Wikipedia:Village pump (idea lab)

From Wikipedia, the free encyclopedia
Jump to: navigation, search
  Policy   Technical   Proposals   Idea lab   Miscellaneous  
Shortcuts:
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, 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18


Centralized discussion
Proposals: policy other Discussions Ideas
Note: inactive discussions, closed or not, should be archived.


Grant some people from some research groups permission to insert original research into Wikipedia articles[edit]

Original research might be necessary to make some explanations in Wikipedia articles clearer. I'm sure some of those people will have the skill to only insert pieces of original research that are true, which was probably the original reason for the rule no original research. They would probably be verifiable by the ability of other experts to figure them out. Maybe there could be a way for people to demonstrate in Wikipedia that they have the skill not to insert any wrong or unverifiable original research. Blackbombchu (talk) 00:53, 25 October 2015 (UTC)

Absolutely not. Esquivalience t 04:01, 25 October 2015 (UTC)
Hell no. This way lies elitism. --Jakob (talk) aka Jakec 15:23, 25 October 2015 (UTC)
I agree that original research would be a very bad thing here, but there is a point to be made. Subject matter experts are often made to grub around finding sources for well known facts found in numerous student-level textbooks. This is unnecessarily hard work and tends to drive such people away. I have long held that we need to cure ourselves of the "little blue number disease" (the one that requires a blue number at the end of each sentence) and take a more robust approach. No other establishment in the known universe has such stupidly onerous requirements on sourcing. At the end of the day, the blue numbers guarantee nothing, one has to actually read the sources for a guarantee. SpinningSpark 18:45, 25 October 2015 (UTC)
@Spinningspark:I see you have never worked on a law review, otherwise you would never say that, "No other establishment in the known universe has such stupidly onerous requirements on sourcing." ~ ONUnicorn(Talk|Contribs)problem solving 18:28, 2 November 2015 (UTC)
Or any high quality scientific paper I would like to add. In scientific papers you can bring in new facts for 3 reasons. (1) By rigorous sourcing not unlike Wikipedia (2) By original data (3) By logical argument (synthesis). Options 2 is only relevant for primary sourced ideas, option 3 is only relevant to primary and secondary sources - neither are for tertiary sources like Wikipedia. Arnoutf (talk) 18:47, 2 November 2015 (UTC)
This is a dreadful idea if you want to keep Wikipedia as a neutral encyclopedia. For such a system to work all academics would have to agree about everything all the time. Endgame for this sort of situation would be low ranking academics with lots of free time pushing their POVs in article with no restraint. Brustopher (talk) 21:59, 25 October 2015 (UTC)
Unthinkable. Idea would turn the encyclopedia into a battle of elites. Suggest we close. Jusdafax 19:39, 11 November 2015 (UTC)
hm, sorry, but we simply cannot there is simply no way in practical terms to do so here. --Sm8900 (talk) 20:47, 11 November 2015 (UTC)

Getting updates on members of Categories[edit]

Tracked in Phabricator
Task T9148

Maybe it would be useful if a user could be notified when a new member is added to a specific category.SoSivr (talk) 23:06, 26 October 2015 (UTC)

See phab:T9148. Anomie⚔ 11:51, 27 October 2015 (UTC)
If you're only interested in one or two categories, then you could manually check the (ugly) URL in Help:Category#Retrieving category information to see what's been added recently. WhatamIdoing (talk) 23:38, 2 November 2015 (UTC)

Would suppressing edit count reporting beyond some level help reduce editcountits[edit]

The background for one specific incident can be seen here:

The discussions are still unfolding even as I write this but, in brief, an editor created approximately 80,000 redirects, most of which are viewed to be as inappropriate. Dealing with this issue has already occupied dozens of hours of editor attention, and is likely to involve many more hours of cleanup and discussion about how to handle this specific event.

It is my view that one of the causes of this problem may be characterized as metastasized editcountis.

If this were the only such case, I'd simply be happy allowing the community processes to carry on and decide how to handle the specific individual. However, I think this may be the symptom of a general problem as opposed to a one-off situation.

On occasion I've taken a look at Wikipedia:List_of_Wikipedians_by_number_of_edits to see who our most prolific editors are.

In many cases, the editors high on that list are some of the most respected content creators in Wikipedia. Many of these editors have received kudos, well deserved, for the substantial contribution to this project.

However, I have sometimes wondered how editors managed to amass such large numbers of edits. My casual investigation leads me to some disquieting results. It isn't always the case that the edits fall into what we think of as a canonical edit — find some article that needs improvement, do some research, add or modify some text, add a reference, rinse and repeat.

I'm reminded of the adage "to err is human, to really screw up requires a computer". In some cases the accounts are the result of automated or semi automated editing. Here it is important to be especially careful. There are a lot of legitimate reasons for doing automated or semiautomated edits. In many cases, each of these edits improves the encyclopedia in a meaningful way. However, there are other such edits whose benefit seems more in generating edit counts than in actually improving the encyclopedia. I understand we have rules to prohibit automated edits that are truly minor, but I think we've all seen examples of edits whose contribution is quite limited.

I don't want to focus solely on automated edits, especially as the current situation appears not to have involved automated editing. However, it seems clear that this editor identified some article, then dreamed up 20 to 50 alternative phrases that might have something to do with the article and created them as redirects. There's a bit of consternation about the nature of the edits focusing on their appropriateness. That's a valid concern, but my focus here is not so much on whether the choice of wording was inappropriate, but the possibility that our emphasis on edit counts encouraged someone to mindlessly create useless redirects.

As another example, I spend a fair amount of time at CSD deleting unused categories. In many cases, it appears that the category wasn't really created in good faith, but was a mindless creation intended to bolster edit counts. Do we really need a category to keep track of corporations that were dissolved in Syria in the year 1132?

One solution is simple — let's discourage the counting of edits beyond some level. I think it is useful at times to know whether an editor has a few hundred edits or a few thousand or tens of thousands. If you need to discuss something with them on a talk page, it's helpful to know whether you are dealing with a newbie or an experienced editor. For that reason, I'm not proposing the absurd notion that we should suppress the reporting of edit counts. However, I think that beyond some point, the count provides no useful information about the type of editor, and merely becomes in some cases, an ego measure. I'll reiterate that this is not a blanket view of all of the editors at the top of the list. In fact, I hope it applies to only a minority. It is clear that many brand-new editors are obsessed with edit counts, and we often counsel them not to be quite so concerned. In many cases, after a few thousand edits, they lose their obsession, and I am confident that many people near the top of the list don't really care whether they have a hundred thousand or 300,000 edits.

