Wikipedia:Village pump (idea lab)

From Wikipedia, the free encyclopedia
  (Redirected from Wikipedia:VPIL)
Jump to: navigation, search
  Policy   Technical   Proposals   Idea lab   Miscellaneous  
The idea lab section of the village pump is a place where new ideas or suggestions on general Wikipedia issues can be incubated, for later submission for consensus discussion at Village pump (proposals). Try to be creative and positive when commenting on ideas.
Before creating a new section, please note:

Before commenting, note:

  • This page is not for consensus polling. Stalwart "Oppose" and "Support" comments generally have no place here. Instead, discuss ideas and suggest variations on them.
  • Wondering whether someone already had this idea? Search the archives below, and look through Wikipedia:Perennial proposals.
« Older discussions, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, 20, 21, 22

Centralized discussion
Proposals: policy other Discussions Ideas

For a listing of ongoing discussions, see the dashboard.

Limited Duty Admins[edit]


As a recently-returning editor, I've noticed a shift in RfA expectations. See, for example, this discussion in K6ka's RfA. At least part of the community will want expectation for previous experience in many different areas, even ones the admin candidate has no expressed interest in, in almost all serious RfA's. There are some fairly strong opinions that the entire process of creating admins is broken expressed even at the highest level. At the same time, there are frequent backlogs in areas that require admininstrator involvement. Clearly we can use more editors invested in these areas.

There have been prior attempts to reform RfA of varying degrees of success, apparently. Instead of reforming the process of creating admins, why not reform adminship, or at least part of it? We have done this before, creating Rollbacker and New Page Patroller rights to improve those areas.

The model of authority I am most familiar with, that of the U.S. Navy, has a feature called a Limited duty officer. I wonder if that is something we might make usage of. The Limited Duty Admin would be a user that has admin rights but only uses them in a limited number of areas. This would not necessarily need a programming change, the way that creating a separate rollback right did.

Imagine that user:RichardRoe has good experience with and wanted to help out in article deletion. They request adminship in only CSD, Prod, and AfD. They show they have experience and good judgment in those areas, and a new account user:RichardRoe_DeletionAdmin gets created and the bit added. user:RichardRoe remains a regular account and any admin tasks are processed through user:RichardRoe_DeletionAdmin. Any attempts by user:RichardRoe_DeletionAdmin to act outside of deletion would be obvious both to users and to other admins, especially "full" admins. In the case that Richard saw something that required emergency admin action outside their area, a simple post on WP:AN would clarify their actions quickly.

This is motivated by a semi-outsider perspective, so I no doubt have missed things or am ignorant of similar attempts. Please fee free to chastise me here;).

Thanks for reading this far. — Preceding unsigned comment added by Eggishorn (talkcontribs) 20:43, 2 January 2017 (UTC)

