Legitimate decreases need to be considered, too.

I personally contribute less to Wikipedia because it usually already has the information, and thus I don't need to edit anything. My point is that lack of participation is inevitable as the wiki matures, and needs to be taken into account in all of this.

Most people I know say the same thing I am. We would rather contribute to other wikis that are missing a lot more content.

I'd second that. I have been thinking about the natural lifecycle of a wiki community.

For static information (not current affairs, etc), there is an obvious sequence of stages:

1. Enthusiasts create shedloads of stubs. Mediocrity is more than good enough. Errors of fact abound because nobody knows any different.
2. New editors are motivated to join. The community expands, the quality improves, the wiki expands.
3. Once all the easy stuff has been done, some editors have been sucked dry and drift away. Others hang around nit-picking, bickering and writing bots to do that for them. Wannabee editors find that it has all been said, their attempts at engagement encounter the stick-in-the-muds and they give up in disgust.
4. Change or fossilise.

So here we are. We can pretend we are still young and world-changing or we can realise that we have grown up. What we need now are the experts who can do the hard stuff - fill in the gaps where enthusiasm isn't enough - and are used to the academic grind of digging out references for every factoid challenged by the "I have the right" grumps.

Don't go bemoaning yesterday's community - go motivate tomorrow's.

Steelpillow 21:49, 11 March 2011 (UTC)

I don't think all really has been said, if your interests go beyond popular, recent United States culture.

That said, you do have a point. While I spotted another needs-improvement-badly article just last week, it's not common that I look up something which obviously should be in Wikipedia, and find it either completely missing or blatantly inadequate.

One thing that would help would be less knee jerk deletionism. The article I spotted - and haven't fixed yet - would quite likely have wound up at AFD if newly created, helping no one. It's a BLP, and while the person is justly notable, the references on the wretched stub article are not to reliable sources. Nonetheless, there was (barely) enough info to answer the question that caused me to look him up.

At the Foundation, we tend to call this the "gold rush theory" -- that there was a rush of fast contributions that filled out most of the important articles, and that now we're in a maintenance mode that should just focus on quality.

That idea tends to ignore two things: first, that we're still have a crazily large systemic bias. Check out the map of geotagged articles in Further Reading for the kind of visualization that will show you what that means, how much knowledge there really is left to catalog.

The reason that new knowledge isn't arriving as fast, and why fewer experienced editors are having to do more of the work maintaining/expanding the articles we already have, is because the pipeline of newbies has been stalled.

In short: if Wikipedia is a constantly evolving project, then we need lots of new faces around to keep it so, or its going to become stagnant. And stagnant, old, out-of-touch content is what is putting the stranglehold on old media institutions unable to keep up. I sure don't want Wikipedia to join them.

The fact that the pattern of stagnation occurs in language editions with very different completeness characteristics completely invalidates the gold rush theory as the primary factor limiting participation or retention. Given that we're seeing similar increases in revert behavior, it's much more likely that communities in many languages have experienced Eternal September phenomena, and have also adopted similar quality assurance mechanisms to deal with those phenomena, leading to similar drops in retention.

Good point about other languages. I was mostly thinking about English in my above reply.

I would think of it not so much as a gold rush as picking the low-hanging fruit. For example in the areas of mathematics that interest me, the well-known ideas need little more than tidying up and, hard for most mathematicians, introducing in simple terms. The less fashionable topics remain poorly represented.

I find it hard to believe that Eternal September has arrived, any more than I can believe that the young have less respect for their elders than they used to, or that policemen are getting younger, or that nobody cooks like my mother used to. I think it is more a case of the old hands having built up knowledge and experience they have forgotten they did not have back in the day. To me, Eternal September protagonists are living evidence of the stick-in-the-mud phenomenon - the ones who dominate Stage 3.

The common pattern across language communities does come as a surprise. But to me it does not suggest Eternal September, rather that the stick-in-the-muds have gone global. I can well believe that the global rollout of tools and policies designed to stifle inexpert enthusiasm is doing just that. Revert rates without context prove nothing: they may be down to young vandals, or old stick-in-the-muds clinging to their territory, or arguments over translating from the original English, who can say without the context. Perhaps many language communities are not so much in Stage 3 as being sucked into it prematurely by the biggest and most influential community - a community that has already got there.

Steelpillow 16:17, 12 March 2011 (UTC)

In addition to the "gold rush" and "low hanging fruit" principles, I would like to add that a major problem on Wikipedia is that articles on small self-contained topics are much more well-developed than core encyclopedic content. While almost all basic topics have articles with sources, many of them are still in a miserable state, because editors need to bring together a wide range of sources and expertise to improve them: it is much easier to work on quirky topics, where there are only a few reliable sources, so that selection bias is not a problem and relevant knowledge can be acquired easily.

Two effects that are common to all languages are stricter BLP policy and competition (for user attention) from Facebook. Both happened at around the same time (2006-2008) as the stagnation of activity. Whether this offers any explanation, I don't know.