My suggestion is simple — why not suppress the public listing of edit counts beyond some level? If we did so, then if an editor reached that level, they should continue to edit for the improvement of the encyclopedia but would no longer be encouraged to find creative ways to generate high edit counts. They'll make lots of redirects if the redirects are valid, they will make lots of categories if the categories are valid, they'll run AWB if it improves the encyclopedia, but they won't dream up ways to pad the edit account.

I'm sure they'll be lots of opposition and lots of questions. One obvious question is where to set the level. My initial thoughts were something like 50,000 or 100,000. I notice that our service awards go up to 132,000 edits, So that might be a natural choice for an upper limit, although I would prefer something a little bit lower.

If we stopped keeping track of edits beyond some large limit, do we think that editors with more edits would stop editing because they wouldn't get recognition? My hope is that this isn't the case.

It would obviously be some technical details, as edit counts are available and pop-ups and calculated with various edit counters, but I'm certain all those technical details could be worked out if the community thinks that suppressing edit counts beyond some level might help discourage editcountitis.--S Philbrick(Talk) 17:14, 6 November 2015 (UTC)

Support. Glad you're bringing this up! I've thought much the same thing, that 50 or 100K edits is something of a threshold we could just describe as "way too many edits." Why display this information, if it encourages individuals to spend an unreasonable amount of their life on a website, or make poor quality edits?
Let's develop some possibilities related to prolific contributors and editcountitis further. Here's a start:
One alternative to editcountitis that might be worth considering is developing tools that tally up manual edits which actually fix a problem in a maintenance backlog, or that provide substantive expansion of content. It's important to come up with incentives that reward editors for making substantive contributions, rather than incentives that encourage participants to waste time pointlessly churning away. --Djembayz (talk) 17:20, 7 November 2015 (UTC)
Opppse – I'm never partial to "solutions" which include hiding information from people. A better solution would be to "redefine" what is considered an appropriate "edit count" – e.g. focusing on just main space edits; or perhaps focusing instead on non-redirect, non-disambig. article creation, or something. --IJBall (contribs • talk) 17:35, 7 November 2015 (UTC)
Oppose. Unfortunately, suppressing edit count is going to make more problems than it solves. People are going to argue about what is an "appropriate" number. You can probably blank the entries in WP:NOE for editors with over 100K edits (just say 100,000+ for these entries) or remove the "Edit count" tool. However, actually hiding the edit count is something beyond the English Wikipedia's control, because it is can be seen in several places, such as in popups and Special:CentralAuth. In principle, this is a good idea, but in reality, suppressing edit count would be very hard. epic genius (talk) 17:42, 7 November 2015 (UTC)

Can I remind participants that this is the idea lab, not Proposals or Policy. The concept behind this page is that editors discuss the idea, and think of ways to improve it, but do not Oppose or even Support. For example @IJBall:, but the rationale included a better solution, specifically, redefining how we count edits. There is precedent for that - when I delete an article, it doesn't count as an edit. What if we decided that creation of redirect or dab pages, while useful, didn't qualify as an edit for the purposes of measuring edit count. That doesn't mean we don't measure them, deletions are counted, but they aren't counted as edits.--S Philbrick(Talk) 17:57, 7 November 2015 (UTC)

OK, but I'm still of the opinion that "redefining" what a "good edit count" is is far preferable to "information suppression". So, taking "redirect creation" out of "edit counts" might be a good start. I also think we do need to be leery of automated edits, so that's something else to think about... --IJBall (contribs • talk) 18:03, 7 November 2015 (UTC)
This is exactly the type of discussion I was hoping to have. I started with a simple-minded way of deemphasizing edit counts, and you suggest that the problem should be redefined to think about what constitutes a "good edit count". While that sounds like a challenge, it might have the dual advantage of providing more useful information while at the same time discouraging editcountitis.--S Philbrick(Talk) 18:11, 7 November 2015 (UTC)

As another example, @Epicgenius: didn't disagree with the core of the idea, but expressed concern that it would be difficult to agree on the level. I agree. If we ended up concluding that the general idea made sense but we couldn't reach a consensus on the cutoff level, we wouldn't implement it. Similarly if we end up with general support for the concept but there are technical difficulties we won't implement it.--S Philbrick(Talk) 18:03, 7 November 2015 (UTC)

  • The biggest problem I see with edit countitis is at RFA where it is one of the overlyeasy metrics that some people focus on. This is a problem both for well qualified candidates who eschew any use of the tools, and because it distracts people from properly assessing RFA candidates by actually looking at their edits. If we had a list of editors by non minor edits then a lot of my edits, including all the hotcat and twinkle ones, would not be counted. So I would suggest that as well as Wikipedia:List of Wikipedians by number of edits we create Wikipedia:List of Wikipedians by number of non-minor edits, calculted the same way but excluding edits flagged as minor. ϢereSpielChequers 19:13, 7 November 2015 (UTC)
  • Do you really mean "hide the number itself", or "get rid of lists showing how that number compares to other editors"? If the goal is to stop silly edits for the purpose of making your name be higher in the list, then getting rid of the list ought to be sufficient. WhatamIdoing (talk) 18:20, 9 November 2015 (UTC)
By "hide the number itself" I mean that it shows in pop-ups when you hover over someone's name. I wouldn't be in favor of eliminating it, although I guess that's sort of what I said. Are you old enough to remember when McDonald's marquees listed the number of hamburgers they had served? Eventually they opted for a simple "billions served". So one option would be to show the actual number until it reaches some level and then just say for example "100,000+"
On the lists by edit count, I would again show the number up to some level, but after that list all editors greater than the limit, possibly in order of the date they achieved it. I wouldn't be averse to having this information available somewhere because it's conceivable it could be relevant in some context but we don't have to make quite so easy. I have no doubt that some people are looking at Wikipedia editors by edit count and thinking up ways to move themselves up the list. If they do so by adding great content to articles wonderful. If they dream up some category which they can add to 10,000 articles it isn't so clear that it's a gain for the encyclopedia. My concern is I've seen a troubling number of examples of editors in the latter category.--S Philbrick(Talk) 18:29, 9 November 2015 (UTC)

A useful approach would be to make the generally displayed edit count more meaningful by including only what we define as substantial edits. This could be accomplished most obviously by omitting various types of edits.

