UTC)
Maybe some community input at Village Pump technical and then mediazilla. Snowolf How can I help? 09:12, 29 December 2007 (UTC)

Another suggestion, unwatched BLPs. If the article is unwatched and has Category:Living people attached, it makes another protected list.↔NMajdantalk 18:32, 28 January 2008 (UTC)

Another suggestion: Special:Recentchangeslinked/Special:Unwatchedpages - at this point, to actually _watch_ them, you'd have to either add all the pages to your watchlist, or make a page with links to all of them (which would disclose the list). —Random832 18:54, 1 February 2008 (UTC)

Good idea - my thought was there are certain things I would add to my watch list (counties and places in Pennsylvania and Ohio), but I like the active version too. I will try to make a suggestion at WP:AN in the next day or two. Ruhrfisch ><>°° 19:20, 1 February 2008 (UTC)
That would be wonderful! It would be especially important to scrutinize unwatched pages for vandalism, and the recent changes list would revolve constantly so that you'll always be able to pick up new pages in areas of interest. Cool Hand Luke 19:22, 1 February 2008 (UTC)
I finally listed this at Wikipedia:Administrators'_noticeboard#Suggestion_for_Special:Unwatchedpages Ruhrfisch ><>°° 04:15, 18 February 2008 (UTC)
OK, the requests are in at bugzilla:13062, bugzilla:13063 and bugzilla:13064 - I have never done this before, so hopefully these are OK. Ruhrfisch ><>°° 04:58, 19 February 2008 (UTC)

A new project...

Before I was syssoped (sp?) a few days ago, another admin gave me the links to a handful of articles here, and I watchlisted them and then sent them scurrying to their appropriate wikiprojects, where there were people happy to be able to watchlist them for me. Now that I can see this page myself, I plan to continue doing this in a slightly more organised way. I will create some pages, probably in my userspace, each referring to a different wikiproject.

I'll watchlist the pages as I pull them off here (copy and paste into a raw watchlist) and then list them according to which project they belong to. Then I'll go and notify each project in turn when there are new articles to be searched. At leats, that's the plan, though currently, there aren't any pages showing up in the list... Anyone got any thoughts? Fritzpoll (talk) 23:16, 29 May 2008 (UTC)

I like the plan. FYI, the page is up again - was run June 1st. It is curently just a list of asteroids whose names start with the number 1 (the list is only 1000 entries long). Ruhrfisch ><>°° 19:32, 2 June 2008 (UTC)

Asteroid Belt

I looked at this and my first reaction as that as I have a fondness for lonely rocks I might watch a few asteroids to move the list along. Then I noticed that there are 10,988 asteroid stubs in total. Ben MacDuiTalk/Walk 11:55, 17 July 2008 (UTC)
I am not sure what else to do here - I thought making 26 plus such lists sorted by first character would solve some of the problems, but no work seems to have been done on my Bugzilla requests. Ruhrfisch ><>°° 17:29, 17 July 2008 (UTC)
It is hard to guess how many pages would actually be on this list in total, but it could be anything between 20,000 and 100,000. In its current state it is completely useless. I wonder if there is any mileage in suggesting a sort by category? Ben MacDuiTalk/Walk 18:40, 17 July 2008 (UTC)
There is a Bugzilla request in to make an unwatched Biographies of Living Persons list based on a cat. I think the problem is that the squeaky wheel gets the grease and no one complains much about this. Perhaps a posing on the Village Pump Technical? Ruhrfisch ><>°° 19:36, 17 July 2008 (UTC)

Might be worth it. I imagine the community would be alarmed if they became aware that a potentially very large number of pages are unwatched, but as suggested above somewhere there is a possible downside to giving it publicity. Ben MacDuiTalk/Walk 08:56, 18 July 2008 (UTC)

I was surprised to see that I could access this talk page when I logged out, so this discussion is open to non-admin eyes (although I doubt many users of any sort come here often). Ruhrfisch ><>°° 11:51, 18 July 2008 (UTC)
Agreed about that. I also suspect that few admins are that interested in this subject or action might have been forthcoming. I therefore can't see any point in flagging the subject up at the Noticeboard. I do however think that a note to the effect that pages to the number of at least four figures, probably five, and maybe even six are not watched by anyone might get some attention at the Pump. I know nothing about the technicalities but it is hard to imagine there is something inherently problematical about creating a list of unwatched pages either by starting letter/integer/symbol, or by some kind of categorisation. If it were not complex I'd favour the latter as I am more interested in say Geography articles than I am in articles beginning with 'E', but either would be a step forward. The alternative would be to simply delete this page - in its current form it is effectively useless and we are kidding ourselves to pretend otherwise. Ben MacDuiTalk/Walk 15:09, 18 July 2008 (UTC)
My thought was if I wanted to check that a place in Pennsylvania or Ohio was unwatched, I could look in the alphabetic list. I agree that categories would be easier. I will try to post something to the Village Pump Technical later today. Ruhrfisch ><>°° 18:12, 18 July 2008 (UTC)
It occurs to me that an IRC discussion might be a more secure way of dealing with the issue but I am not a practitioner of this dark art. I know someone who is if you want me to give it a try. Ben MacDuiTalk/Walk 09:10, 19 July 2008 (UTC)
I have also never been on any IRC channel, but it sounds like a good idea. Please go ahead - discussion could also take palce here once people know more. Ruhrfisch ><>°° 13:15, 19 July 2008 (UTC)
Said old chum seems to have left Wikipedia in despair but I'll see if I can find someone else. Ben MacDuiTalk/Walk 14:54, 19 July 2008 (UTC)

As of 11:58, 16 July 2008 "There are no results for this report", which makes discussion a shade harder. For the benefit of anyone joining this debate the last posting of the list was clogged up with asteroids stubs beginning with the integer 1, of which there are clearly many hundreds, if not thousands.Ben MacDuiTalk/Walk 15:11, 19 July 2008 (UTC)

I have seen this "no list" before - in some ways it is an even better argument for changing the system - none at all is worse than 1000 asteroids. Unfortunatley I just got bogged down in an FAC and need to do some rewriting. Perhaps the Village Pump Technical is the way to go - I brought this up at AN/I after I became an admin and all I got out of it were the Bugzilla requests. Ruhrfisch ><>°° 19:26, 19 July 2008 (UTC)
Take a deep breath, folks. First off, the asteroids really aren't that bad. There is a bit of a trick to this, and there are a few of us doing it. Consider creating an acknowledged alternate account specifically for watching these pages, and when the list appears (as it does about once a week or so, depending on the wonts of the developers), copy it over to the alternate account. (I stop in Word in between to strip off everything but the article name.) I have several thousand pages on mine, many of them asteroids and planets, and I check it once a day or so to catch vandalism. Remember that BLP-related vandalism can occur on any article; some of the worst is sneaky and put into articles that don't give the appearance of being related to a person. Since I started working on this, about two months ago, I have found one BLP related vandalism edit, and only about half a dozen other vandalistic edits. Even with all these pages on the watchlist, it's rare that I see more than 10 page changes a day; when I do, it is usually because a bot has gone through. It's a worthwhile endeavour, to ensure all pages are watched by someone. Don't forget, though, that plenty of pages are "watched" by users who have long since left Wikipedia. Risker (talk) 20:24, 23 July 2008 (UTC)
Thanks very much Risker. I have nothing against asteroids (and would love to visit one someday), but my guess is that inhabited places will have more vandalism since people actually live there. Just for the sake of argument, let's say Zelienople, Pennsylvania is unwatched (it's not - I just added it to my list). If I could somehow get a list sorted by Pennyslvania boroughs, then I would see this right away. By your plan, assuming only 1000 unwatched articles per number and letter and one new list a week, I would have to wait 35 weeks to see if it was unwatched, and would have to have a watchlist with 35,000 articles on it to get there. My guess is there are many more articles on most letters and numbers and it would take longer and I would have an even bigger list to watch. While your method works, it is not the most efficient way to do this. I also have thought about the watched only by the departed problem, but will worry about that more when this is better resolved. Thanks again, Ruhrfisch ><>°° 00:45, 24 July 2008 (UTC)
Watching the top pages each week only results in many people watching the tip of the iceberg. This needs to be fixed so that the submerged unwatched pages beneath can be watched, especially in sensitive categories. Incidentally, I think 100,000 is a very conservative estimate given how we've been gnawing away at the beginning of the list since it was created. Cool Hand Luke 00:48, 24 July 2008 (UTC)
All I can say is that I'm doing my bit. If the list was automatically generated as soon as everything on it had moved to a watchlist, it would all go much faster. I figure 100,000 is probably a bit high. I'd be more than happy to have all those celestial bodies on my alternate account watchlists (incidentally, about 15% of these numerical articles aren't asteroids) and am looking at an alternate way of getting them onto my watchlist so that this list will move more quickly. If there was a willingness on the part of, say, 100 editors to all take 1000 articles, and we could make arrangements with the devs to get the list to regenerate more quickly, it could be taken care of in no time. Admins would have to assist regular editors who are willing to help (my betting is there are quite a few who would be willing), but it is pretty straightforward. Risker (talk) 01:15, 24 July 2008 (UTC)

