Wikipedia talk:Page Curation

From Wikipedia, the free encyclopedia
Jump to: navigation, search

Page Curation tool bar[edit]

My Page Curation tool bar is not appearing check.Hison Here (talk) 08:48, 21 May 2014 (UTC)

Yeees, you reported this above. Did you try searching for a 'curate this article' link on the side? What browser/OS are you using? --Ironholds (talk) 04:53, 24 May 2014 (UTC)
My curation tool bar is stoped showing from last 2 weeks.Hison Here (talk) 12:53, 24 May 2014 (UTC)

French Wikipedia[edit]

This extension has been mentioned on the French WP Village Pump and I said a consensus should be shown before we ask for it, so a poll page was started and so far almost everybody seems to want it. Next day on the talk page someone noticed that bugzilla:48552 to make the extension "agnostic" and suitable for our WP has the lowest priority, and on bugzilla:42322#c1 in a similar case details were asked about the local deletion process but the installation was not done.

Could somebody tell us more there on the poll talk page: is this hopeless, or could it be adapted to our deletion process, or could we just ignore the deletion part and keep the rest? If you do not know French, there are enough people on the French WP who know English and will be willing to translate. Thank you in advance: Oliv0 (talk) 05:40, 24 May 2014 (UTC)

Toolbar is AWOL[edit]

Working on this page: https://en.wikipedia.org/wiki/Kirby_Burkholder

No toolbar popped, no "Curate this page" link on the lefthand side. Chrome, up to date. Also, I noticed on another page that the toolbar only showed up if I opened visual editor and then closed it. Save me Ironholds! Nathan T 21:37, 28 May 2014 (UTC)

I'm not seeing it either - but it was marked as reviewed on 8 May. Is it appearing in NewPagesFeed despite that? Ironholds (talk) 07:38, 30 May 2014 (UTC)
I have the same trouble with Not Censorship, But Selection. No toolbar, no link to mark the page as checked, but for this page the log says it's not been checked. QVVERTYVS (hm?) 19:21, 13 August 2014 (UTC)
And there wasn't a "curate this page" link? It was apparently reviewed, but after this post. Ironholds (talk) 12:44, 15 August 2014 (UTC)
I'm not seeing this toolbar anywhere. I remember using it at some point, but I haven't seen it in ages. (Now I can't unreview pages either.) QVVERTYVS (hm?) 18:58, 25 August 2014 (UTC)
Hold on, I found it! I had to set the skin to Vector and language to English, and then Page Curation re-appeared under "tools" (not "toolbox" as this page suggests). QVVERTYVS (hm?) 19:11, 25 August 2014 (UTC)

I'm having trouble[edit]

Hey friends, I'm having a trouble. I want to add something in the script used by Page curation. I frequently use Page curation to review pages. And honestly I always use Page curation to request articles for CSD. But I often face a problem to keep track of those article. (I sometime have 100 edits per day) And I have 1800 articles on my watchlist. So, I can't track if the article is deleted or the deletion tag has been removed by the author. So I want to add some code to the MediaWiki script so that it can create a subpage similar to User:Jim Cartar/CSD log created by TW. And it will update whenever Page curation is used to add a CSD tag. I think it will also help other users to keep a record. Jim Carter (talk) 18:17, 3 June 2014 (UTC) P.S. Just leave me a talkback when some respond to this proposal. Thank you. Jim Carter (talk) 19:11, 3 June 2014 (UTC)

Well, that seems reasonable, but it's not actually a script - it's a full extension, subject to all the normal code review processes and deployment protocols as much as any element of MediaWiki. So it might be a pain. One thing I can suggest doing is looking at the page curation log; if the link is red, it got deleted, if not, it's been denied or the user has removed the tag (although there's actually a bot that should be catching tag removals by the author). Ironholds (talk) 21:51, 3 June 2014 (UTC)
Sounds good. Thanks Ironholds. Jim Carter (talk) 06:53, 4 June 2014 (UTC)

Another[edit]

I have another problem. I often use my mobile phone (Android 4.2) to edit pages (even this edit is made by my phone). So I was thinking if MediaWiki developers can enable Page curation for mobile too. Thanks. Jim Carter (talk) 07:06, 4 June 2014 (UTC)

Hmn. I don't think that's currently on the books, although it would be nice. I know the design (big round buttons) was intended to be used on tablets, at least. Ironholds (talk) 15:10, 4 June 2014 (UTC)
I believe the mobile interface is in the process of being rewritten as a skin, so that all your functionality (such as user scripts) works properly. For now I think you have to use the desktop skin (a link at the page bottom) in order to use this tool. —Gryllida (talk) 05:17, 9 July 2014 (UTC)