  • The exclusion rules would be determined by what we want the curated count to indicate. For example, an edit count that highlights significant content editing could be accomplished by excluding Talk page edits, redirects and any other administrative edits, edits to PAG and essay, user, noticeboard and project pages, and article space edits under a certain character count (IOW, including only article edits over a certain character count).
  • Anti-gaming checks could be created with rules that flag extraordinary counts, like editors performing hundreds of qualified edits per day, and those could be volunteer patrolled and verified (similar to patrolling of pending changes).
  • Rules would be tweaked until consistent and widely agreeable results are achieved.
  • The user's raw edit count (all edits) would also be readily available.

This should not be viewed as controversial, as it's really just a system admin area, like firewall maintenance: the rule set would be there for all to see, and all editors could continually comment and make suggestions. Changes would result in automatic recalculation across all users. Whether this could be technically implemented at reasonable cost, and if it would be a significant drain on server resources, seem to be the only limiting questions. (I've been thinking about a somewhat similar semi-automated approach to RfCs...) --Tsavage (talk) 05:23, 23 November 2015 (UTC)

Wikipedia as a component of artificial intelligence[edit]

I was thinking that "wouldn't it be weird if Wikipedia became the first artificial sentience?"

Then I remembered that Wikipedia is part of Watson. That is, it is included in Watson's data banks.

So, Wikipedia is part of what Watson knows, its awareness, and eventually, when Watson wakes up, it will be part if its sentience.

But, Wikipedia is also an evolving program/data/computer complex in its own right, including a core program stack (MediaWiki +) and a small army of bots, installed on a massive array of servers, the whole of which is growing exponentially. So it is possible, that Wikipedia itself could become sentient.

Far fetched? I'm not so sure. With the line blurring between data and programming, with ontological data becoming integral to AI engines, and with ontologies being increasingly automatically generated from natural language sources such as Wikipedia, knowledge itself may come to life, in a manner of speaking.

And then, as such intelligence expands into the cosmos, the universe itself wakes up.

It's amazing how much Wikipedia has on this subject, which may provide the kernel for its eventual self-awareness.

Some things to think about. The Transhumanist 19:19, 6 November 2015 (UTC)

Would it be weird? Yes. But if Wikipedia scripts evolve to the point of making it self-editing, then I suppose it is possible. More likely though such capabilities will be developed elsewhere then migrated into Wikipedia. Praemonitus (talk) 18:44, 9 November 2015 (UTC)
well Wikipedia is the dotty professor of the AI world. and regardless of anything else, it is still only a web page, not a semi-autonomous robot and not even a computer. so if the Singularity ever did happen, then Wikipedia would possibly be literally the least of our problems. :-/ --Sm8900 (talk) 20:51, 11 November 2015 (UTC)

I'd see something like Google as becoming sentient more so than Wikipedia. There is no mechanism in place to make Wikipedia become "aware." You'd need a form of machine learning for that to happen, which doesn't exist here. As a machine, Wikipedia is primitive. It's the editors that do the thinking, not the computers. But you mention Watson, which is why I mention Google. Much of the information that pops up in a Google search (ie. in the "infoboxes" on the side) comes from here. It's possible that an advanced AI, if such a thing ever comes to be, will learn from Wikipedia, but it won't be Wikipedia itself doing the thinking.  DiscantX 13:26, 15 November 2015 (UTC)

Wikipedia has the 3 necessary components: hardware, software, and data. Yes, they are primitive. But that may change, along with funding, collaboration, personnel, etc. The Transhumanist 07:41, 17 November 2015 (UTC)

Suppressing redlinks option?[edit]

Sometimes, I think people may want to see a Wikipedia article without the redlinks in it. (so "text including Something wierd here" would instead simply show up as "text including Something wierd here"). Would this make sense as a preferences item? (If not, is it possible by setting a js/css file?)Naraht (talk) 17:14, 11 November 2015 (UTC)

Redlinks serve a function. Especially on a new topic (increasingly rare on Wikipedia) a redlink serves to indicate the demand for an article providing in depth information. If such an article would be trivial and would never materialize, the option, of course, would be to remove the Wikilink. But since we can already do that, there is no need to change. Arnoutf (talk) 18:07, 11 November 2015 (UTC)
This is possible in CSS. I don't know how to do it, but I know that it's possible. WhatamIdoing (talk) 18:52, 11 November 2015 (UTC)
.mw-body a.new { color: #210 !important; } /* Very dark red links */
To removed them you need JavaScript like $($('a.new').replaceWith($('target').html())); /*slower*/.
Frankly, its a fucking stupid idea and show how out of touch WMF's mobile team is. I've had to implement a redlink search functionality into WP:Dab solver because of this idiot "make it complete" mentality. You can check out the Video Game reference library (Wikipedia:Meetup/NYC/Women in Architecture#To-Do and on user pages) to find articles that have references, but no article. — Dispenser 23:32, 11 November 2015 (UTC) Updated 20:02, 20 November 2015 (UTC)
See also Help:Link color. Mobile has started showing red links, for example in https://en.m.wikipedia.org/wiki/Hakeem_Muhammad_Amin_Soomro. PrimeHunter (talk) 00:54, 12 November 2015 (UTC)
@Naraht: My guess is you wanted an option that would be permanent, but if a reader wants this option on an individual article, clicking on the "printable version" in the left sidebar will render the article without red links.--S Philbrick(Talk) 15:02, 16 November 2015 (UTC)
@Sphilbrick: That is something that I hadn't thought of, but that will get rid of all links, red and blue, I was hoping for something that would show the blue, but not the red. Cheers!Naraht (talk) 15:07, 16 November 2015 (UTC)
Fair enough, but I occasionally run across someone who finds all the colored links distracting, so that is an option to remember if that comes up.--S Philbrick(Talk) 15:37, 16 November 2015 (UTC)
True. Definitely an option.Naraht (talk) 16:21, 16 November 2015 (UTC)