D'oh! Thanks @Mandruss:. Eggishorn (talk) (contrib) 20:57, 2 January 2017 (UTC)
See Wikipedia:Perennial proposals#Hierarchical structures for some background on unbundling administrative privileges. (Note this doesn't mean that further unbundling can't take place, but understanding what has been previously discussed is helpful.) Wikipedia talk:Requests for adminship#Planning for a post-admin era is the most recent thread on the RfA discussion page with some discussion on unbundling, and Wikipedia talk:Requests for adminship/Archive 244#Moderator proposal links to what I believe is the most recent proposal made on that discussion page. isaacl (talk) 22:35, 2 January 2017 (UTC)
I like unbundling and I would take on some limited duty myself, while being completely unprepared for the whole mop. I have not read all prior discussion, but the opposition arguments I've seen have been unconvincing. When I have advocated a try-it-and-see approach as opposed to a WP:CRYSTAL approach, the responses have either been absent or of the nature, "Yes, but [repeat of WP:CRYSTAL]." En-wiki is risk-averse in the extreme. ―Mandruss  04:38, 4 January 2017 (UTC)
People have put the admin bit on too high a pedestal. The old "it's no big deal" concept needs to make a comeback and quick. Trustworthiness is the only factor that matters for RfA and that should be the central focus. A large portion of editors have become unreasonable in how they factor content creation towards trustworthiness and I think it's currently the single worst element of the RfA process.
As for "limited duty admins", I barely do any admin stuff. I don't have the time or desire to do a lot of it; something I truthfully stated in my RfA and which caused many people to oppose me. The admin work I have done I think has been very beneficial to the project yet despite my history establishing that I am a respectable editor here for the common good, many people were willing to forego giving me a chance based on their own negative biases that fear the worst. Bottom line: we need to get better at identifying trust. The community lets too many people through we shouldn't make it (look at how often admins end up at Arbcom) and are presumably denying people who should make it. I think the process is inherently biased against those with cautious/meeker/transparent personalities and biased towards those with with bolder or Machiavellian personalities. An example of "machiavellian" behavior I would give is those who answer RfA questions, not from the heart, but by seemingly writing in the way that they suspect the community wants to hear.
There's no onus to have to use the tools. It's not like becoming an admin means you must do admin work, which is how many editors seem to treat RfA. This is especially true since it's unreasonable to expect the editors to be fully aware about what they can (or want to) do until after they get the tools. In fact I think the current RfA process encourages people to jump head-first into the fray when I think a "dip your toes" approach is much wiser. "Don't walk before you can run" should be the parting advice after a successful RfA.
Anyway, I don't really think it's the RfA process that's broken so much as the mindset of the participants. (Mandruss, I've seen you around and have a favorable impression so if you know your editing history to be clean, you should have them too.) Jason Quinn (talk) 09:52, 4 January 2017 (UTC)
After hearing many complaints about WP:AE I have started paying attention to ideas for reform that links to a user space scratch pad file and anyone's additions there are welcome. We might get more admin participation while at the same time cultivate community trust, by creating a class of admin that have AE rights, limiting the number of them, appointing them for short-ish terms, and emphasizing that re-appointment depends on particular factors the community really wants to see exhibited by those admins. On the list of factors, I propose the #1 thing should be community assessment as to how well the candidates arrive at their opinions through the application of the principles and remedies articulated in the arb ruling that authorizes DS in the first place. NewsAndEventsGuy (talk) 17:21, 4 January 2017 (UTC)
PS Having read comments here and at the "perennial proposals" page, it's seems pretty likely that a feeler "what if" (like this good faith thread) will produce a lot of no's, just because history repeats. However, if a proposal is packaged in outline and/or graphic or tabular form.... has a short concise bullet list of goals to be achieved... and inventories the various admin tools to be parsed between admin the different admin levels .... then the proposal will hopefully get some more specific opinions ... they might still be "no", but it will require more thought to back that answer up. NewsAndEventsGuy (talk) 19:04, 4 January 2017 (UTC)
I agree that adminship is WAY too precious and that RFA needs reform, but I don't know that unbundling is the way to go. "I don't intend to use this tool" is not the same as "I cannot be trusted with this tool", and we should assign adminship based on trustworthiness primarily (i.e. do we believe a person will misuse a tool). It's not an award we give to people for doing certain things right at Wikipedia, it's a set of tools we give to people who we believe won't abuse them. We NEED to get back to that mentality, and dial back the strictness of RFA, but I don't believe unbundling helps that at all; it merely increases the workload and complexity of adminship (because now we need to consider 3-4 different RFA processes rather than merely one) without solving any of the problems of RFA (unrealistic and unhelpful expectations of what qualities a good admin needs). If an admin has no use for the protection tool, they can simply not use it. I see nothing useful of specifically not giving it to them. It does no harm. --Jayron32 18:58, 4 January 2017 (UTC)
Not that I want to constrain discussion, but I was not thinking so much about further tool unbundling. Rather, I was thinking along the lines of what might be called responsibility or area unbundling. The Limited Duty Admin would have access to all the same tools as a "regular" admin, preventing any programming changes. The only change would be in community-granted expectations. This would address the "You don't have enough content creation/AIV presence/deletion experience/whatever" objections used in RfA as roadblocks to otherwise-qualified candidates. It would also address the "An admin needs to be trusted everywhere" objections. If the community doesn't trust the candidate on vandal-fighting and the candidate doesn't want to vandal fight, why make vandal fighting an issue? Thanks. Eggishorn (talk) (contrib) 19:51, 4 January 2017 (UTC)
So a candidate states that they will never be involved in Admin work in area X but within days starts working in area X. What happens next? Leaky Caldron 20:03, 4 January 2017 (UTC)
Another user reports it on AN and, if no explanation is forthcoming, the new admin is blocked and life goes on. No big ArbCom mess needed. Eggishorn (talk) (contrib) 20:18, 4 January 2017 (UTC)
What happens next? Next, the community takes a close look at how they could have been completely wrong on trustworthiness. As long as we demand perfection, no improvement will be possible; that's the one thing we can be sure of. ―Mandruss  20:23, 4 January 2017 (UTC)