WikiProject tagging and rating[edit]

PageCuration should support tagging and rating pages during review. A user script does the tagging and rating. I think that these two processes — tagging at least — would be useful to do during a page review. --Gryllida (talk) 05:14, 9 July 2014 (UTC)

Gryllida, this was previously discussed in Wikipedia talk:Page Curation/Archive 1#wikiprojects, where WereSpielChequers noted that WikiProject tagging wasn't all that important (relative to categories), as the active WikiProjects patrol their categories of interest anyway. APerson (talk!) 13:03, 7 August 2014 (UTC)
Actually I really like that suggestion. Would be so useful to be able to tag something with the right wikiproject(s) during the review process. It's very easy to miss doing so and then you get the occasional gentle scolding message on your wall for not doing so.... Mabalu (talk) 13:29, 7 August 2014 (UTC)
A reasonable enough suggestions but unfortunately many patrollers refuse to use the features curation already has and most patrollers won't do everything they're supposed to do anyway. Until NPPer becomes a coveted user right for the hat collectors it's unlikely that patrollers' performance will be improved. Kudpung กุดผึ้ง (talk) 11:48, 20 August 2014 (UTC)

Some non-obvious things[edit]

Hi all. I discovered this tool today when a summer student I'm looking after was told that their user page had been reviewed. The notice didn't make it clear what the review was about, and it looked initially like it was a review of the user rather than of their user page. It would be good to make the echo notification a bit more obvious, and perhaps so that it links back to this page or a general page about what 'reviewed' means?

Also: enabling the curation toolbar is really non-obvious. Shouldn't it appear under Preferences/Gadgets or Beta? If it's there, I can't spot it. It seems that you turn it on by visiting the new pages feed, then clicking on one of the articles, after which it sometimes appears and sometimes doesn't, and not always in the same place on the page? (At one point it appeared on the bottom-left of the page rather than the right-hand side for some reason.)

Hope this feedback helps! Thanks. Mike Peel (talk) 15:58, 6 August 2014 (UTC)

It sometimes doesn't appear? I mean, it should always appear on new articles. Ironholds (talk) 23:46, 6 August 2014 (UTC)
Mike, there should be a link that says "Curate this article" under the "Tools" section in the left-hand sidebar.
About the bottom-left position: I've gotten that error a few times myself, and am interested in knowing what causes that. APerson (talk!) 12:50, 7 August 2014 (UTC)
APerson - I can't see that link... Thanks. Mike Peel (talk) 14:42, 7 August 2014 (UTC)
Mike, it might have already been reviewed, then. APerson (talk!) 15:43, 7 August 2014 (UTC)
Hmm, OK, transient links are confusing. :-/ Mike Peel (talk) 16:26, 7 August 2014 (UTC)

The Page Info panel should take into account tags on the article[edit]

Steps to reproduce:

  1. Open the Page Curation toolbar on an uncategorized article.
  2. Tag the article with {{Uncategorized}} (with the "Add Tags" panel).
  3. Wait for the page to be reloaded.
  4. Open the "Page Info" panel.

Expected behavior: The "Page Info" panel shows that the page has no categories but that the page is tagged as such (perhaps by not using the red text and/or suppressing the red number on the toolbar).

Actual behavior: The "Page Info" panel ignores that you've already tagged the page, adds "No Categories" to the "Potential Issues" section and shows an alert on the toolbar.

APerson (talk!) 12:47, 7 August 2014 (UTC)

The effect of this tool on newcomers[edit]

I've been looking at the articles by new users recommended for deletion in the last couple of days and am rather alarmed to see that 40% of these recommendations are made within 10 minutes of the article first appearing (based on a sample of about 50). Indeed I've seen 8 examples where it's happened within 2 minutes. I think it's fairly obvious that this behaviour will have a very negative effect on the new user. In the example that alerted me to it, they hadn't made another edit in 4 months. I won't be able to give a comparable estimate for a month or two as stats are unavailable to me once the article has been deleted.