Yeah, that would work. Hell, if we could get this damn thing refreshed even daily, we could probably chomp down the list in a year. But I should point out that 100,000 is not a high estimate. This list rarely gets past "1" I've hit random page over over 100 times, and have not yet seen something that low. I'm guessing that the proportion of articles starting with "1" or lower <1%, and if unwatched pages are randomly distributed throughout, we would expect well over 100,000. Cool Hand Luke 12:05, 24 July 2008 (UTC)

July 25th list

Just to let folks know I have added the July 25th list to my alternate account watchlist, so nobody else needs to do so if they do not wish.

As my list is building, I am noting certain trends in the articles included:

  • Military units and ballistics
  • The arts, including songs, albums, and award shows (15th annual...)
  • This year in history of a certain country/literature/etc
  • Notable addresses

While I don't have time to do this right now, I am inclined to sort some of these into their own categories and approach editors interested the these general subjects to "take over" these smaller sublists. Thoughts? Risker (talk) 12:27, 26 July 2008 (UTC)

It sounds like a good idea to me, although the sorting could be a lot of work, and something that would ideally be done automatically. Ben MacDui 13:15, 26 July 2008 (UTC)
Yes, thanks for doing this. Ruhrfisch ><>°° 14:41, 26 July 2008 (UTC)
Co-ordinating and splitting up topics is a fabulous idea. Cool Hand Luke 14:44, 26 July 2008 (UTC)
I suppose we could get a bot to do it - there is one which splits up new articles by topic so presumably the same could be applied to this page. Hut 8.5 15:11, 26 July 2008 (UTC)

The following entry appears in the Signpost Technology report: "The information in Special:UnwatchedPages is now available to administrators via the API (using list=allpages with apfilterwatched). (r37748, bug 14020)". Sounds good, no idea what it means. Ben MacDui 15:46, 26 July 2008 (UTC)

Yes, that is good - if the API can access the information then bots can get at it and process it. Hut 8.5 16:00, 26 July 2008 (UTC)

France

I've left a discreet note at WProject France about the unwatched pages here. It remains to be seen if anyone replies. Ben MacDui 19:44, 28 August 2008 (UTC)

Just found this page - added some to watch

Somehow I have never seen this special page before. A good idea indeed, it it is refreshed often enough. Even though my watchlist is already way too long, I understand that many of these unwatched pages do not/should not change too often so I'll add some of them to my watchlist so at least someone watches over them. I'll also peruse the list periodically to find hidden/unseen vandalism as I've read in the previous comments. Cheers! -- Alexf42 16:54, 10 September 2008 (UTC)

Create a counter of people watching a page

Okay, in Wikipedia:PERENit is noted that this would specifically make unwatched pages vulnerable to vandals.

Simple issue (with a little software tweak): You do not give a count. You only give (as an example) one of the following three possible responses:

This page is watched by:

  • Less than 10 editors
  • 10-100 editors
  • More than 100 editors

This would make sure that there are no invitations to vandals, yet I could be sure that I am not leaving a page unwatched that I don't really care THAT much about if I am number 11+ Ingolfson (talk) 08:56, 15 September 2008 (UTC)

Opening things up

What was the rationalle for making this special page require admin rights? It doesn't make a lick of sense to me - Don't make users ask for the tools if all they want are mundane information-gathering tools like this.

Alternative:

Possible workaround: instead of creating this sock account, program the software to allow any user to view all unwatched pages in a "watchlist-style" format, so that it shows each page in the order they were last edited. --M@thwiz2020 20:44, 4 February 2006 (UTC)

The proposal made by M@thwiz2020 in the 2006 RfA cited above seems like a reasonable in-between, if there really are legitimate reasons to hold this close. That said, though, I would certainly prefer to just see this page opened up to registered editors. Potential damage from vandals building massive watchlists from here would be largely mitigated by the suggestion from Ingolfson above. MrZaiustalk 08:22, 21 September 2008 (UTC)

The current problem is not so much who has access to the page as that it is totally dysfunctional. It is updated very irregularly and the size limit makes it largely useless. Until this is fixed I can't see any point in extending the franchise. Ben MacDui 09:53, 21 September 2008 (UTC)
Yep; the page should be functional before its worth expanding. But a watchlist-style feature would be a great improvement. We would add Special:Recentunwatchedchanges. Through that page, all of the unwatched pages would be visible on a rolling basis, and admins could adopt pages off that list as they peek their curiosity. Cool Hand Luke 20:01, 21 September 2008 (UTC)
To answer your original question, the reason it was admin only was that (when it worked) vandals could pick-and-choose from it to find places they wouldn't be spotted. – iridescent 21:52, 3 October 2008 (UTC)
That's the reason that, if this were to be fixed, it would be nice to also have Ingolfson or Mathwiz's proposed fixes implemented. That said, if Special:Statistics can differentiate between active and inactive users, it would be extremely useful to add that caveat to Ingolfson's proposal or, if this page is fixed in its older format, to also ignore watchlists of inactive users. If it's the product of (Total number of watchlists)*(Total number of pages) that's causing this issue, it might actually make it a good deal easier to bring it back and expand it per Ingolfson's proposal if the tool only looked at active user watchlists. MrZaiustalk 06:42, 4 October 2008 (UTC)

Incidentally, one of User:Ruhrfisch's best bugzilla suggestions back in February has apparently been coded and works. Who do we talk to about making it live? Cool Hand Luke 18:42, 4 October 2008 (UTC)

I think either contacting User:Geometry guy or User:CBM (Carl) might be a start, or raising the issue at the Village Pump technical or perhaps WP:AN. Ruhrfisch ><>°° 03:25, 5 October 2008 (UTC)
In the meantime, any non-admins are welcome to ask me for a list of pages to watch. John Reaves 22:53, 6 November 2008 (UTC)

A way to handle unwatched pages

Note: This proposal has now been archived, please start a new discussion if you wish to comment.