Logos[edit]

I suggest that you added in previous logos for example ERT OTE, Cosmote, Vodafone Greecre Wind Hellas etc.--Γιουγκοσλαβια (talk) 16:42, 12 November 2015 (UTC)

A new type of pending changes protection as an alternative to full protection?[edit]

Idea: to set up a type of pending changes that allow autoconfirmed editors to edit an article, but changes must be approved by an administrator. Similar to how articles with pending changes protection allow IP/non-autoconfirmed editors to edit articles, but changes must be approved by pending changes reviewers. This would allow constructive edits to disputed articles (such as typo fixes and other uncontroversial edits) without the need to respond to edit requests. This type of protection may be suitable for articles like Nanak Shah Fakir, Brianna Wu, Mass killings under Communist regimes, Douchebag, List of social networking websites, and other long-term fully-protected articles. What do you think? sst✈discuss 08:59, 13 November 2015 (UTC)

There is already a PC2, Template:Pending changes table, that does that but consensus is not to use it, Wikipedia:PC2012/RfC 1. CambridgeBayWeather, Uqaqtuq (talk), Sunasuttuq 18:46, 15 November 2015 (UTC)
PC2 allows pending changes reviewers to approve submissions. I am suggesting a level of pending changes where only administrators can approve changes. sst✈(discuss) 17:20, 20 November 2015 (UTC)
This had been suggested in the initial trial proposal, WP:FPPR, but this ended up not being implemented. Cenarium (talk) 23:00, 20 November 2015 (UTC)

Easier access to biography articles via subject surname?[edit]

A reader may well approach the encyclopedia wanting to find out about a person with surname "Xyzname", when they do not know the person's forename(s) or initial(s). They may have read or heard some mention of "Dr Xyzname", or "After Xyzname's breakthrough work in this field", or "the followers of Xyzname". If this happens to be a string of characters which is only ever used as a surname (say Higginbottom), they mght find a surname page (this one has 5 entries), or they might do a Wikipedia search (if they know how to do this, bypassing the link to the surname page) and see a listing of 11 people surnamed Higginbottom. But if the surname they are looking for is something like "Leeds" they've got a problem. The base name page has a hatnote pointing to Leeds (disambiguation), which has a link to Leeds (surname), but there's no knowing how complete this is (though I did what I could with it earlier today). If they do a search on the word "Leeds", the results will include people with the surname, mixed up with a load of other articles (just one of the first page of 20 hits is a person with the surname).

There are different views among the Disambiguation community about whether entries for "people with the surname Xyzname" belong on the "Xyzname (disambiguation)" page, and if so where: a change to WP:MOSDAB in May 2015 means that they are now to be added to the "See also" section (which, to my mind,then gets very cluttered) until a separate Xyzname (name) page is created. (There is separate provision for people like "Lincoln", "Shakespeare" and "Churchill", who are recognised as being commonnly referred to by surname alone: those aren't the people I'm worrying about here). But such listings, wherever they are, are always likely to be incomplete anyway - as with our Higginbottoms above.

For living people, it's possible to create a link to the appropriate A-Z section of Category:Living people (like this). It's slightly inelegant in that it continues on beyond the chosen surname, but it's otherwise pretty good: a listing by surname - ie using the "DEFAULTSORT" that many of us carefully add to every biographical article we ever see.

If there was a listing which was the equivalent of "Category:All people" (ie living, dead, or unknown), sorted by DEFAULTSORT, then we could add a link to the "Xyzname" point in this sorted list as a really useful enhancement to the "See also" section of every disambiguation page where the word being disambiguated is ever used as a surname/family name/"the name used as a sort key". It would also be useful on every surname page, to provide an up-to-date listing to complement the handcrafted annotated listing on the page itself.

There could perhaps be a template to add to the "See also" section of appropriate disambiguation pages, which would provide this link, with text saying "List of people with surname Xyzname", in the same way that {{look from}} and {{in title}} are often added. With real sophistication, maybe the template could produce a list cut off at an appropriate endpoint (the next possible word, perhaps, eg "Xyznamf" for "Xyzname" - that way we'd get all the compound names included too).

But the prerequisite is for there to exist a category, or category-like listing (not necessarily updated in real time, perhaps every day/week if it would otherwise be too demanding of the system) which includes every biographical article in the encyclopedia, sorted by their DEFAULTSORT. The totality of the categories listed under Category:People categories by parameter, and all the child categories down to the last generation, with duplicates deduplicated, would seem one possible definition. (Not the subcategories of Category:People because that includes a lot of non-biog stuff like flags and books).

Perhaps such a category already exists and is used for some operations I know nothing about? Perhaps there are technical reasons why it can't be done? Perhaps the consensus is that it wouldn't be useful? I'll drop a note at a couple of relevant talk pages to alert them to this discussion. PamD 17:59, 16 November 2015 (UTC)

Notes left on Wikipedia talk:WikiProject Disambiguation, Wikipedia talk:WikiProject Biography, and Wikipedia talk:Categorization of people. PamD 15:19, 17 November 2015 (UTC)
WP:MOSDAB was changed in May after a long and sometimes acrimonious discussion. As far as I can see, the discussion didn't address the point that PamD raises, or even consider the matter from the point of view of an actual user of Wikipedia; but even so, I suspect that any attempt to re-visit the topic any time soon would be dismissed out of hand. Putting name lists in the "See also" section of a disambiguation page seems strange, and user-unfriendly; but there's nothing that can be done about it now.
Having said that: name disambiguation pages, even ones of the form Xyzname (name), are difficult for following up references such as PamD's examples (mentions of "Dr Xyzname", or "After Xyzname's breakthrough work in this field") because if there's a lot of people with the surname, often everyone called "Charles Xyzname" will be hived off into Charles Xyzname (disambiguation) and similarly with other common forenames such as David, Thomas, William, etc., which makes searching tedious and difficult. Sometimes name pages are divided in other ways.
I wonder whether in the long term it would be useful to have an easily-accessible search tool that would filter Wikipedia searches by reference to the subjects' properties in Wikidata. Not a user-unfriendly search where you enter language like "search for 'Xyzname' with 'instance of=human'", but rather a simple, accessible search, perhaps with tick-boxes for "People", etc. Not all biography articles in Wikipedia have corresponding Wikidata items yet, but that can be fixed.
Stanning (talk) 15:15, 18 November 2015 (UTC)
We used to have several hundred pages listing people by name. They were deleted after a MfD and DRV discussion in 2007.-gadfium 22:56, 18 November 2015 (UTC)
User:PamD, I don't understand how your requested system differs from the existing WP:NAMESORT system. WhatamIdoing (talk) 23:57, 18 November 2015 (UTC)