I'm not arguing about the intrinsic worth of any of these articles but rather that the tool is too powerful in the hands of inexperienced editors. e.g. when one presses the Delete button on the Curation Toolbar, it alerts the user that the article is only 2 minutes old, but fails to point out that it's by a new user (although stated policy is that it shouldn't happen within 10-15 minutes.) Even experienced editors make mistakes when they first put up a new article and if WP was fulfilling its duty of care, the response to the first mistake should be an offer of help not a brusque "We don't want you" message. That may be impracticable but one thing that's obvious is that this tactic is hurting particularly those from developing countries whose first language isn't English. The documentation for these tools have even omitted the old Don't bite the newcomers adage from the New Pages page.

Even in a couple of days I'm learning to distinguish those editors who make timely and helpful suggestions from those who don't seem to take any care. Is the use of these tools subject to any monitoring by WP? If it isn't, could I suggest that such monitoring is needed. Chris55 (talk) 15:21, 18 August 2014 (UTC)

Well, I'd make three points.
The first is; 50 people does not a sample capable of statistical significance make. The standard there is 0.95 so you'd need...a minimum of double that even before we get into things like standard error. This is not to trivialise your experimence, merely to note that it is not necessarily representative. In fact, yes, a lot of newcomer articles are deleted - User:EpochFail does a lot of research in this area and I think the percentage I've heard him throwing about is around 80%.
Second; NPF is indeed powerful. But the workflow mandates a message to the user that makes clear how to appeal and who to talk to. The alternative is a mix of Special:NewPages, which uses much more complex messaging and doesn't always mandate them at all. I'm not saying it's great, but it's better than the status quo - and the "brusque we don't want you" message is actually based on kind of a lot of A/B testing as to how to make the messages as non-discouraging as possible. There is, of course, a limit on how far you can do that when they're negative messages :(. If you have suggestions for improve them, though, note them! The developing countries point is interesting - I have access to our db in my professional capacity and so could actually run an observational study there if you'd be interested in the quantitative results.
Third; yes, it is subject to monitoring by the community. You'll note as well as review, it contains an unreview button, and everything is very deliberately logged. Ironholds (talk) 02:53, 20 August 2014 (UTC)
Your first point is obviously true. On the 2nd, I've suggested the tool could identify newcomers (including when the next button is used). It could also delay listing them for 5 or 10 minutes. I realise there are some categories of attack articles that need to be dealt with immediately, but your citing of EpochFall's findings are frightening. The Delete button is very tempting and in some ways needs less thought to use it than some of the others. Maybe only one or two possible reasons should be available for very young articles. Some criteria need more time to be evaluated. That's for starters. Chris55 (talk) 08:04, 20 August 2014 (UTC)
Hey folks. Thanks for the ping. Chris55, I'm stoked that you are looking into this. I agree that it is a problem. The mean time between edits is 7 minutes and the median time between edits is more like a half hour (due to log normal distribution), so if we CSD tag something that is less than half our old, we probably (1) caused an edit conflict that confuses the new editor and (2) tagged something while the work that would make the tag unnecessary was being done anyway. Given that most CSD's are A7s (no assertion of notability) and not spam/vandalism/attacks, there doesn't seem to be a need for this haste. I think that a critical misunderstanding baked into the NPF is that, once a page is created, it is immediately ready for review. If we could somehow delay NPF for 30 minutes after page creation I think that we'd see higher quality page creations at time of review (therefore less deletions) and more retained good faith newcomers. This seems to be worth an experiment. Ironholds, do you know who I might contact to pitch this idea and run a limited test? --EpochFail (talkcontribs) 08:21, 20 August 2014 (UTC)