This watched-pages proposal might take the strain off this page, among other things. Or, it might be unfeasible. Please comment there if interested. –MT 23:45, 21 April 2009 (UTC)

Restriction

I understand why this information is restricted, but surely opening this up to other trusted editors (like we do with Rollback, Accountcreator, Autoreviewer, Abusefilter) would be a good thing- mainly so these pages actually appear on someone's watchlist! It just occurred to me- the irnoy if nobody replied to this because it wasn't watched! HJMitchell You rang? 21:14, 21 January 2010 (UTC)

It doesn't actually work at the moment for anyone. I suggest we get that fixed first. Hut 8.5 22:20, 21 January 2010 (UTC)
Ah, yeah. That might be an idea! I wasn;t aware of that when I made the above post! HJMitchell You rang? 22:31, 21 January 2010 (UTC)
Or else it truly is empty, after some people speedy deleted all unwatched pages (hey, who would notice it, they are unwatched anyway!). Just kidding, I hope... Fram (talk) 09:54, 22 January 2010 (UTC)
The problem is that even when it worked, it was nearly useless as it only showed the first 1000 unwatched pages, which I recall all started with numbers. I also agree that this could be shared with trusted users, if it ever worked. Ruhrfisch ><>°° 16:20, 22 January 2010 (UTC)
The reason for the restrictions is mentioned a few times on this page. However, my main concern is the "1000 limit" - I'm not sure why this can't be worked around, as the list isn't generated on the fly. Even if the processing load would be large, it could perhaps be run once a day - and a way to see pages beginning with "A", "M", etc would be useful! -- PhantomSteve/talk|contribs\ 08:36, 19 April 2010 (UTC)
Yea, the 1,000 limit makes this almost worthless. There may be pages I'm interested in but generally not ones that start with special characters or numbers. Vegaswikian (talk) 19:29, 7 May 2011 (UTC)
Agreed. It would be impractical to scroll through the entire number of pages , but it could be divided alphabetically, or even show a random assortment. Then we could see how useful it would be to the admins at least, and think about extending visibility. DGG ( talk ) 20:17, 2 November 2011 (UTC)

I was asked to comment

The number of "What links here" incoming backlinks an article has is a useful proxy for how many pageviews it gets, which currently takes way too long to query

I think the single biggest improvement to this page would be to have it sorted by some rough measure of page readership / popularity. When I look at it today (sorted alphabetically) I find it unexciting because I don't want to clutter my own personal watchlist with a bunch of pages that aren't statistically speaking likely to be very damaging for the simple reason that they have extremely low readership. At the same time, I worry a lot about "What are the most popular pages that no one is watching?"

There are many more intelligent improvements that we can imagine on top of that. What about a list of popular pages unwatched by any trusted person? What about a list of popular unwatched pages that have some history of vandalism already? How about weighting BLPs more heavily? Algorithms could be developed to do a reasonable job of picking out interesting things for people to watch. But I see no reason why we should make the mistake of waiting for or asking for something perfect, when something simple could be done quite easily.--Jimbo Wales (talk) 15:58, 13 April 2012 (UTC)

I agree one could also get more intelligent in terms of detecting potential vandalism. For example, if someone has vandalized a page, and it has been detected by a bot (and reverted), then that same person makes an edit (the last edit) that is *not* reverted by the bot, it could still be vandalism and probably needs to be looked at. One could devise a sort of 'risk' score, as a weighted-metric, by adding up how many watchers, date of last edit, number of previous vandalism events, and number of page views, etc. Then this list, sorted by top 500, would give you a good starting point of pages responsible eds should take a look at. --KarlB (talk) 17:54, 13 April 2012 (UTC)
I am sure there are many useful things that could be done e.g categories by popularity/Project/nation state (there were large numbers of French articles on the list at one point for example) etc. etc. Different editors will have varied interests and enthusiasms and if the list can be split into smaller chunks it should not be unmanageable. However, whilst it may well be the case that "something simple could be done quite easily" to date it has proven hard to get anything done at all. Is it reasonable to assume, Jimbo, that you will use your influence with the technical folks to get them to make a start? Ben MacDui 18:59, 13 April 2012 (UTC)
  • This was my rationale behind allowing access to editors that have already been approved for rollover, reviewer, file mover, etc. (yes, I'm one). Or combining with a minimum years/edits criteria. At least some vetting has taken place, greatly reducing the odds of the user becoming a vandal. It could begin fixing the problem today. No extra classification of editor is needed to begin getting the benefits today. If you can't trust a rollover editor to not vandalize, then they shouldn't have rollover. Fixing the list, how to sort, those are interesting ideas with obvious merit, but it isn't required we do that first. It can be tweaked along the way. Dennis Brown (talk) 21:56, 13 April 2012 (UTC)

Note that you can use the API to obtain the number of incoming links to each article from article space (namespace=0) relatively efficiently. 70.59.20.190 (talk) 20:07, 13 April 2012 (UTC)

  • Here's a thought (which may, admittedly, be a teeny bit OT, but...). What about using some sort of metric involving the number of reverts performed by Cluebot NG (talk · contribs)? Since ClueBot patrols recent changes, it doesn't really matter how many watchers the vandalized page has. Comparing the top hits from ClueBot (perhaps mixed in with some other vandal patrol page stats?) with the top unwatched pages (somehow) ought to provide some useful "needs to be watched" lists, I'd think (Maybe the top ClueBot pages per 1/incoming links?).
    — V = IR (Talk • Contribs) 06:43, 14 April 2012 (UTC)
  • I think the difficulty here is knowing what criteria can best be used to search for pages which are not currently watched (but should be). We've identified a number of plausible criteria (e.g. age of account at time of page creation. Could we envisage some kind of tool that allows patrollers to select a number of criteria they are interested in and then filter and drill down into the matching content? We'd be building a generic exploring tool rather than anything tied to any specific metric or vandalism technique. --Salimfadhley (talk) 09:56, 28 April 2012 (UTC)

A suggestion

I think that we have three main lines of defence against vandalism:

  1. Automatic controls, such as editfilters, [semi]protection of the most vulnerable articles, title blacklist, and so on.
  2. Recent changes patrol (and perhaps NPP &c too).
  3. Watchlists.

Each of these three layers of protection is good, but not perfect, at catching vandalism and some other kinds of problematic editing. Each layer helps compensate for gaps in the other layers. However, having lots of unwatched articles weakens that third layer of protection (admittedly the gap involves less-active articles with few incoming links). It would be really helpful if we could shrink the list of unwatched articles, to reduce the risk of sneaky vandalism persisting in the long term.