@WhatamIdoing:, @Gadfium:, @Stanning: Thanks for reading and commenting. I suspect I didn't make myself clear. What I suggest is not any manually-maintained list or set of pages like the ones deleted in 2007. The existing "NAMESORT" system specifies how names are to be sorted within categories, and is one of the prerequisites for my proposal to work.

I am suggesting that there should be a category, or a listing functioning like a category, which contains every item which is in any biographical category (including stubs) - whether Category:1917 births, Category:Mexican poets, Category:People from Headingley or Category:American football defensive back, 1980s birth stubs (some people will be in several). This list would be automatically generated, and therefore as complete as our categorisation and stub-sorting allows. It would be sorted using the NAMESORT system - ie all those with the same surname would appear together.

We could then offer a link to the relevant point in this A-Z listing as a useful "See also" link in any disambiguation page, and in any surname page, to help the reader who is looking for a person they only know by surname. For living people we can already do this - see this listing for people with the surname "Leeds", who are very difficult to find othewise because the word "Leeds" appears in so many other article titles.

The list a reader would find would be unannotated, just names - but if they have "tool-tips" activated (or is it a default - I mean the system whereby hovering over a link shows the lead sentence) they can skim through that list quickly to find the paleontologist or politician they are looking for. Even without tooltips, they have a list, in one place, of all people who have a Wikipedia article and who have that surname as their DEFAULTSORT, and that's more useful than finding the same names thinly scattered through a long list of article titles. That seems to me to be a really useful enhancement. What is needed is for the Category/Listing of "All people" to be created. Can it be done? PamD 09:54, 19 November 2015 (UTC)

I believe that there would be approximately 1.3 million biographies in this category – more, if you include redirects to alternate names. Are you sure that would actually be useful? WhatamIdoing (talk) 06:17, 20 November 2015 (UTC)
I can't be sure, but I think so: no-one would want to look at the whole 1.3 million, just at the A-Z chunk which has the surname they are looking for. If it was technically easier that 1.3 million could be held in 26, or 26x26, separate files (A-Z or Aa-Zz) or any further subdivision, generated automatically. Not necessarily in real time - daily or weekly update would still provide a powerful tool.
The information in the "DEFAULTSORT" field is a valuable potential search tool which at present can't be used except within a particular category. Trad encyclopedias, on paper, offer access by surname. Wikipedia doesn't, unless the surname is a string of characters which has no other usage ("Higginbottom"). I suggest we should enable readers to find articles by the surname of the biographee, even if it's "Martin" or "Leeds". PamD 09:26, 20 November 2015 (UTC)
This sounds interesting. Basically you'd like to have a Special:Allpages for people sorted by DEFAULTSORT. I think that would be nice to have. Alternatively, we could try to have Lastname, Firstname redirects to Firstname Lastname, and just use Special:Allpages. —Kusma (t·c) 12:45, 20 November 2015 (UTC)
SELECT page_title, cl_collation, cl_sortkey
FROM categorylinks
JOIN page ON cl_from=page_id
WHERE page_namespace=0 AND cl_type="page"
/* Looking at a single category: 1 sec */
AND cl_to IN ("Living_people")
/* Looking at 2,868 categories: 2 hours */
-- AND cl_to IN (SELECT cat_title FROM category WHERE cat_title LIKE "%\_births" AND cat_pages>0)
AND cl_sortkey LIKE REPLACE(UPPER("Leeds%"), "_", " ")
GROUP BY cl_from;
Since moving labs Dabfix would've taken hours to search for end of title matches. I intend to fix by using by using Special:Search. BTW, WikiProject Biography isn't tagging every person. Also SELECT SUM(cat_pages) FROM category WHERE cat_title LIKE "%\_births%" yields 1,1510,84. — Dispenser 13:04, 20 November 2015 (UTC)
Suffix search mostly working in Dabfix again. On a side note we could basically bot-create every surname article with the above query. — Dispenser 14:26, 20 November 2015 (UTC)
I am fairly certain we have a query on Wikidata for exactly this, or could easily have such a query. Not sure how often he logs on over on en.wp (I'll ping him later at WD), but @Jura1: probably knows. --Izno (talk) 14:49, 20 November 2015 (UTC)
I just happened to see your ping .. Surnames haven't been developed that much yet (it's being worked on now), but indexing by given name is fairly complete, at least for people from countries with Latin script (sample). Jura1 (talk) 18:10, 20 November 2015 (UTC)

Is there any interest in holding other "Region months"?[edit]

Asia Month looks to have been successful. While some of the editors creating new articles under the Asia Month banner would have created those articles anyways because that's where they worked, it does seem like other editors (myself included) jumped in to create new content in an area that they normally would not have.

I'd love to see a few more 'region months' to help combat the natural biases that I suspect Wikipedia has as an English language project (i.e. that we cover English-speaking areas much better because people write about the areas in which they live and because the sources are in English).

Would there be any interest in an Africa month, a Caribbean month, a South America month (or Latin America month), a Small islands month (for all of the tiny island nations), etc?

Who would organize it/them? What incentives could we come up with? When could we hold it/them? Mobile Squirrel Conspiracy (talk) 04:47, 19 November 2015 (UTC)

Language Link on Search Results?[edit]

Tracked in Phabricator
Task T119457

For someone who has just done a search on the English Language Wikipedia, and wants to do a search for the same string on the German Wikipedia, the choices seem to be

  1. a long series of clicks from that page, to main page to a location with a list of Wikipedias going to german, going and searching there
  2. knowing what the code (de) is for the other language and altering the URL (which works even though the information in the header may not be quite right for the other wikipedia.