Facebook might compete for people's time, but I don't think it substitutes for the reasons you edit Wikipedia.

For me I started out reading Wikipedia and had added a little content about my local suburb, but the motivations for those edits was curiousity, just to experiment with this Wikipedia thing. For the next couple of years, I just did little edits here and there, probably because I was annoyed with something I throught wrong or a bit incomplete. I had some of those edits deleted for reasons I never quite understood.

Then I read (via one of the donate banners) of the Jimmy Wales quote:

"Imagine a world in which every single person on the planet is given free access to the sum of all human knowledge. That's what we're doing."

And that changed my perspective of what Wikipedia was and what it was capable of. It was a "road to Damascus" moment for me. I became much more actively interested in writing new articles on topics that weren't there. Now my personal mission has been to try to collate information on missing topics and write at least a start class article.

I still don't necessarily like the way Wikipedia operates (technically or socially), but I'm probably more thicked-skinned about it. I'm here for a reason and I will persist because I am a "true believer" in the Wikipedia vision.

Perhaps we need to understand more about *why* people edit Wikipedia and therefore what will help them to become/stay active editors (assuming their "why"s are positive ones).

There are some interesting posts about why women edit Wikipedia on the Facebook "Wikipedia Women" group (you can search for it there). - PKM 23:50, 12 March 2011 (UTC)

It's not clear how a study of editor retention relates to systemic bias, in particular geographic bias. Most of the editors we are worried about retaining have little or no interest in writing about under-served geographic regions. To get more articles about a place like Botswana, we would have to attract more editors from there, something we have never been able to do very well, whether or not we retain the existing editors.

The other problem is that systemic bias is not solely a feature of Wikipedia, but also of our available sources. Recruiting a bunch of editors from, say Botswana isn't going to lead to the same density of articles we get for similar sized regions in the US and Europe because the source raw material is much sparser, both online and in print. To overcome Wikipedia's geographic bias, we'd basically have to give places like Botswana a modern publishing industry, which would probably require solving the North-South economic divide. Good luck with that.

If Wikipedia really wants to cover under-served regions of the world, then we probably need to relax our requirements for notability and against original work, for those areas. A lot of our "missing" information might not be published at all, or not in English. It's hard enough for native English speakers to defend an article with sources in English against deletionists; imagine sending Botswanans into that snake pit. In areas of the world with inadequate sources, Wikipedia would probably have to take on more of the work we currently delegate to the publishing industry.

It seems to me that part of the systemic bias is due to deletionist behaviour, particularly these (bad) deletion reasons:

• I'm not interested, therefore it's not interesting
• I've never heard of it, therefore it's not notable
• I looked for sources using an anglicized name, and found nothing; therefore there are no reliable sources
• I can't read the sources (I only read english) and have never heard of them, therefore they are unreliable
• This is bizarre (not my native customs), therefore it's "fringe" and should be deleted

Eliminate this kind of illogic, and there'd still be some bias, based on editors' interests, but wikipedia article density would come closer to actual source density.

And that's enough; I don't think anyone expects wikipedia to go beyond that. And realistically, it wouldn't match even actual source density - too few people who read language X chose to edit language Y's wikipedia, even when Y is english. (But heavens knows, for those who do read multiple languages, there's lots of low hanging fruit still left to add to any wikipedia, including the english one.)

1. Do we really NEED more articles on Botswana? Do the readers WANT more articles from there? I think the bigger issue is the youth bias. there are a lot of older readers of the Internet, but Wiki is youth-oriented in various ways (even the micro font, check out the ref 2.0 toolbar for the worst).

2. I think participation is very important. And I totally HEART having a decent edit user interface...it's a freaking CRIME that we are behind 1995 MS Word. Sooooo much time wasted for writers in the interface (even this edit window, look how tiny it is). And the lack of comprehension of how this impacts ergonomics reminds me of a bunch of Unix users, who think everyone should just enter C: prompts and why they heck would want want to click on stuff.  ;)

3. All that said, we should think about our work product and our READERS first. Yes, don't kill the golden goose. But the objective is NOT to have a well populated site. That is a means to an objective. Intead we should think about what our content is and how often it is used and not used and what criticisms are made of it. Although I'm not crazy about a bunch more Botswana content, I did appreciate that the foundation fellow was at least trying to talk about what sort of encyclopediea we should be (where our weak points are) on the output side.

TCO 20:45, 14 March 2011 (UTC)

I don't know about Botswana, but my Ethiopia-related articles appear to be well received. The last two articles I created which made DYK achieved over 1,300 (Bakri Sapalo) & 1,400 (1960 Ethiopian coup attempt) views. A lot more people are interested in African topics than you might think.

I actually totally HEART an attitude of wanting to know what is going in around the world. We have a world of Starbux-zipping multiculturalists who have no concept what Burton did undercover. Oooh-rah! The world is strange and interesting and diverse...much more than the Wilsonian neoliberals think.