So,

  • Would adminfolk object to giving me a piece of this list, maybe a couple of thousand, so I can watchlist them...?
  • More generally, would it be possible for other known/trusted editors to request the same thing? (If you don't trust me, fine, but surely there's somebody else you'd trust to keep an eye on some articles). bobrayner (talk) 13:57, 18 July 2012 (UTC)
Why not try your hand at RfA Bob? -- œ 08:05, 23 July 2012 (UTC)
You make it sound so easy! I don't think I'm ready to run the gauntlet right now, and the end result of an RfA may not lead to me being able to see Special:UnwatchedPages. :-)
Anyway, there are probably other editors out there who might want to get a slice of Special:UnwatchedPages, and who could be trusted to watchlist that slice; should they all have to try RfA? It's worth considering the general case. bobrayner (talk) 08:41, 23 July 2012 (UTC)
  • Anyone? Anyone? Bueller? bobrayner (talk) 10:21, 1 August 2012 (UTC)
  • As to the list, if you just add some of these 1000 to your own watchlist, won't they just drop off and new ones replace them? Dennis Brown - © Join WER 13:16, 1 August 2012 (UTC)
    • Well, the actual pile of unwatched pages is unlikely to be infinite; It's just that the visible list is the top 1000 of the pile (I haven't seen the page but that's what others say). Taking the top 1000-2000 off the top of the pile may make a new 1000 visible on the list, but the overall pile will have shrunk. If I (or some other editor) took on all the previously-unwatched articles from (say) Aardvark to Abacus, then the new list would show remaining articles between Abacus and Achæmenid. (That's just a human-readable example. I presume the boring reality is that it's in asciibetical order, and the top few thousand probably start with the number 0 or with some odd punctuation mark). bobrayner (talk) 14:01, 1 August 2012 (UTC)
      • {{palmface}} I keep forgetting the list is only viewable by admins.... I've participated in a discussion about opening it up, but it never reached consensus. That was many months ago, I forget where. Dennis Brown - © Join WER 17:30, 1 August 2012 (UTC)
Note: This can still be easily gamed. For example, a determined vandal could grab this list, register an account, and watch a bunch of pages (thus removing them from the list); then go on to vandalize at will. A single watcher doesn't solve the problem - it needs to be more intelligent (like, who is the watcher, how active are they, etc)--Obi-Wan Kenobi (talk) 16:16, 1 August 2012 (UTC)
Our primary defence against that particular risk is that the page is only visible to admins, and we make the slightly optimistic assumption that admins are less likely to be "determined vandals". :-) Nonetheless, I'd like to watch a part of this list. If adminfolk think that I'm going to vandalise them, or - more generally - think that encouraging experienced editors with no record of vandalism to watch some neglected articles would increase risk in some other way, I'll give up now. bobrayner (talk) 17:14, 1 August 2012 (UTC)
Right; but there was a discussion above about opening this up. If we did open it up, it should probably be opened to trusted users, perhaps users who had amassed a certain number of edits, or who had some brief peer review of their edits.--Obi-Wan Kenobi (talk) 18:10, 1 August 2012 (UTC)
I think we could agree on that. Opening it up to everybody sounds nice in principle, but in practice I think it could enable embarrassingly-persistent sneaky vandalism. Not sure where the best threshold might be, but personally I think enwiki is not very good at sudden process change (especially where privileged access is concerned) so it would probably be better to start with something adhoc and see what consensus forms over time... bobrayner (talk) 22:19, 1 August 2012 (UTC)

Fix Unwatched Pages technical issues and de-restrict visability

The fact that there are atleast 1000+ pages that are unwatched could is a big issue for me, IMO. How many of these thousands of pages contain BLP issues, or copyright issues, or just contain blatant vandalism that has been there for months, or worse, years? If this was open, and we started a project to get trusted people looking at this list (Rollbackers for example) we could reduce the number of pages that are 'redundant' and help to keep Wikipedia clean of issues that could harm or destroy its reputation.

Is there some kind of concensus we can come to, to allow minor de-restriction to people who are trusted, but still restrict for non trusted editors? Mrlittleirish 13:42, 11 April 2012 (UTC)
At the moment the page is not a very useful tool since it only lists the first 1000 pages. That doesn't get you very far (the last entry is 1972–73 Copa del Generalísimo at the moment). To be an effective antivandalism tool the number of pages on the list would have to be drastically expanded. I do think it would be a good idea if we could widen the readership, perhaps by granting it to reviewers, but given the current length of the list the issue is largely irrelevant. Hut 8.5 15:34, 11 April 2012 (UTC)
Well, we could start on some kind of catagorisation tool? If we get them all sorted alphabetically that would help? Mrlittleirish 15:46, 11 April 2012 (UTC)
I'm not sure you understand the problem. The special page sorts the list of unwatched pages alphabetically and displays the first 1000. It isn't possible to get the page to display anything more than the first 1000. Since there are far, far more than 1000 unwatched pages on Wikipedia this makes the page virtually useless, especially as those 1000 pages are all very low risk. Until this problem is fixed there's little point trying to open the page up. Hut 8.5 18:23, 11 April 2012 (UTC)
"Until this problem is fixed there's little point trying to open the page up.." completely agree. This problem has been around for ages and no-one seems to have come up with a solution. On the one hand, it seems crazy that potentially very large numbers of pages are unwatched, on the other bringing the issue to the attention of a wider public has its risks. I don't know it has ever been raised at Village Pump, but that's a possible way forward. Ben MacDui 19:24, 11 April 2012 (UTC)

I fully understand the problem Hut. I know only the first 1000 can be viewed, and like you said, we need to get that fixed before opening it up. So where do we start to try get it fixed? The amount of unwatched pages that are currently un-noticed could potentially lead to lawsuits and reputation impacting effects if there are BLP or copywrite issues that have not been stumbled upon. They may be low risk pages however, they are still pages within this project. Mrlittleirish 20:33, 11 April 2012 (UTC)

If you open it up, then eds will probably watch those first 1000, so then the next 1000 will become visible. In any case, this is a great example of security through obscurity. Letting confirmed eds, or eds with a certain number of edits, see this page seems reasonable. --KarlB (talk) 00:21, 12 April 2012 (UTC)
  • From a tech point, could it be opened up for non-admins who are page-movers or rollbackers, who can adopt 100 pages each, perhaps? There is at least a little vetting that occurs to get those tools, so accessing for vandalism is greatly reduced. Ideally, enough people would watch enough pages that the 1000 page limit would eventually become meaningless. Still good to fix, but it wouldn't be required to do first. Dennis Brown (talk) 12:38, 12 April 2012 (UTC)
  • So instead of offering what we can do, and what won't work, where can we go to begin consensus on actions to be taken? If the first 1000 pages all begin with the number '1' then, imagine how many pages with the number '2' are unwatched, and even the number '3'. Need I go on? We need to make a push. Mrlittleirish 12:49, 12 April 2012 (UTC)

We would like to get the ball rolling to try and fix bug that only shows the first 1000 unwatched pages. We propose that we reach consensus to sort all unwatched pages alaphabetically, and open up Special:UnwatchedPages to trusted editors, such as rollback or reviewers. Mrlittleirish 08:16, 13 April 2012 (UTC)

Support

  • I definitely Support this proposal, especially the 1000 page limit which has gone ignored for years. Plus if this were to be opened up to trusted editors I believe it would help with, rather than hinder, the fight against vandalism, as well as maybe even lead to discovery and elimination of long-stale unknown vandalism edits and/or hoax articles. -- œ 08:28, 13 April 2012 (UTC)
  • I will Support as per my comments above. Mrlittleirish 10:10, 13 April 2012 (UTC)
  • Support fully Ruhrfisch ><>°° 10:35, 13 April 2012 (UTC)
    • Just to clarify, there needs to be some way to be able to tell what pages are completely unwatched. It also seems clear that the number of unwatched (and little watched) pages is likely very high, so allowing trusted users access to this infomration once the problem is fixed in some way seems like the only practical solution. I support this porposal, but would support others that had the same effect and solved this issue. Ruhrfisch ><>°° 19:09, 13 April 2012 (UTC)
  • Support Obviously.--Gilderien Talk|Contribs 10:45, 13 April 2012 (UTC)
  • Support - even before fixing the 1000 page limit, which still needs fixing, but isn't required to gain the advantages of having more eyes on the articles. Let dev do what dev does best, and trusted editors do what they do best, each on their own timetables. Dennis Brown (talk) 13:01, 13 April 2012 (UTC)
  • Support - we may want to break this RfC into two pieces
  1. Whether non-admins should be given rights to the existing page, and if so, which ones
  2. What improvements we can make to the page