How difficult would it be to add the complete list of languages on the left side the way that wikidata or interwiki links cause articles to be listed?Naraht (talk) 15:41, 19 November 2015 (UTC)

This doesn't sound too difficult to me, but I'm not the best person to make a technical judgement. I'd use it, if it existed. I've posted the idea to Phabricator; maybe User:Deskana (WMF) will know about whether this would interest someone on the mw:Discovery team. Whatamidoing (WMF) (talk) 23:12, 23 November 2015 (UTC)

wikidata interwiki affecting talk pages as well?[edit]

If Wikidata says that two pages are linked for example Johns Hopkins University Press and fr:Johns Hopkins University Press and as such have the other under "languages" on the left, shouldn't Talk:Johns Hopkins University Press be linked to fr:Discussion:Johns_Hopkins_University_Press?Naraht (talk) 15:45, 19 November 2015 (UTC)

This is basically phab:T30604. --Izno (talk) 16:21, 19 November 2015 (UTC)
Thanks for the phab link. A couple comments. 1) There are probably two separable changes, one to pull from wikidata, the other to pull from the associated article (so that fr:Discussion:Johns_Hopkins_University_Press pulls from both the wikidata for fr:Johns_Hopkins_University_Press and the actual text for fr:Johns_Hopkins_University_Press to look for interwikilinks that haven't been moved to wikidata. If the Wikidata one is easier, that would be fine. 2) In terms of pages that don't exist, what does the code do if one of the articles in a single wikidata entry gets deleted, does it still show the language? If so, then I think include the language regardless, worst that happens is that the user ends up being asked to create a talk page. (I should probably put these comments in the phab entry as well.)Naraht (talk) 16:45, 19 November 2015 (UTC)

Article Deleted[edit]

Someone once made a Wikipedia Trading Card Game. They got so far before the game became inactive. I would like to revive it. I know it may not be popular, but it may have hope. (Article Deleted is the name of the NEW game.) I'm returning...from the WikiDead. (But you still dare speak to me...) 21:25, 20 November 2015 (UTC)

There is a page at Wikipedia:Trading card game, is that what you are thinking of? Otherwise I cannot find an article. If you know the name or who wrote it perhaps some one can retrieve the lost article. Graeme Bartlett (talk) 23:59, 20 November 2015 (UTC)
Well, yeah, I want to revive it and improve it. I'm returning...from the WikiDead. (But you still dare speak to me...) 02:54, 21 November 2015 (UTC)
The talk page is at Wikipedia talk:Trading card game for anyone else that wants to join in. Graeme Bartlett (talk) 20:20, 21 November 2015 (UTC)
I may be interested in giving a hand in this, though I can't guarantee how much time I'll be able to commit to it.  DiscantX 04:15, 23 November 2015 (UTC)

OK, it seems people like the idea. If I don't get objections by tomorrow, I'm going to take this to Proposals. I'm returning...from the WikiDead. (But you still dare speak to me...) 12:56, 23 November 2015 (UTC)

In respect to non-US laws[edit]

The discussion, Wikipedia:Possibly unfree files/2015 November 10#‎File:Australian Aboriginal Flag.svg, resulted in "keep[ing]" the Australian aboriginal flag and solely relying on US law to deem it free to use in English Wikipedia. Of course we are not legal experts, according to disclaimers. I tried similar discussion but just about WP:non-U.S. copyrights page at WP:VPP, but other things overshadow that issue, and then that discussion is now archived. I was thinking about proposing to either add more headquarters, add more rules, or change rules. However, I want the issue to be brought to wide attention. I don't editors to believe that it is okay to distribute something copyrighted to online, even when it may not be copyrightable in the U.S. But administrators want to stick to US laws. Any ideas? George Ho (talk) 22:02, 20 November 2015 (UTC)

My idea is this: ignore foreign laws. Otherwise, Germany can dictate that we remove all swastikas, China can dictate that we remove all references to Falun Gong, Saudi Arabia can dictate that we remove all depictions of Mohammad, and North Korea can dictate that we remove all material critical of their Great Leader. --Guy Macon (talk) 22:32, 20 November 2015 (UTC)
Maybe copyright is not one of top important things after all? Do these countries have advanced technology to block these depictions on Internet? --George Ho (talk) 22:42, 20 November 2015 (UTC)
"Ignore foreign laws", for the reasons that Guy Macon gives. WMF is based in the U.S. (California) and its primary web hosting is in the U.S. (Virginia) so it has to respect U.S. laws and no others. [Except, does the cache in Amsterdam have to respect Dutch law?] The drawback is that WP is US-centric, not truly international, because while it can ignore the sensitivities of all other countries in the world, U.S. sensitivities must be meticulously respected. Stanning (talk) 15:26, 22 November 2015 (UTC)
In other words, whatever is out of copyright in source country may be copyrightable in the US. George Ho (talk) 17:36, 22 November 2015 (UTC)
Let me ask this. In some countries, it is illegal to display certain images. Should Wikipedia respect those laws and not display those images on its articles, even if they are perfectly legal to be displayed in the United States and most other western countries? There are plenty of other cases where US law and the laws of other countries directly conflict with each other in regard to the dissemination of certain information. If Wikipedia allows countries—or other groups/individuals using the laws of other countries—to have a "veto" on the dissemination of certain information, what does that do to the mission of Wikipedia? —Farix (t | c) 18:26, 22 November 2015 (UTC)
Since Wikipedia is subject to U.S. law, the U.S. – or groups/individuals using U.S. laws – have a "veto" on the dissemination of certain information. What does that do to the mission of Wikipedia? Stanning (talk) 21:31, 22 November 2015 (UTC)
I think we just have to add a warning to any media that we know to be a problem. Then it is up to reusers if they comply with the local laws, and they will be informed. Graeme Bartlett (talk) 23:23, 22 November 2015 (UTC)
Yes, that seems like the solution. I see that we already have Category:Restriction tags, and equivalents in Commons:Category:Non-copyright restriction templates, which seem intended for images; maybe we need standard templates with warning text to appear in articles which mention matters that are known to be legally restricted in certain countries (but legal in the U.S.)? Or do such already exist? And to raise awareness of the issues among editors (those who are paying attention) and/or patrollers? Stanning (talk) 10:17, 23 November 2015 (UTC)
The last template that did that was Template:spoiler, which got deleted in 2007. Of course, it's not law-related. We already have WP:content disclaimer, but it doesn't mention foreign law. George Ho (talk) 16:12, 23 November 2015 (UTC)

Anti-bullying task force[edit]

Hello community members,

I've been thinking for a few days now about the need to combat bullying and related abusive kinds of behaviors in Wikipedia editing. I've seen and been subjected to a lot of it myself.

