Wikipedia talk:Beware the myth of Wikipedia written by passing strangers

Is this intentional?

I was going to help work on this essay, but I'm confused. Please excuse my ignorance, but I'm not familiar with all of Wikipedia's various types of subtle humor. This essay seems to make all the same mistakes that Jimbo made as pointed out in the "Who Writes Wikipedia?" study conducted by Aaron Swartz. It uses the same 80/20 rule, the same 500 people write Wikipedia and Jimbo knows them all. This essay also uses just the edit count numbers—instead of looking at the substance of each edit—to determine which group contributed the substance of each article, which Swartz pointed out as Jimbo's major failing in his study.

If I had to guess, I would say that this was intentional because of all the coincidences with Swartz' study and the fact that his study isn't mentioned anywhere in this essay. So should I be adding things like Twinkle and Huggle help improve stubs in to GA's and FA's and similar things like that? Confused. 64.40.54.29 (talk) 19:16, 13 July 2013 (UTC)

I have to agree this page is just strange.

In fact, as I understand it the edit counts being seem to be general ones i.e. including edits outside article space. While I don't deny most of these things are important to creating an encyclopaedia, it seems clear it isn't what the supporters of the idea this study seeks to rebut are talking about. (Reading more carefully think I'm mistaken on this part.)

And as the IP has said, it seems to go by number rather than the substance of the edit, so someone who takes 50 edits to write a stub or makes 50 spelling corrections has done a lot more than somene who takes 5 edits to change a stub to an FA, and both have done less than someone who vandalises 100 times. While I don't think it's too helpful to get too much into value comparison, supporters of the idea aren't talking about gnomish stuff, probably not even adding wikilinks, formatting refs, or 'most of the rewrites to match format guidelines and template features', and perhaps even wikifiying but instead writing the encyclopaedia content. (It definitely seems clear anything contributed by a bot isn't something supporters of the idea are referring to.) Just to re-emphasise, I don't think anyone is denying all this is important, simply that it isn't what people are referring to when they talk about nom experienced users contributing a lot of the cotent.

Similarly it talks about article creation, yet as we all hopefully know IPs can't create articles and while there is AFC, it's hardly surprising that registered users are more likely to create article for this reason alone. Not to mention while I'm a deletionist, I do acknowledge that there must be times when an article is deleted even though the subject is notable simply because the writer couldn't establish it as such and it's not surprising this is less likely to happen to experienced editors. For that matter, experienced editors are more likely to be able to see gaps which they understand we should cover (where as an IPs gap may be something we won't cover). It may even be the case that an experienced editor may be more likely to create a stub since they may think this within the wiki ethos whereas an inexperienced editor may think an article with little content is not worth creating. And when we talk about articles created, are we including disambig pages and redirects which while useful are unlikely to be what people are thinking about? Either way, who created an article doesn't tell us where the content came from.

I would definitely be interested in decent studies looking the idea, but it seems clear this isn't it. To be fair, it isn't easy to do. You'd either need to take a large enough random sampling of edits and analyse what was contributed (which would likely include checking the editor history to make sure the content wasn't removed). Or a large enough random sample of articles and analysing the history to determine precisely who contributed what. Either way, a decent study even working full time could easily take weeks or more.

As I said earlier, I don't think anyone is denying all the gnomish and other stuff isn't important, nor that most of it probably comes from experienced editors, but even if you want to look what percentage of it comes from who, you'd still need to analyse the edits rather than just going by the number and ultimately it's probably even harder to actually come up with some useful measure.