In any case, I do think that with 1000 articles, it is already useful. It probably wouldn't be that hard to change the code to generate a list of 5000 instead, as a first quick change.--KarlB (talk) 13:37, 13 April 2012 (UTC)

  1. Support the technical fix, of course. Oppose extending access rights until after a more thorough discussion of the risks and benefits.  Sandstein  15:27, 13 April 2012 (UTC)
  • Surely it's blindingly obvious that, if possible, the technical fix should be implemented. If it can't be implemented, we may as well scrap the whole thing, because it's useless at the moment. On opening up the page, I generally agree with Karl and Sandstein – we need a better discussion about how extending access would work (e.g. how would these trusted users be selected, would they need a new userright, what would the potential for abuse be, etc.) before we get into a straight support/oppose vote on the subject. Jenks24 (talk) 17:27, 13 April 2012 (UTC)
  • Cautious support. I am very much in favour of finding suitable technical fixes (without which I agree any extension of the franchise is probably a waste of time) and once it has become clear what can be so achieved in having a larger cadre of watchers involved - especially as I fear the number of pages probably exceeds 100,000. However, these steps need to be taken in that order. Ben MacDui 18:42, 13 April 2012 (UTC)
  • Support for the fix and not sure about opening that page for more editors. Can somebody explain me why this page has this restriction? mabdul 18:03, 17 April 2012 (UTC)
  • Strongly and unreservedly support technical fix; support opening to non-admins as per proposer. St John Chrysostom Δόξατω Θεώ 18:19, 20 April 2012 (UTC)

Oppose

  • This list is pretty much useless and there is no reason to allow more people to see it, until it gets fixed. And, quite frankly, I don't think we need any more user rights for the benefit of editors who are looking for more hats to collect. Salvio Let's talk about it! 11:14, 13 April 2012 (UTC)
    • Part of the solution suggested is to get it fixed, then open it up to editors who already have rights, such as rollback or reviewer. I'm not sure if you read the discussion above the notice either. Mrlittleirish 11:18, 13 April 2012 (UTC)
      • Then have the list fixed and then start a discussion again. Until then, it is perfectly useless to discuss opening it up. And, to continue being quite frank, there are a lot of people who have been granted the rollback flag (and even more who have received the reviewer bit) who I would never trust to see such a page – if it were fixed and became actually useful –, as some admins grant rollback to people who have reverted some fifty vandal edits (and why should they not? It's not like rollback allows you to do anything a non-rollbacker can't do just as easily). Salvio Let's talk about it! 11:33, 13 April 2012 (UTC)
        • While I respect anyone's right to oppose, this is a bit of a chicken and the egg problem - which comes first? There have been multiple requests to fix this page at Bugzilla for over 4 years with no action. When I contacted the Wikimedia Foundation recently, one of the WMF staff suggested a RFC to show consensus that this should be fixed, suggesting that would be the only way to get action on the problem. Ruhrfisch ><>°° 11:41, 13 April 2012 (UTC)
        • This discussion is regarding both the fixing and the opening up of the special page. Mrlittleirish 11:44, 13 April 2012 (UTC)
  • Oppose; proposer has not convinced me this would not simply result in the opening of another attack vector. Opening the (rather useless) page and "fixing" it to be able to view all unwatched pages is not a solution. There are many thousands of unwatched pages and I am not convinced that the proposer will be able to find and convince enough editors to watch them all (and, even then, once a page is watched by one person this is not a great solution). I think this is more of a hat-wearing thing than any useful proposal. TO put it another way; this proposal doesn't solve the problem it relates to (protecting unwatched pages from attack) and so I suggest that a more careful proposal to solve the correct is issue is raised. --Errant (chat!) 14:10, 13 April 2012 (UTC)
    • Instead of outright opposing, why not suggest a better way to make it better? This RFC is not just to open and 'fix' it. It is to build consensus on how to resolve this issue as a whole. Everything can change, including proposals. Mrlittleirish 14:18, 13 April 2012 (UTC)

Neutral

Comment

This is effectively pointless because the page is, currently, useless. There are many thousands of pages that are unwatched - and this page doesn't present a solution to that, even if opened up. What would be better is a recent changes filter (and patrol mechanism) for unwatched pages which could then be opened to a wider user group.

But before that development occurs there is no point to having an RFC about this page (because all that will happen, if successful, is the page will be opened and it will just sit there, uselessly).

Develop a full technical proposal, get community support, then ask the devs to implement it. I suspect the foundation will show interest because these pages are a significant risk factor. --Errant (chat!) 10:46, 13 April 2012 (UTC)

All these pages can/should be placed on WP:Pending changes protection, if/when it is switched back on. - Youreallycan 10:48, 13 April 2012 (UTC)
The easiest way to do that would probably be to get further development on the Pending Changes extension so it is automatically applied to unwatched pages. I agree; it is the way forward. --Errant (chat!) 10:51, 13 April 2012 (UTC)
Errant, getting full community support for a technical change here is exactly what this RfC is about. Did you read the pre-discussion above? Several bug reports were filed over several years and still nothing. This is the best way to get the issue out there. -- œ 11:17, 13 April 2012 (UTC)
I don't think that getting this specific page to work as intended will be beneficial. And I doubt that it would be approved by the devs even with a successful community RFC. --Errant (chat!) 11:33, 13 April 2012 (UTC)
If the pages are opened up to admins or rollbacks or some other group, they could be on a summarized recent changes, like this. tedder (talk) 14:40, 13 April 2012 (UTC)

It's not clear to me what the votes above are for. On the one hand the title says it's about opening the page up to non admins, but those who are voting "oppose" to that are then being told the proposal is wider than that, without specifying how wide. In other words the vote seems to be either support = open up to non admins / oppose = don't or support=do something (anything) / oppose = do nothing. I could do with some clarity about what exactly the proposal is that's being voted on.

I certainly agree that the status quo is an undesirable state of affairs but I agree with the opposers that just opening the page as it stands up to non-admins is not the solution. waggers (talk) 15:17, 13 April 2012 (UTC)

I have changed the title, overall, it is proposed that we gain consensus to first, put forward the technical issues with the special page (only able to view first 1000) then, when a fix has been implimented, to revisit the restriction and grant access to trusted users who are non-admins. If you need any more clarification, please ask. I can redraft the whole proposal if required. Mrlittleirish 15:21, 13 April 2012 (UTC)
Thanks, that helps. I don't think being able to see more than 1000 articles in the list is going to help at all. What's really needed is something different, like the suggestion above of implementing pending changes on all unwatched articles, and/or
  1. an "unwatched" filter on Recent Changes
  2. a tool that lets you see unwatched pages by category
It's worth remembering that even if an article is on somebody's watchlist, it isn't necessarily being watched. There are lots of inactive user accounts around and if an article is watched by one of those (and nobody else) it wouldn't show up in this list but is in effect unwatched. So the problem is bigger than the thousands of unwatched pages, and getting them all on someone's (anyone's) watchlist won't solve it. waggers (talk) 15:42, 13 April 2012 (UTC)

I have no comment about who should or shouldn't be allowed access to this; but I do imagine that there must be tens (hundreds?) of thousands of unwatched pages. Presenting all of them wouldn't be much good; how far would you get clicking through a list that big? I suggest modifying the current tool, instead, to show, say, 500 random titles from the list. There's nothing like a bit of surprise to keep people engaged. -- Hex [t/c] 18:03, 28 September 2012 (UTC)

Fix these articles, then worry about the other unwatched ones

Cross-posted from the administrators' noticeboard:

If you're looking for pages to defend, you don't need access to Special:UnwatchedPages. There are tons of little-watched articles that have been completely ruined over time by vandals, well-meaning people with extremely poor English skills, people who don't know what should and should not be included in an encyclopedia, people who insert lengthy, unwikified screeds that feature ALL-CAPS sections (and may or may not be copied and pasted from another website), and people who accidentally drop random junk such as '''Italic text<big></big>[[Example.jpg]]''' and their IP signatures into articles.