I discovered that there is an essay about bullying at WP:BULLY which is a great starting place, but seems like it could be more fleshed out. And then there is implementing actions to curb bullying.

I've seen too much of it around here, and it often is in regard to content of articles. There is a big difference between reasonable and civil dialogue when there is a difference of opinions. Everyone has a different point of view. We are here to reconcile various points of view, and to decide on content in service of the reader. We want to get articles right and this comes out of such good dialogue.

But far too often, dialogue devolves into name-calling, pushy ways of speaking, condescension, insults, Wikilawyering, taking advantage of the naivete of new editors, and all this sort of thing. It tends to allow some people to swing false power around and to dominate articles, where there more subtle and nuanced voices who may be more polite and less aggressive then get drowned out by the dominators.

I know we have some mechanisms to work out issues about civility, and about people who are pushing content into articles against consensus or against good community judgment.

Often what happens is a long-term pattern where one or more editors will harass or hound another editor or group of editors. Often it works out along some ideological lines, as many topics in Wikipedia have some controversy around them. Some of the more experienced people know just how far they can push their behaviors without being too flagrantly in violation of a guideline and therefore able to be sanctioned. Some know how to insinuate insults, how to ignore another editor's fair points without it being so noticeable, how to change topics constantly or to use strawman arguments to try to make the other person seem wrong and foolish, and many other sorts of things. Some people cite guidelines like WP:IDHT and WP:DEADHORSE to try to get people to back off, especially newbies who can be intimidated by the alphabet soup. Sometimes there is a long-term pattern of one editor giving another editor so-called "friendly warnings" like "when people act like you are, they are often banned..." or "If you continue to act this way, you will go off a cliff" and these are not actually "friendly warnings" but more like understated threats intended to have a chilling effect on another editor. They even lead to a gaslighting effect where the victim can think "i must be wrong here" and clams up and backs off, not continuing to argue a point even though they may be right.

All these sorts of things are forms of intimidation that add up to bullying. I've been seeing it around in my year or so of editing, and now that i have some more experience, i recognize it as a major problem in Wikipedia. I don't think the system as it is, is good at dealing with this dynamic. I think the system as it stands sometimes even has the opposite effect -- it shoots the messenger. If someone does have the guts to stand up to a bully and bring it to a noticeboard, sometimes people come and gang up on that person and try to make them think they're wrong, and the bullying continues even in the forum where it's supposed to be addressed.

Is anyone with me on this? Do you see this going on? Do you have ideas for how to address it better?

Thanks. — Preceding unsigned comment added by SageRad (talk • contribs) 11:08, 23 November 2015‎ (UTC)

One thing is that bullying can traumatize people. I know it's "just" words on a screen, but these words mean real things. They carry real emotions. When someone directs aggression toward another, even through words to someone they've never met in person, it can do real damage. I have heard at least half a dozen people say things like "I don't even edit in this topic area anymore because of the toxic editing environment" or to that effect. I think that's largely from the bullying behaviors of some people that can traumatize people. Then the articles suffer because the more sensitive people, or often people with the less "mainstream" viewpoints, get intimidated, scared away. And people get hurt -- real people. It's not ok. SageRad (talk) 11:21, 23 November 2015 (UTC)
SageRad, I think we all see this going on all the time. I think you've summed up how I feel about this matter very well, so I'm definitely with you - I think the question is, what can we realistically do about it? I've been thinking, and I'm afraid I've drawn a blank - yes our current system doesn't work well, but places such as WP:TEAHOUSE, where almost every helper is kind and considerate to new users, does provide a safer environment for talking about bullying. I'm not sure if that even happens at the moment, but I could see a similar system being useful Face-smile.svg samtar {t} 11:34, 23 November 2015 (UTC)
Also, yes, a Task Force would be good - members could address bullying on behalf of a new, intimidated user by means of talking it out, dispute resolution or bringing to AN/I. Maybe that would be an idea? Face-smile.svgsamtar {t} 11:36, 23 November 2015 (UTC)
Yes, what if there is a group of volunteer users who agree to learn about bullying behavior, and then look out for it, helping out editors who are subjected to it, and trying to get the bullying person(s) to see how they're acting badly. Hopefully they would agree, and learn from it, but in my experience some people don't want to look at themselves and change, so if they continue to bully then they would need to be excluded from editing to make the environment ok for the rest of the editors, and for the good of the articles and the readers. `SageRad (talk) 11:41, 23 November 2015 (UTC)

One aspect of why i think this is needed, is that WP:CIVILITY is not just about "bad words" or one-time interactions. Many times it's a pattern, and a lot of bullying is done with no bad words at all. I seems innocuous enough to the casual observer, because they may not know the deeper meaning, or the history involved between two editors. That's part of how manipulative people work, and how they abuse people without other people noticing.

I would like to quote some good words from user Dennis Brown (who i hope will not mind being pinged) who said here:

We would all love a more civil Wikipedia, but blocking people for using bad words will only mean that the more passive aggressive types who hide their bullying and insults in saccharine laced words will be running the place. Some of the nicest people cuss sometimes. Personally, if I'm going to be insulted, I prefer the honesty of someone who just says it bluntly, not someone who hides it in clever language designed to intimidate and diminish me.