Nil Einne (talk) 18:52, 14 July 2013 (UTC)

Worked example

I just walked through the edit history for a fairly typical, randomly selected stub (and stubs are fairly typical, since more than half of Wikipedia pages are stubs). Here's what I found:

1. Two unsourced sentences added in 2006. Editor's 29th edit out of 108 ever.
2. Infobox added.
3. Infobox corrected.
4. Bot (added a WP:General reference to support infobox contents).
5. Stub tag.
6. Disambiguated link.
7. Interwiki link.
8. Grammar correction (zero content change).
9. Bot.
10. Infobox addition.
11. Interwiki link.
12. Change stub tag. Introduce category error.
13. Correct category error.
14. Vandalism.
15. Revert vandalism.
16. Change stub tag.
17. Add navbox.
18. Changed category.
19. Expanded infobox.
20. Copyedit (zero content change)
21. Add picture.
22. First-ever change to factual article contents, 2010: changed "considered to be the same species as" to "considered as a subspecies of". Still unsourced. Approximately user's 30,000th edit out of nearly 43,000 to date.
23. Add external link.
24. Add interwiki link.
25. Change image.
26. Bot.
27. Bot.
28. Bot.
29. Vandalism.
30. Vandalism.
31. Revert vandalism.
32. Bot.
33. Interwiki link.
34. Bot.
35. Update infobox link.
36. Update infobox source.
37. Update infobox link.
38. Change stub tag.
39. Bot.
40. Bot.
41. Bot.

That's it. There were exactly two edits that added actual content to the body of the article. One of them was by one of those "passing strangers", who contributed two sentences here and never made it into the "highly active editor" category (defined as 100+ edits in a month). The other was by an editor who has made thousands of edits, who contributed approximately two words.

The other 39 edits were about 20% infobox tweaks, 20% interwiki links, 20% stub and category links, and 10% vandalism and its repair, with two minor copyedits and two changes to the picture.

Who would you say actually wrote this article? WhatamIdoing (talk) 22:42, 14 August 2013 (UTC)

Tree for forest. Stubs may make up the largest percentage of articles; but they're certainly not the type of article your average user sees. The average user is far more likely to be looking for sparrow than Arabian Golden Sparrow; far more likely to be looking for Hungary than Kőröstetétlen. Try doing this study for a vital article. (If you take a very old vital article like periodic table, don't take into account the initial stubbiness; that's from an old era when most of the vital articles were missing and passing strangers could improve on that. The situation now is quite different and this mostly happens for much less important stubs, if at all. A plausible exception might be news, an area of WP I'm not so versed in yet: I wonder what happens there.) Double sharp (talk) 08:06, 25 November 2013 (UTC)

Closed society

More often in chatting with civilians I encounter the opposite assumption, that Wikipedia is edited by a carefully selected group of insiders who skillfully block the insertion of alternative viewpoints and new ideas. I lack the motivation to expand this into a mythological essay, however. Jim.henderson (talk) 14:08, 24 November 2017 (UTC)

I suspect that this is partially due to the authoritative-looking encyclopaedic style we write in and partially due to people's bad experiences adding in their pet fringe theories (which would be wholly deserved). The bar for a new user to come in and actually improve things. without doing some serious research has long since gotten very high, which is the entire point of this essay. Combined with what looks like a procedural maze if you see it all at once – which you will if you promote your own POV or some fringe theory and get warned and blocked – it is no surprise that the idea that WP is edited by a group of insiders conspiring against such viewpoints springs up; all it takes is for human psychology to do the rest. (See especially point 62 on User:Antandrus/observations on Wikipedia behavior, regarding the "inexhaustible source of folly": I can't possibly be wrong – all of them must be.)

As another example of the 1-edit-myth at work: I watch all 118 articles on the known chemical elements, many of which are already FA's and GA's thanks to the efforts of a few dedicated users, and my general impression is that most IP or new-user edits at this stage are either vandalism or minor copyediting (the second of which is fine and helpful, but doesn't really add content, and the barrier to doing so is still just as high). The one exception I can think of is when new elements get named (which really is news). This is admittedly just from memory (not from actually trawling through 118 page histories), but it is illustrative. Outside news, the early days of WP when passing strangers could offer significant help on content have long vanished. Double sharp (talk) 10:55, 7 March 2018 (UTC)

P.S. In the interest of completeness I should note that the "few dedicated users" I mention for the elements include myself, though since self-praise is never advantageous I note that the adjective is mostly intended for the other ones. ^_^ Double sharp (talk) 00:07, 26 March 2018 (UTC)