Renala Khurd is such an article. Chak 4GD is another. There are literally thousands of such articles. Some of these articles are on the unwatched pages list, some are not, but they all desperately need the community's attention. I spend many hours each week trying to cull the junk out of such articles, but I would greatly appreciate the help of anyone who has the time, skills and inclination to assist in the cleanup. 28bytes (talk) 19:16, 13 April 2012 (UTC)

Hitting "Random article" is a really easy way of finding these unloved pages, if anyone's not interested in working on a more curated list. Steven Walling (WMF) • talk 18:52, 16 April 2012 (UTC)
This is a good, but very slow, way of getting around this. We could do with a more curated list so we can actively pick out articles that we know about, and watch and help improve them. I spend an hour a day clicking random article to find something of interest and I just come across stub after stub of no interest. Mrlittleirish 10:04, 17 April 2012 (UTC)

December 2012

BEANS leads me to post this here.

We have a vandal who knows CSS. Little though I want to feed the troll, I have to say that his work screws up the presentation of a diff to his work, and its workings can only be seen in a diff from (undoing) his work (e.g. here).

I've given him a 31-hour break. I imagine that he wants to make fuller use of his skills. If or rather when he pops up again, don't hesitate to give him a longer rest from editing WP.

This post is of course completely irrelevant to the proper topic of this talk page. Feel free to delete it after a week or so, or of course to move it somewhere more suitable that I haven't thought of. -- Hoary (talk) 06:21, 2 December 2012 (UTC)

I've come across editors like this before, but the technique varies. I have applied WP:REVDEL to all edits by this user. --Redrose64 (talk) 09:45, 2 December 2012 (UTC)
Thank you. Just after posting this, I was going to post a link to it within WP:AN/I, but I there noticed the thread "Filter 139", which looks relevant. If I understand it correctly (and that's a big "if"), filter 139 is designed to prevent this kind of vandalism, but the IP found a workaround for it. I think that my understanding of CSS is up to the task of working out what's what, but I have never encountered filters before; all in all I don't want to tinker. So I sent email to Shirik (who started that thread), inviting him or her to look here. -- Hoary (talk) 12:23, 2 December 2012 (UTC)
Thanks for doing this and posting about it. FYI, this page is visible to anyone (even IPs) if they know about it - only UnwatchedPages itself needs admin privileges to view. Ruhrfisch ><>°° 19:43, 2 December 2012 (UTC)
Damn. I mean, thank you for telling me. -- Hoary (talk) 13:24, 3 December 2012 (UTC)
This page is even watched by some non-admins. But, hey, it's a fairly obscure page so don't worry about it too much... bobrayner (talk) 13:29, 3 December 2012 (UTC)
I was also like, wait - don't we have a filter for this? I temporarily unRevDel'd the edit on British Royal Family in order to examine it against the filter (otherwise it pretends that new_wikitext is empty), and it does indeed match 139. It's either some workaround (in which case it's a major bug) or more likely just that they got lucky and hit the condition limit. -- King of 09:35, 3 December 2012 (UTC)

Toolserver watch

You'll need to sign in with your watchlist token and need 500+ edits or be listed on m:Toolserver/watcher. Watchers are considered inactive if they haven't edited or logged in the past 30 days. The inactive watcher article list (enwiki only) is updated once a month. Discuss. — Dispenser 23:42, 1 August 2012 (UTC)

If you're wondering why the index wasn't updating for months, it's because I was diagnosing (and forgot) a bug that excluded unwatched pages with zero inactive watchers. This has been corrected and I've implemented and tested the monthly updating code. I'm sorry if this error had any negative effects. For those curious, the problem was the NULL from the LEFT JOIN cascaded through SUM() and broke numerical inclusion comparisons. — Dispenser 05:59, 7 December 2012 (UTC)

Article limit again

Just wondering if anyone was working on this? KillerChihuahua 14:01, 8 December 2012 (UTC)

MZMcBride would be the most likely to know who, if anyone, is working on this. I'll see if a Bugzilla has been filed later. MBisanz talk 16:22, 8 December 2012 (UTC)
I filed a Bugzilla (or two?) several years ago - this seems to be very low on the priority list. Ruhrfisch ><>°° 18:29, 8 December 2012 (UTC)

Pending Changes

Now that Pending Changes is available again, how about activating Pending Changes on these articles? Andreas JN466 16:25, 24 January 2013 (UTC)

RfC: Visibility of the special page

Should non-admins be given the right to access the Special:UnwatchedPages special page and if so, which ones? smtchahaltalk 08:34, 27 May 2013 (UTC)

There already has been a discussion about a year ago regarding the visibility of this special page, and although the majority of users (though not very numerous) were in support of allowing non-admins to view it, it wasn't clear exactly which ones and the change was hence never implemented. I thought I should bring it up, because I personally believe that the visibility of this page is that big deal. Of course that doesn't mean everyone should have access to it; there could be a separate user right flag the members of which could access this special page, or maybe rollbackers and/or account creators could be given this right by default. smtchahaltalk 08:35, 27 May 2013 (UTC)