User-performed deletion of user subpages[edit]

It seems like a bit of a waste of time for admins to have to go through and delete pages in people's user space instead of letting them do it themselves. I can't see any real potential for abuse here, as {{db-u1}} requests are usually granted right away anyways. goose121 (talk) 05:01, 8 January 2017 (UTC)

Sample scenario for "why not" (1)Go to any article (2) move it to your user subpage (3) delete it ... — xaosflux Talk 05:07, 8 January 2017 (UTC)
These do not take much admin time. It just needs a check of the history. Some other speedy deletes are much more intensive in work, eg checking for copyright infringements, and seeing if there is a good revision in the history; or checking for a banned user, when they are not blocked or banned - just a suspected sock. Graeme Bartlett (talk) 23:28, 8 January 2017 (UTC)
@Xaosflux: that doesn't work: speedy deletion criteria apply only if all revisions of a page meet those criteria. In the case you describe, the pre-move revisions obviously fail the criteria. – Finnusertop (talkcontribs) 23:32, 8 January 2017 (UTC)
I was referring to the existing software capabilities, which do not care about prior revision owners when performing a deletion. — xaosflux Talk 00:47, 9 January 2017 (UTC)
It would probably be difficult to make the software check for incoming page moves; and this trick has even tricked human admins. עוד מישהו Od Mishehu 18:53, 10 January 2017 (UTC)
Then have software just check to see that the page started out in the person's userspace / has ever left the person's userspace. Dustin (talk) 20:35, 10 January 2017 (UTC)
WP:PEREN#Grant non-admins admin functions within their user space. BethNaught (talk) 23:55, 8 January 2017 (UTC)
At that section: "if users could delete pages in their namespace, they would be able to move pages to their user space and delete them." – At least in theory, having a move log check / edits by user check doesn't seem like it would be too complex, and it would ensure that that problem never occurs. Perhaps someone else could assess the actual difficulty of such an implementation? "Gives the impression of user space ownership" – If users don't de facto control their own userspaces, why are they allowed to request deletion of pages? I get that users aren't allowed to outright violate general Wikipedia policy in their userspace (spam, personal attacks, harassment, general vandalism, etc.), but if a user is allowed to request deletion of userspace pages, I don't see why barriers should remain in place that slow down this process if faster alternatives that still prevent abuse are possible. "and has been rejected by the developers" – They seemed to be partly following the "own" bit (which I mentioned earlier), which seems to be flawed reasoning in my opinion (it could at least be expanded upon). Really doesn't strike me as compelling rationale. But alas, I am not the person writing the software. Are there significant software barriers, perhaps? Dustin (talk) 20:35, 10 January 2017 (UTC)
Perhaps it would be possible. But so far the administrators who actually do these deletions have not complained about the workload, so it's highly unlikely for anyone to bother actually doing the work to make anything like this happen. Time to drop it. Anomie 22:36, 10 January 2017 (UTC)
Its a good idea and we have had consensus more than once in the past to do it. But not the IT resource to make it happen. As I remember the ballpark figures, providing we put in the logical safeguard of not allowing this for a page that had been moved, were that it would save us admin time equivalent to appoint an extra admin a year. I'm hoping that eventually we can get this actioned, as much because it would empower everyone to do something they should be able to do as because it would lighten the admin load. ϢereSpielChequers 18:00, 11 January 2017 (UTC)

Navbox request template[edit]

It would be great to have a template that could be placed on the talk page of a category or list, in order to request that the Wikipedia pages listed be turned into a navbox. Submissions would go into a category (similar to Template:Infobox requested and Template:Logo requested).