Maybe Ironholds can point us to the maintainers. A possible improved design would have several 'entry points' to the process: "immediate" only for flagging attacks, copyvio etc, but a second at say 30 minutes or 1 hour which allows for a fuller range of reviews. Many people would prefer to do the second, but the present tool only allows a single entry point. A third entry at 7 days to review the effect of the first two would also be useful. Chris55 (talk) 09:38, 20 August 2014 (UTC)
NPP is our first and most important firewall against totally unwanted rubbish - and that is unfortunately a very high percentage of the 1,000 or so non-autopatrolled pages that get created round-the-clock every day. Thanks to Ironholds and a few others who fought for and worked on the development of New Pages Feed and the Curation Toolbar, it was the best piece of core software to hit Wikipedia since sliced bread. Unfortunately it's still only as good as the people using it and that has been the main problem all along because ironically, such 'maintenance' areas are magnets to very young, very new, and very inexperienced users.
Until NPPer becomes a coveted user right for the hat collectors, it's unlikely that patrollers' performance will be improved. I believe that before any improvements are made to the software. we should be looking at that angle - and until that happens we will continue to lose some newcomers who create good faith but inappropriate pages, who might have nevertheless stuck around and created something worth keeping. Kudpung กุดผึ้ง (talk) 12:16, 20 August 2014 (UTC)
I really like the suggestion of delaying the article going up for review for 30 minutes or so, to give it a chance to shape up properly - in principle it's kinda awful that a page can be trashed on within seconds of creation, before it's had a chance to be edited up properly. I do wonder how many of the 80% deleted pages by "newcomers" were created by sockpuppets or disruptive editors previously banned under other IDs - probably not that many really! I thought Kudpung กุดผึ้ง's comment in their edit summary that it's usually new users who patrol pages was an interesting point too. I have certainly seen quite a few pages "approved" through review that shouldn't have been. When I've (not all that regularly I fear) been page patrolling I try to look for all new pages without categories in order to add appropriate cats (particularly important for BLPs) and have come across quite a few inappropriately cleared pages such as completely unreferenced BLPs or obvious copyright violations that were waved through. It's been mentioned qute a few times, but when you tag a page with issues, it REALLY should NOT automatically mark it as reviewed unless you want it to be reviewed. If such tags added during the review process are removed then the page should automatically be marked unreviewed again - if the issues have been fixed then it should pass. Mabalu (talk) 13:15, 20 August 2014 (UTC)
Whenever I can, I spend about an hour on NPP a day. Not to patrol the pages, but to patrol the patrollers. On average I have to ask at least one patroller a week to stop patrolling until they have either leaned to read and write English or reached 8th Grade. On very rare occasions I have had to threaten topic banning or even blocking until they understand that Wikipedia is not a game. The sooner the patrollers realise that Wikipedia is neither a MMPORG nor NPP an old fashioned pin-ball machine, less crap will slip through the net, and fewer good faith creators will get bitten. Kudpung กุดผึ้ง (talk) 14:05, 20 August 2014 (UTC)