  • Support - This was one of the reasons I wanted the admin tools was to help support some of the special pages that non admins cannot see. So if this was a user right, that would allow people to see these, then that would be one less reason to need to go through RFA again in a couple months. I don't think it should be totally public because there is a lot of room for abuse but it could and should be a user right like Rollback or File mover. Kumioko (talk) 15:53, 27 May 2013 (UTC)
  • Oppose - the work needed to create and manage another user right isnt' justified by any benefit of allowing non-admins to see this page - especially given the 1000 article limit that's been discussed many times on this page. This page really isn't all that useful. Creating a whole process just so that non admins can view the page, discover that for themselves, and then get on with their lives is unnecessary. Trust me: it just isn't worth it. WaggersTALK 07:58, 28 May 2013 (UTC)
I do agree creating a separate user flag just for this special page isn't justified. How about giving the right to existent user flag members, like rollbackers? Rollback is apparently somewhat a big deal, and it does seem relevant here; a user with a good knowledge of what vandalism is and what's not and can be trusted with rollback should also be trusted with this special page, right? This page may not be useful for preventing vandalism, but it can be useful for improving pages that are not bothered about. Whatever it is, I still strongly believe that this page – useful or not – should not be accessible to administrators alone, because administrators apparently don't have much use for it. Besides, how is this page useless if it is also a potential target of abuse (to the point that only administrators are allowed to view it)? smtchahaltalk 08:22, 28 May 2013 (UTC)
I could support giving access to rollbackers. To answer that last question - the page would be extremely useful to vandals to see for obvious reasons - that's why it's restricted. But as it stands it's less useful to constructive editors. If the unwatched pages could be categorised somehow, and of course if we could see the full list, it would be much more so. WaggersTALK 08:48, 28 May 2013 (UTC)
I guess that if even rollbackers (there are over four thousands of them; that's about four times the number of administrators) were allowed to view this page, there could be fewer unwatched pages and hence more unwatched pages available to be viewed. smtchahaltalk 09:03, 28 May 2013 (UTC)
  • I would support moves to open this page up to more editors. I think that the 1000-item comment is missing the point; it's a display limit, not an absolute limit on how many unwatched articles can ever be known. If I took 1000 unwatched pages and added them to my watchlist, subsequent viewers of this page would see different pages listed here, and most importantly there would be 1000 fewer unwatched pages. bobrayner (talk) 22:53, 29 May 2013 (UTC)
  • Support as long as it is limited to an existing class of editors like rollbackers (do not make a new class just for this). I just checked and the 1000 article list only goes as far as articles beginning with "197..." Ruhrfisch ><>°° 03:01, 30 May 2013 (UTC)
  • Support for rollbackers. No point creating a new right. AIRcorn (talk) 11:15, 30 May 2013 (UTC)
  • Support for rollbackers (or a similar group, whatever), but I do believe this is pointless. I don't believe that the 1000-item display list will ever be whittled to the point where anyone can see even all said articles beginning with A, and perhaps, it'll never even get to the point where we can see the first such entry. It's my recollection that watchlists start getting unusable around 10,000 entries.... --j⚛e deckertalk 20:25, 30 May 2013 (UTC)
Mine's at 13,000. Do you mind if I opt out? --Redrose64 (talk) 20:41, 30 May 2013 (UTC)
I have a little under 22, 000! Kumioko (talk) 20:43, 30 May 2013 (UTC)
Wow! Okay, I've kept paring mine back to 9,000 or so every time it got near 10K. I stand corrected! RedRose... I don't think anyone was suggesting that anyone have to look at the special pages, or change their watchlist. Maybe I'm just being confused, that happens a lot.  :) --j⚛e deckertalk 21:39, 30 May 2013 (UTC)
Well, Bobrayner did put "If I took 1000 unwatched pages and added them to my watchlist ...". --Redrose64 (talk) 21:45, 30 May 2013 (UTC)
How on earth can one have a watchlist that big? When would one sleep? Herostratus (talk) 04:53, 2 June 2013 (UTC)
My watchlist currently covers 3556 pages. These are mostly pages vulnerable to neutrality problems so they have have a modest level of activity. Pages which haven't yet acquired any watchers are likely to be much quieter (I suspect many of them will be rote-created articles about obscure villages and retired sportspeople &c) so 1000 of these wouldn't make much difference to my watchlist workload. — Preceding unsigned comment added by Bobrayner (talkcontribs) 10:58, 2 June 2013 (UTC)
  • Jeez, one result of this would be for some people at Wikipediocracy or 4chan or wherever to get whatever right is necessary to access this page and publicize it among their buddies. This would be a vandal's or troll's dream would it not? (Of course, Wikipediocracy may already have some admin accounts, so maybe this wouldn't matter. (As a practical matter, I'd like to see this list so I could take some articles off it.) Herostratus (talk) 04:53, 2 June 2013 (UTC)
  • Yes; we have to balance that against the abuse of unwatched pages in a very bad manner, which is an ongoing problem. Recent changes patrol &c is a fairly effective control but doesn't catch 100% of abuse; watchlists are better at catching abuse but don't cover 100% of articles; increasing the reach of the latter control would widen a different vulnerability (which already exists; it's not new). We have to make a decision on the margin. Personally, I think that opening up this list a bit more would be a net positive as long as we still have reasonably high standards for who can see it. bobrayner (talk) 18:38, 8 June 2013 (UTC)
  • Support I can see the benefit of having another group of users being able to access this list because it will lighten the load on the admins. -- MisterShiney 17:03, 4 June 2013 (UTC)
  • Support per Ruhrfisch. If the list is that long, clearly it's grown beyond the control of the admins. I don't want to open up opportunities for vandals, but I would guess the more clever nefarious characters have already gained access. It's high time we sweep out the cobwebs. Chris Troutman (talk) 04:01, 7 June 2013 (UTC)
  • Comment: I have 2508 main space article names in my Watchlist, but I'm not watching them! My first "watch list" is the articles in my most recent 500 contributions since changed by someone other than me. I think we need a better definition of unwatched: specifically pages edited by IPs and other possible vandals (always assuming good faith), that have not been automatically reverted by bots, that have not been viewed by N logged-in users (not including possible vandal accounts) where N may vary on the size of the article. Mark Hurd (talk) 04:13, 9 June 2013 (UTC)
    That might not be a helpful definition, since listing articles according to those criteria would require a great deal of complexity and technical change, whilst we already have an existing resource which shows all pages that aren't watchlisted. Perhaps, if you're not actually watchlisting them, it would be better to remove articles from your watchlist rather than reëngineer wikipedia to suit a new definition of watchlisting which isn't actually based on watchlists? bobrayner (talk) 12:26, 9 June 2013 (UTC)
    You may be right that I should consider removing my watchlist, but what about all the other editors that haven't checked their watchlists, as mentioned by Ruhrfisch below. An updated version of this page should at least consider only counting watchlists that have been checked recently... Mark Hurd (talk) 02:09, 12 June 2013 (UTC)
    It would be nice if that became technically possible, but in the meantime we don't need this to be an absolutely watertight control; if we reject improved controls whilst waiting for a perfect control, there's something very wrong with our approach to risk. Our other controls have their own gaps - for instance, there's no guarantee that every recent change is seen by an RC patroller, let alone that they understand what they're looking at - but we can live with that because we have multiple, overlapping controls which collectively do a good job of reducing the risk of malicious edits. Watchlists are already a good line of defence; not perfect, but good, and this proposal would narrow the gap by reducing the number of pages which are not effectively watchlisted. bobrayner (talk) 10:39, 19 June 2013 (UTC)
    Users watching pages aren't always active indeed, but this also duplicates work. There is no good way to have an effective patrolling system without a way for 'trusted' users to manually mark an edit as patrolled and having a list of unpatrolled pages (never patrolled) and oldpatrolled pages (previously patrolled, but not the most recent edit), as suggested at WP:Patrolled revisions. Unfortunately, this would need development. Cenarium (talk) 02:53, 29 June 2013 (UTC)
  • Comments There has been no technical support for this page in the 5+ years I have been an admin (and no real progress on the Bugzilla requests I filed either). This page is currently of little use because only the first 1000 unwatched articles are listed. I just checked the edit histories of 5 random articles from this list. None had any edits marked as reverting vandalism (human or bot) and only one was above what I would consider stub level - all had edit histories that displayed in one page (and I only looked at the edit summaries, nit the actual edits). One suggestion that has been made here in the past is to make a special account that is just to monitor these unwatched articles. Another observation is that an article may be on the watchlist of one user who has not edited in many years, but will not show up here. Ruhrfisch ><>°° 13:09, 9 June 2013 (UTC)
I would also suggest that this list probably contains some redirects and Disambig pages that need not be on there. Perhaps someone can do some investigation into that? Kumioko (talk) 02:30, 12 June 2013 (UTC)
No, redirects and disambiguation pages need watching too: more savvy vandalism can be hidden in those for much longer if they're not watched. Mark Hurd (talk) 03:11, 12 June 2013 (UTC)
Good point, thanks for checking. Kumioko (talk) 03:29, 12 June 2013 (UTC)
I checked and seven of the first 500 entries have the word "disambiguation" in the title. Ruhrfisch ><>°° 03:35, 13 June 2013 (UTC)
  • Question What are the reason(s) to not just give access to everyone? CombatWombat42 (talk) 02:35, 13 June 2013 (UTC)
  • Oppose. It's useless in its current form (according to the admins that have it), but a nefarious account or 2 with just "rollback" permissions could easily start adding 1,000 items at a time to his own watchlist, until tens of thousands of items are mischievously removed from this potentially useful feature. –Quiddity (talk) 21:45, 14 June 2013 (UTC)
That's a point, but I guess it's reasonable to expect rollbackers not to do that (because rollback right is apparently taken quite seriously). smtchahaltalk 09:56, 19 June 2013 (UTC)
  • This will be a hugely unpopular position, but I think the page should be open to everyone, completely unrestricted. I'm surprised at the number of people here advocating security through obscurity. NinjaRobotPirate (talk) 20:31, 19 June 2013 (UTC)
  • Yes, I can see that argument; it's something that I considered. The problem I have is that we're assuming that hiding information makes these pages more secure. It may very well do so, but having an army of anti-vandals with access to the information would undoubtedly help, too. However, since Wikipedia is not an experiment in democracy, I doubt that we'll ever find out. NinjaRobotPirate (talk) 01:40, 20 June 2013 (UTC)
  • Once someone adds one of these articles to their watchlist, it disappears from this page the next time it is updated (so the "army of anti-vandals" would no longer have access to the same information). This means a patient vandal who had access to the list could add many of these articles to their watchlist, then wait for the list to be updated, and then essentially have free license to vandalize (as long as they did not make edits of the kind that one of the bots catches). While many of these articles are low traffic, there are some that I assume vandals would love to know they could mess up with slight chance of detection. Ruhrfisch ><>°°
  • There is a risk that a patient and subtle vandal could gain access to this list and have the same "free license" that every editor currently has - that is to say, they're still at risk of getting caught by RC patrol &c. (And once one suspicious edit is caught, past contribs get investigated too, so long-term accounts have rarely been responsible for much vandalism). I think this risk is outweighed by the benefit of genuine watchlisting by a much larger number of editors. bobrayner (talk) 03:11, 28 June 2013 (UTC)
  • Support for rollbackers per Ruhrfisch etc.. However, I suspect that the list is so huge its not going to make a lot of difference. It is indeed all but useless at present and none of the various technical solutions proposed over the last 5 years have ever been actioned. Ben MacDui 12:53, 23 June 2013 (UTC)
  • Support for rollbackers. Glad I read the Signpost this week. With the number of unwatched pages we have on this wiki, more people having the ability to know this and keep an eye on them is exactly what we need. TCN7JM 22:37, 27 June 2013 (UTC)
The following discussion is closed. Please do not modify it. Subsequent comments should be made in a new section.