Daylen (talk) 19:21, 8 January 2017 (UTC)

Probably never to be serviced. My suggestion, is that you make such requests at the applicable WikiProject. --Izno (talk) 21:18, 8 January 2017 (UTC)

Some way to sync up glossaries with articles?[edit]

I'm not exactly sure where the proper place to propose this is, but I had an idea that would make technical articles much more readable. What if there was some feature so that when you click on or mouse over a technical term (perhaps one not warranting its own article), a concise definition of the term would appear (something like WP:Hovercards but for glossary definitions). Ashorocetus (talk | contribs) 17:03, 9 January 2017 (UTC)

New Editors - or are they?[edit]

I do a few bits and pieces of reviewing the work of new editors, typically looking at their first 4-5 edits. Generally speaking, they fall into three categories:

  • Some are obviously constructive. Their first few edits will be making useful corrections to articles or expanding an article which is (I guess) of interest to them. A few of these I template with something appropriately welcoming, but generally I just let them get on with it.
  • Some are obviously unconstructive. You know the sort of thing; either it's spamming the same URL into half a dozen articles, or user XYZCorp is editing a page coincidentally called XYZCorp, or it's about lunch-time in Wales and half a dozen new editors all turn up making "funny" edits to a Welsh school's article, or someone, hilariously, replaces a whole article with the word "gay" repeated about six hundred times. These get reverted, warned, and if it's particularly unconstructive twinkled to AIV.
  • Some have edits that appear constructive, but... there is no way on earth that that is a new editor. These accounts begin their careers by making elaborate changes to infoboxes, or go around adding the same category to dozens of articles, or launch straight into detailed policy discussions on talk pages. They frequently have very elaborate user pages within their first couple of dozen edits and are sometimes prolific editors - often averaging 50-100 edits per day for weeks or months. They often make multiple page moves in their first few days and sometimes show a detailed knowledge of the MOS.

What do people do about that last category? They do, reasonably often, end their days indeffed as sockpuppets, but without any particular evidence to link them to any master account, it's not clear to me what can be done. There seem to be certain people around who have a, well, encyclopaedic knowledge of the styles of the usual suspects, but I'm not one of them.

For all the obvious reasons I'm not going to link to any examples of active users I feel fall into the third category, but User:Johny5000, User:Team61ROLL and User:White Flower are the sort of thing I'm talking about. Going to [1] and checking out the recent editors with thousands of edits will also demonstrate the kind of pattern I'm talking about.

Is there anything to be done? GoldenRing (talk) 13:48, 11 January 2017 (UTC)

beats me. Still I would like to applaud your effort hereNewsAndEventsGuy (talk) 14:00, 11 January 2017 (UTC)
Thing is, we get so used to dealing with the users who create an account and immediately plunge into editing - getting stuff wrong and generally attracting a pile-on of helpful advice and/or warnings - that we tend to forget how much of Wikipedia you can look at without actually editing. I for one spent a good long time looking through project pages and guidelines before I ever saved my first edit, and I suspect that a lot of other users do as well. There's no proscription against lurking for a few months and learning how things work before getting stuck in, and I expect a lot of people do exactly that - we don't want to introduce anything that will discourage them because frankly, they are exactly the sort of users we want to attract. Sockpuppets and illegitimately-returning users are generally picked up on other issues than their immediate competence at editing, and what you're suggesting would also have repercussions for those users attempting a clean start, as well. The system you're asking for already exists - it's called WP:AGF! Yunshui  14:12, 11 January 2017 (UTC)
(ec) @Yunshui: It's a fair point. There are a couple of subcategories of the third type, most of which are perfectly acceptable: Some are coming from other non-en wikis and know their way around the software pretty well (saw one of these today). Some I guess have been editing as IPs and know their way around. Some start by doing one of the tutorials and hit the ground running that way. And others, as you say, have just been lurking for a long time. But others tend to get several hundreds or thousands of edits in (those three accounts I linked managed well over 1,000 between them) before someone tumbles to the fact their a sock of a banned user, and someone ought to be going back and reviewing those thousands of edits to see what needs reverting. I suspect no-one is, because it's a very big job. In a way it makes me wish CU was just a form where you could pop a name in and it would generate a report - but of course that's never going to happen and for good reasons. Or maybe just that CUs would accept, "That's a new editor? BS!" as a reason to run CU. Also, it seems, not going to happen. GoldenRing (talk) 14:44, 11 January 2017 (UTC)
There are also the promotional undisclosed paid editors who drop in a complete article on their first edit. For those, even if the article is not an obvious {{db-advert}} you may want to consider pointing them to WP:PAID or placing a {{uw-paid1}} inquiry on their talk page. Sure there are editors who 'spent time learning Wikipedia before making their first edit' and nearly every one of these accounts will say that is what they did if challenged but 98%+ of the time that is BS so the actual WP:BITEing risk is very low. JbhTalk 14:42, 11 January 2017 (UTC)