────────────────────────────────────────────────────────────────────────────────────────────────────@Kudpung: you've provided this theory (that we have a large number of patrols by inexperienced editors, or a large number of inexperienced patrollers) before. I've gathered data to validate or invalidate it and found it incorrect, and presented that data to you. In the absence of an explanation as to why the data is missing something I would suggest making a different argument.

For the sake of argument I did some more recent data-delving, specifically looking at the last 3 months of patrols (for anyone who reads R, the script can be found in this gist).
Working out the density of patrols or patrollers by edit count is difficult because we have a few editors with extremely high edit counts, which pushes things off. But if we log-scale the values we find that most patrollers, and most patrols, are done by people in the middle of the range of edit counts. Generating quantiles for each class (distinct patrols, versus distinct patrollers) indicates that that middle-of-the-range is between 500 and 11k edits - only 25% of editors had edit counts below 470 edits, and only 25% of patrols were made by people with fewer than 1017 edits.
Patrols that got reverted/unreviewed is a smaller dataset (only ~1k results, rather than 20k, in that period) and the non-log-scaled results are similarly non-useful, but log10s of the reverted editors and the reverted patrols again shows that they're in the middle of the range; newbies are not the people who get things wrong. In fact, quantiles for unique editors who were reverted show that only 25% of them had editcounts below 712 edits, and quantiles for reverts overall puts the lower 25% bar at 880.
So if there are problematic patrols, they're less than 5% of cases. And they're mostly made by people more experienced, in terms of edit-count, than the base level of patroller experience. Ironholds (talk) 23:19, 20 August 2014 (UTC)
Oliver, yes we have had this discussion many times over the last four years. Unfortunately, 'only' 25% of editors had edit counts below 470 edits, and 'only' 25% of patrols were made by people with fewer than 1017 edits rather tends to prove my points (i.e. therefore, at least 25% of patrolls are potentially incorrect) rather than discredit them; furthermore, your stats do not and cannot take into consideration patrollers' age and/or maturity, block logs, rejected creations, CSDd/PRODed creations, other templated warnings or informal polite requests to improve whatever they are doing, and the NPF software does not mandate anything at all - most patrollers only do the bare minimum, even addressing only the low hanging fruit, hovering with their finger on their mouse button every time they refresh the feed, and that's why they pounce too soon, hence my analogy with computer shoot-'em-up games.
AFAIK, I am the only admin (or any other user for that matter, perhaps with the exception of DGG) to continue more or less systematically monitoring the quality of the NPP process for years, and if I were humanly capable of patrolling every single patroll and the editing history and user page of every single patroler, you will find again that your 5% statistic is way, way, under par. This underlines yet again that some contractors and/or staff are often reluctant to accept what the community is telling them and prefer to rely on some math and scripts that bear no relation whatsoever to reality. Ironically, we need more people to patrol the patrollers, which I am sure you will agree would be kind of ridiculous. What we do need is some kind of encouragement to patrol along with some clearly defined criteria of experience.
In contrast to the Foundation who still appear to be only interested in reporting overall growth in user registrations and page creations, as a volunteer community member I am committed to examining the quality of new pages as well as the competency of those who patrol them. At the moment, NPP in spite of its excellent new software, is still practically as dysfunctional as it ever was; far too few genuinely competent editors are prepared to go there, and even fewer exploit all the excellent features that the Foundation built into it. Based on what they actually appear to do whether experienced or not, the vast majority of of patrollers have never read WP:NPP and don't and won't until they are asked by an admin to do so.
We have a similar issue with AfC for which however, on my initiative, at least a minimum set of qualifications has been adopted for reviewing article submissions. NPP is by far the more important process and we cannot rely on surmised WMF stats that attempt to demonstrate that there is nothing wrong with the quality of patrolling. With all due respect, I would suggest that the Foundation leave these issues for the community to identify and resolve, and just address the bugs/improvements in the software as and when required. Kudpung กุดผึ้ง (talk) 05:13, 21 August 2014 (UTC)
Re paragraph 1: you're saying not many edits equates to potentially incorrect? Again. people whose patrols are reverted tend to be more experienced than the norm, not less. There's no real indication from the data that it's the newcomers screwing things up. Re paragraph 2: well, first, contractors are staff, second, you'll note this is my personal account, and third, you'll note that it's not like I'm inexperienced at patrolling. If you're arguing against scripts and for anecdata, I have a big pile of anecdata too, and it doesn't agree with your conclusions ;p. It's clear from your post, however, that this is not going to be a productive conversation, and so I'm going to bow out of it rather than waste the time of you, myself and the potential audience. Ironholds (talk) 14:53, 21 August 2014 (UTC)
Lately I've been focussing on AfC more than NPP, but I check NPP every few days. I don't review systematically (tho I did at the old NP a number of years ago), but try to spot someone doing it drastically wrong and checking what else they have been doing there. Kudpung's results are correct, and further studies will verify the details--we do not need such a high standard as .95 significance to be reasonably sure we need improvement. Waiting for more studies when there is a serious problem, and there are obvious partial solutions, is the typical bureaucratic device to avoid addressing a problem. Given that WMF is inherently a bureaucracy, I prefer solutions the community can adopt on its own. I see this discussion as a consultation on what would be best to propose, or at least best to first propose as a partial solution. step.

We should require the patrollers to demonstrate competence, and to remove from patrolling those people who do not show it. Access to NPP and the curation bar should be limited to the same standard as AfC. This is a direction that is worth pursuing. (it's limited, because the raw NP log is still available, but that's not so attractive to beginners) Kudpung, your opinion? Ironholds, we need your continued input. You have more experience than Kudpung or I on what is practical to program. DGG ( talk ) 21:14, 21 August 2014 (UTC)

Delaying A7 at NPP would help, and this is a policy change that the community can adopt by itself. (It would need to be implemented by programming, but if there is a clear policy requirement for something, a programming solution has always been found). It would be more effective to delay A7 altogether, at NP as well as NPP, and we could certainly try proposing it. I think there's a reasonable chance of adoption. Again the community could adopt this by itself. I suspect it will be more difficult to program, but it should be possible. And at least it can be enacted as forbidden, which will give us a good response to those who do it.

Delaying NPP altogether has the danger of delaying action on more critical problems than A7. This is a real difficulty, because the immediate removal of dangerous material is a potent deterrent. In the past, this has prevented any such solution to the problem. It might be more acceptable now, because we could do it on NPP while retaining it for NP. And we have much more effective edit filters that in the past--the extensive amount of utter junk present 7 years ago when I started doing this is much diminished.