Yes, the article describes a myth that faded years ago, partly replaced by one with surprisingly better basis in fact. I spent this afternoon coaching at an edit-a-thon in Brooklyn, where newbies had been led to believe that this was their chance to write Wikipedia, specifically new articles, and more specifically biographies of living artists who don't have one yet. Wrong on more than one level. First, anyone can write something in an article anytime; no rule says they need our help. Second, they need experience or help, to start a new article on anything. Third, a new BLP is the most difficult thing they can tackle; even with our help they are unlikely to find the refs they need. Fourth, the suggested biographies are mostly unarticled because others have tried and failed to find good refs. So, it's pretty much doomed to futility. Hmm, this is only slightly related to the topic at hand; I need to bring it to more relevant attention. Jim.henderson (talk) 04:01, 11 March 2018 (UTC)

@Jim.henderson: I recall an interesting Signpost op-ed draft (sadly never published) that touched on this perception, although its focus was a little different (it was on how to change this perception among millenials specifically). I wonder how much prior experience many of these newbies have with writing encyclopaedic or even just descriptive and neutral pieces outside Wikipedia. It's not as easy as it looks to marry the three ideals of being complete, being neutral, and being at an appropriate level, but knowing that only comes through experience – especially because, like everything else, we usually see the cases where it is done really well, and those are exactly the cases that look as though they were easier to write than they really were. Newbies not usually knowing this would explain some of the ambition, although I suspect that more of it could be explained by misconceptions about the purpose of Wikipedia. Unfortunately I am not sure there is a better answer to that than simply explaining and pointing to policy.

I would think that a better thing to suggest to most newbies is not to write new articles, but to rewrite old ones that aren't so good, but getting them to dare to do that runs squarely into this assumption – either they won't dare to do it because they feel the insiders have done it already, or they'll add alternative fringe viewpoints, get correctly reverted, and then the myth is confirmed (which is fine for fringe-theory promoters, but the problem is that it's not only them who believe it). The reason I suggest this is that if an article hasn't been created by now, its subject is either at a really high level (where you really would need a lot of qualification or at least a lot of reading on the subject to even begin to write), or it is a relatively new field of inquiry (and then it is really hard to be neutral). To combat this you would need greater awareness that Wikipedia is a work in progress in that it doesn't just need new articles, but it needs improvements to its old ones. It would be interesting to see the relative activity of old hands and new editors on existing and new articles. But at this point, I agree that we've left the topic of the 1-edit-myth and moved on to outreach. Double sharp (talk) 15:29, 24 March 2018 (UTC)

@Double sharp: thanks for the draft link from a few years ago; it is well-written and has not gone obsolete. A somewhat similar discussion is underway in Wikipedia:Autoconfirmed article creation trial/Request for comment on permanent implementation by which starting a new article will require Autoconfirmed status. As it happens a hard week has left me with an illness and an injury both of them minor but enough to keep me away from editor coaching and similar contact sports for another week. Some of the time can go into writing an essay that's been fermenting in the back of my mind. At first it was to be, "the Art + Feminism project is futile" or "Editathons are useless" but maybe "Wikipedia already has all the articles it needs" would be closer. Of course I don't believe any of these things, or at least not that starkly, but the last of these suggests adding some stern language into WP:Your first article. Jim.henderson (talk) 19:35, 25 March 2018 (UTC)

@Jim.henderson: I love the wittiness of the phrase "editor coaching and similar contact sports"! ^_^ I have already supported restricting article creation to autoconfirmed editors, as its results in the trial seemed very positive, although I might sometimes wonder if instead of the good message of "article creation is tough and not what you should start on" it might perpetuate the bad message that Wikipedia is somehow closed. It would once again be interesting to see what people expect Wikipedia to be like and how far that expectation differs from reality. (I suspect that a lot of it comes from our style, which is a reflexion of our mission, but I am not sure how you could prove this. Maybe looking at other wikis, but for most of them I think the effect of differing missions would be swamped by the simple difference of scale.) I support the addition of some stern language, even if I likewise don't completely believe all those pessimistic statements. Looking forward to the essay! Double sharp (talk) 00:01, 26 March 2018 (UTC)