This whole exercise is also leaving me tempted to start WP:Taxonomy of new editors - only slightly tongue in cheek. GoldenRing (talk) 14:49, 11 January 2017 (UTC)

That would actually be very useful. Particularly to help train new page patrollers/reviewers. JbhTalk 14:53, 11 January 2017 (UTC)
@Jbhunley: A start can be found at User:GoldenRing/Taxonomy_of_new_users GoldenRing (talk) 17:24, 12 January 2017 (UTC)
  • Maybe there should be a WP:Database report for "New user articles" or "Users less than X days with greater than Y edits". This might also help sniff for WP:ECP gamers. --Izno (talk) 15:06, 11 January 2017 (UTC)
    • I guess this is roughly how I use the edit count part of [2] GoldenRing (talk) 17:14, 11 January 2017 (UTC)

One thing to remember: Some of these cases may be genuine clean starts. We must respect this right, and not start assuming that these users are blocked or restricted users trying to get around their block or restriction. Only when we have clear evidece of this my we start to make such assumptions. עוד מישהו Od Mishehu 21:55, 11 January 2017 (UTC)

  • This reeks pretty strongly of assuming bad faith right off the bat. As someone that was the target of a witch hunt that ended nowhere you really need to understand that what Yunshui describes probably happens a lot more often than you think. Remember, the Wikipedia and Help namespaces are completely indexed and anyone can easily Google anything they are looking to read up on and find information on it. My first edit was to a table, after reading extensively on Help:Table and Help:Table/Introduction to tables. I also used the preview button before actually saving, which made my first edit pretty much perfect from a syntax standpoint. That edit was, of course, brought up in the witch hunt as "proof" that I was not a new user but I had never edited Wikipedia before that. If there are problems with their edits, of course gather evidence, request a CU, file an ANI report, tell them what they are doing wrong, do something. But if there is nothing wrong don't start a witch hunt and potentially chase off an editor that is working to improve. Especially if there is no evidence of any wrong doing. That is assuming bad faith and should be avoided. --Majora (talk) 03:37, 12 January 2017 (UTC)
    • @Od Mishehu:@Majora: - Please understand that my current strategy with these users is exactly to AGF and let them get on with it, because I have no way, early on, of distinguishing those that are going to be disruptive from those who are genuine. I understand that some come from other wikis, as User:xaosflux notes, though these are often good enough to note this on their userpage. I also get that some people are careful and painstaking about making their first edit perfect, though it would be unusual for these types to amass dozens of edits in their first 24 hours for the obvious reasons. Regarding clean starts, that right is not absolute - if the user is banned or blocked, they are not entitled to a clean start. I guess I posted this here to see if anyone had tips on sorting the disruptive ones from the genuine ones. GoldenRing (talk) 14:15, 12 January 2017 (UTC)
  • One thing to look for is: Do they have edits on other WMF projects? (Check Special:CentralAuth). I know I can personally be found showing up on other projects and could appear exactly as "class 3" if this wasn't checked by that local community. (e.g. w:ba:Махсус:Өлөштәр/Xaosflux) Not all cross project users will have a global user page either. — xaosflux Talk 04:39, 12 January 2017 (UTC)
  • The way to spot a cleanstart by someone who is not entitled to a cleanstart is to keep an eye on your watchlist and especially pages where certain blocked or banned editors were problematic. If you see someone new turn up and continue the disruptive behaviour that someone else just got blocked for then file an WP:SPI. Otherwise best to AGF. ϢereSpielChequers 16:26, 12 January 2017 (UTC)

Why don't we use PCPL2 or PCPL2 /w Semi anymore? I see how they could be very useful and I am not sure why they aren't used.[edit]

Like the title says, I am confused about why we don't use Pending Changes Protection Level 2 or Pending Changes Protection Level 2 with Semi Protection. I see many cases in witch these could be used, and In many cases they could even stop edit warring. Can someone please answer? — Preceding unsigned comment added by Creeperparty568 (talkcontribs) 22:07, 20 January 2017 (UTC)

There is no community consensus for it to be used. See Wikipedia:Village pump (proposals) § Make PC2 no longer available to admins. — JJMC89(T·C) 22:31, 20 January 2017 (UTC)