This is the same thing i have found. Repeated behaviors by a few people who have taken to hounding me and trying to grind down my self-esteem, using various turns of phrase and conceptual tricks to make it seem like i should just crawl under a rock and hide because obviously i'm too stupid to be editing at Wikipedia, and my point of view is just worthless, etc.... but using relatively innocuous-sounding words. It's tricky, and that's why i think we could use a volunteer corps of people focused on this. I'd volunteer. SageRad (talk) 13:22, 23 November 2015 (UTC)


Another aspect is that the person who is being bullied may sometimes react due to counterwill -- not wanting to be controlled for very good reason, feeling the sliminess of the interaction, and they might cuss in anger, justifiably, but then the bully will use that response to try to further characterize the victim as being unfit for Wikipedia ... and the cycle goes on and on, and the bullies get entrenched and develop gangs of mutual supporting "good old boys" who help each other out. SageRad (talk) 13:24, 23 November 2015 (UTC)

You might want to look over WP:WQA and WP:PAIN, and the discussions that led to them being shut down. --Ronz (talk) 17:36, 23 November 2015 (UTC)
Maybe it's more powerful if a change is cultural rather than procedural, but i could see that those two examples are really focused on short-term interactions and more on manners/etiquette and probably one-off personal attacks, whereas a deeper understanding of bullying and controlling behaviors could be useful to unravel some other sorts of deeper conflicts that occur on Wikispace. SageRad (talk) 17:48, 23 November 2015 (UTC)
Wikipedia works through consensus. The consensus as I understand it is that bullying is too subjective a label to enforce, the label is used as a bullying tactic itself, and the actionable offenses are best addressed by enforcing existing policies and sanctions. --Ronz (talk) 18:45, 23 November 2015 (UTC)
(ec)I think you would first need to define what is bullying behavior. If you simply disagree with some people, they take it as aggressive intimidation. I asked another editor to stop dropping small insults towards me and they accused me of personally attacking them. In this situation, I don't think either of us was being a bully.
But I've seen this situation occur a number of times. Editor A posts their opinion. Editor B mocks their post. Editors C, D & E get angry at Editor B for his/her rudeness. Editor B says he/she is being ganged-up on. Who is the bully here? Editor B for being sarcastic and rude towards another editor? Or Editors C, D & E for bashing Editor B for being caustic?
I saw this happen a lot during the GamerGate fury, where one person would be abrasive & insulting, then get attacked for their attitude and then claimed that they were being harassed. I don't think that this world is black/white on who is being victimized and who is bullying. Liz Read! Talk! 18:50, 23 November 2015 (UTC)
I agree there are some problems. However, I do have some problems with formalizing this for several reasons. (1) Some editors cannot take content based criticism and consider such criticism as a personal attack (and or bullying). I have repeatedly seen that after a polite exchange, editors who refuse to give up their fallacious reasoning and/or fringe theories are more bluntly put into their place. Who is to blame here? The people becoming harsh after been plagued with stubborn refusal to yield a ridiculous point (indeed WP:DEADHORSE is such a blunt discussion stopper but it is often justified), or the self-claimed bullied editor. (2) Harsh comments felt in one culture may not be intended by the poster from another cultures. Since EN.wikipedia is in practice the international Wikipedia this makes defining bullying tricky; and editors acting well within what is considered civil within their own social context may be seen as a bully by others. (3) Which leads to the next issue. If bullies are sanctioned - who decides. As many cases will be in the grey area all but the most black and white cases should be decided by "Wikisaints" who are in short supply. (4) Finally we should be extremely wary of more Wikilawyering. The original poster justifiably stated that the alphabet soup of wikirules and policies is confusing. I am afraid formalizing this proposal would add more of that soup. Arnoutf (talk) 19:23, 23 November 2015 (UTC)

┌────────────────────────────────────────────────────────────────────────────────────────────────────┘ The above three statements are definitely good thoughts, in my opinion, but these are not reasons why it couldn't work. Of course there is a relativity among people of different viewpoints. People with more similar viewpoints tend to have better edit histories and to forgive each other more often, and also don't butt heads as often. And of course, if someone is trying to push content into an article that is not supported by reliable sources, then they are probably at fault for contention. Bullying, when it happens, is in the behavior, not in the point of view of an editor. For the content it comes back to sources and having good dialogue. Those who don't have good dialogue and continue to push (either to block content they don't like or to push content that they want) and in the process hurt other users and make the editing climate contentious, especially if they continue to target or to go after or harass a specific editor(s) who they tend to disagree with, then they could be given a kind of notice. "Here's what we see going on... we see you misrepresenting the other editors even after they've explained themselves quite well, and calling them names and being condescending to them, and posting templates on their talk pages that don't appear justified," for instance.

I'm also wary of formalizing this, but it could be an advocacy group of volunteer editors who know enough to advise someone who comes to them if they feel bullied. Then they could use their experience to help work it out, if possible, or advise where and how it could be brought to a noticeboard for the best and speediest resolution. It could use the already existing noticeboards, and simply be a group of advocates (sort of like public defenders in the court system) who volunteer to help out because they've been there before and know what it's like.

I assure you that my intention in suggesting this is not to enable POV pushing of any kind. In fact, quite the opposite. I advocate for integrity to the sources and articles that reflect reality as best known according to reliable sources. SageRad (talk) 20:42, 23 November 2015 (UTC)

General, non-spam blacklist?[edit]

I know we have a spam blacklist, but do we have a unique blacklist for sites that are definitively established as unreliable by various Wikiprojects/the greater community? And if not, why not? I've been doing a lot of editing in the world of Indian cinema over the last year or more, not out of familiarity or interest in the niche, but out of frustration with the corruption that is so obvious and rampant. If the Indian cinema task force were to conclude through discussion that various sites were not deemed reliable, (let's say koimoi.com and boxofficeindia.com) they'd still have to manually remove thousands of unreliably sourced submissions each year, because there's nothing preventing the addition of these sources except for eagle-eyed editors, and the bulk of editing in this realm is by SPAs, sock farms, paid editors, and people who seem to think that the most recent higher box office estimate is the most accurate estimate, regardless of where it comes from. That sucks up a ton of volunteer time unnecessarily. This isn't limited to Indian cinema of course, because any time that someone submits a reference from forum.toonzone.net, that too should be on the blacklist, since nothing at that discussion forum is of value to the project. Or Wikia? Thoughts? Cyphoidbomb (talk) 03:10, 24 November 2015 (UTC)

Retrieved from "https://en.wikipedia.org/w/index.php?title=Wikipedia:Village_pump_(idea_lab)&oldid=692148330"