More generally, when encouraging new editors to become active in WP process, we need to arrange some means of either teaching them or checking that they have educated themselves. The safest deletion process for beginners to work on is AfD discussions, where there will be other people looking in a very visible way, and usually bad arguments are corrected. WP has always relied on the correction of errors by cooperative editing of many people, and the problems with NPP (and AfC) is that they rely instead on single individuals. The model simply does not work in such cases. Individuals can only be relied on if they are screened, tested, and audited. The usual argument against this, is who will have the time to do it, but looked at from a broader perspective, it saves time in the end because there will be fewer errors that need extensive work & discussion, and more new editors will stay with us. (One of the clearest of all research results, is that people whose first article is deleted very rarely continue). DGG ( talk ) 21:14, 21 August 2014 (UTC)

I'm neither a programmer, commenting in my official capacity nor assigned to the team that handles page curation support; I do research :). User:Legoktm is points 1 and 3, however. Ironholds (talk) 22:19, 21 August 2014 (UTC)
DGG & Kudpung, I agree that any change to the software that would affect process should correspond to consensus of those involved in the process (e.g. NPP). However, what I would like to propose is that we run a limited experiment for the purposes of observing the effects of a change. This experiment could/should be designed to only affect a small percentage of new page creations and it would only need to run for a short period of time (a week or so depending on the proportion of page creations) in order to observe a substantial effect (if there is any). I've already secured a few volunteers would be willing to manually work to ensure that new page creations that are more troublesome (e.g. spam and personal attacks) would not survive the 30 minute delay -- not enough yet, but it shows that there's interest in supporting an experiment. I think that the results of such an experiment could be critical for identifying some of the implications of a policy change before we consider making it.
This all assumes that we can get the engineering support necessary to run the experiment. That's yet to be determined. --EpochFail (talkcontribs) 10:34, 22 August 2014 (UTC)
To be quite honest I don't think that running more experiments is going to make us any wiser. Over the last 5 years every single aspect of NPP has been brought up and ruminated upon until it seemed as if the idea makers had run out of steam. What we got out of it all by demonstrating to the the top people in the Foundation through live demonstrations of what actually happens at NPP and without all the mind-numbing stats from p[eople who try to prove us wrong, was the new New Pages Feed and its Curation Toolbar. And that happened surprisingly quickly - it was like going to bed and waking up in the morning and there it was, as if Santa Clause had brought it down the chimney in the night. And it was beautiful. In spite of all the other gimmicks that have been thrust on the community such as VE and MV and weird feedback tools etc, as I've said before, the new NPP suite was the best thing to hit en.Wiki since sliced bread, because it's such an important process, BUT, and it's a VERY BIG BUT, it is only as good as the people who use it.
Tweaking the software to accommodate users who shouldn't be patrolling in the first place is not the way we should be approaching the shortcomings of the NPP system. We should ideally be looking instead at ways to attract a better class of editor to the task and ensuring that s/he is fit for the job.
A few years ago Scott and I practically rewrote the NPP instruction manual and still hardly anyone goes near it. OK, it's long and complex, but not so convoluted that a user of normal intelligence and motivation would feel intimidated by it. The problem starts when newbie editors realise that Wikipedia is not only the encyclopedia anyone can edit, but is a so a free web site anyone can tinker with and go on a button mashing spree. That's what happens when they discover that 'New Pages Feed' link in the tools section of the sidebar. They go and experiment. They don't go via WP:NPP or WP:DELETION, so we are reaping what we have sown, and IMO it's not the way we should be recruiting maintenance workers. Let's face it, designing and building better motor cars and roads may fractionally reduce the number of fatal accidents (and it doesn't here in Thailand). but it definitely does not make people better drivers. Kudpung กุดผึ้ง (talk) 11:35, 22 August 2014 (UTC)

Toolbar for older pages[edit]

Can this toolbar (PC features) be made to appear on pages which are a bit older? It seems to me that it works only for very new pages. I often review articles which were create days before, like those at User:AlexNewArtBot/PolandSearchResult. --Piotr Konieczny aka Prokonsul Piotrus| reply here 09:30, 22 August 2014 (UTC)

Same here. Filtering by user doesn't work, and I want to unreview a page that was created in June. QVVERTYVS (hm?) 19:20, 25 August 2014 (UTC)

How do I get rid of the page curation toolbar?[edit]

About two hours ago, when looking for Special:NewPages, I accidentally visited Special:NewPagesFeed, and now, when I visit certain pages like Dina Rae (singer), the Page Curation bar appears, which I don't want, because it hides content and you have to scroll the page up and down to see what's behind it. Wikipedia:Page Curation#Curation Toolbar says that I "can close that toolbar by clicking on the 'x' icon—or minimize it by clicking on the icon to the right" - but there is no an 'x' icon, and of the seven icons, which are all in a vertical column, which is the "icon to the right"? --Redrose64 (talk) 19:29, 6 September 2014 (UTC)