What about an option to store them to a page that can be viewed using the related changes feature? That way, you can still have your personal watchlist that you care about, and also watch these millions of pages. Thoughts? -Mysekurity 07:03, 6 January 2006 (UTC)

Second. Having admins add 100+ pages to their watchlist just doesn't scale. — Ambush Commander(Talk) 03:35, 19 January 2006 (UTC)
Absolutly. I would watch even more pages if I didn't have to weed through them to see my "really important" ones. On a somewhat related not, are we ever going to get to see what comes after the first 1000? I for one would like to know if Yukon is watched...or even Advertise- Trevor MacInnis (Talk | Contribs) 05:24, 19 January 2006 (UTC)
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.

(Merged with much earlier section with the same name.02:43, 15 April 2012 (UTC))

Here's an idea. I'd have to look into the actual technical implementation, but there must be a way to make this list work with Related Changes. That way anyone with an interest in checking these pages can simply pull up the list whenever they want, and there's little if any security concerns.
— V = IR (Talk • Contribs) 02:40, 15 April 2012 (UTC)

As an aside, to answer that earlier question from 2006, Yukon has 106 watchers at the moment. waggers (talk) 15:10, 16 April 2012 (UTC)
117. Respectfully, Tiyang (talk) 11:00, 1 July 2013 (UTC)

Yet another opening proposal

How about this: The UnwatchedPages itself is only visible to admins, but rollbackers can request visibility for pages relating to a certain topic (i.e. relating to their WikiProjects). Looking through the discussions above, I've seen a lot of back-and-forth about this issue, but it seems like a restricted opening would really help to lessen the load and this idea hasn't been tried yet. Course, then the issue would be knowledge of this page (which I only found through the Counter-Vandalism Unit), but that seems like a minor one compared to the others. Supernerd11 :D Firemind ^_^ Pokedex 00:23, 28 March 2014 (UTC)

Alas, I think that could be difficult to implement, because the underlying data is just a flat list of pages - it's not a table with columns for category, wikiproject &c. So, how would Special:UnwatchedPages know which ones to show you? bobrayner (talk) 00:32, 28 March 2014 (UTC)
@Kevin Gorman, Anna Frodesiak, Materialscientist, and Kudpung: First, the earlier proposal (unwatched pages should be opened to rollbackers) RfC was never closed although a bot removed the RfC banner, so I would posit that the consensus supported such devolution of admin powers. Second, I supported that proposal and would support any proposal that lowered our potential vulnerability. Finally, the one snag is that even if you went through manually to determine which articles belonged in which WikiProject, I'm not sure all those unwatched pages belong in a WikiProject and many WikiProjects have been only semi-active or completely inactive since users started leaving in 2009. (For example, you might have a AfD discussion that is unwatched, doesn't have an identified WikiProject, or falls into a WikiProject like History that's dying a slow death.) Chris Troutman (talk) 00:52, 28 March 2014 (UTC)
  • I think Bob and Chris both raise good points re: the feasibility of making it Wikiproject-specific. Both previous proposals did have pretty strong support amongst those who commented, but didn't receive very many comments for a proposal that involved devolving admin powers, and are both stale enough that I at least wouldn't want to try to close one as successful. From a practical standpoint: this isn't currently a very useful page, and would require something approaching a complete rebuild in order to become useful - unless the Foundation or a volunteer dev wanted to pick up supporting it, I don't think there's much point in debating whether or not we should devolve it to rollbackers - it's just too useless to be worth the words spent discussing it. To give you an idea, currently the last unwatched page you can see is 1961_Wilkes_200 - it doesn't even extend the list far enough to see pages that don't start with a number. Kevin Gorman (talk) 01:32, 28 March 2014 (UTC)
The sad truth is that this page itself lists only 1000 entries and does not even get to unwatched pages starting with the letter "A". There is no interest in the part of the developers and powers that be in making this easier to use (there has been a Bugzilla request in to fix it somehow for years and years). Ruhrfisch ><>°° 01:35, 28 March 2014 (UTC)
That is a separate problem. It would be good if we could fix list depth or allow people to start at an arbitrary point in the alphabet; it would also be good if we could let a wider range of trusted editors access the list. It would be even better if we could do both, but they are separate problems, and the existence of one is not a reason to avoid solving the other. bobrayner (talk) 01:43, 28 March 2014 (UTC)
What about implementing a search function for just the UnwatchedPages, like WP:AFC's got? At least some that are later in the alphabet (or in the alphabet at all) could be put on a watchlist, if only found by dumb luck. Supernerd11 :D Firemind ^_^ Pokedex 01:53, 28 March 2014 (UTC)

How does this work?

What's the metric for counting a page as "unwatched"? Zero total watchers? Zero active-editor watchers? Or some threshold low number of watchers?

And why can I see only the first 5000 of these articles? I am sure there are more, because #5000 has a title starting with the character "1" just like the 4999 before it. ~Anachronist (talk) 23:45, 12 October 2017 (UTC)

Anachronist - Yes, pages listed here have 0 watchers - no user accounts have the pages listed here added to their watchlist. I know that this is an old discussion, but I wanted to answer the question nonetheless :-). ~Oshwah~(talk) (contribs) 17:18, 31 August 2018 (UTC)