There have been a lot of people who have tried to figure out what the magic sauce was that made Wikipedia so popular. Two Sheds recently linked to this article The contribution conundrum: Why did Wikipedia succeed while other encyclopedias failed?. It mentioned E2 near the start, and implies we're a failed encyclopedia, which I guess I can get behind, because E2 was never meant to be an encyclopedia.
E2 is intended to be a web of ideas, connected both by intentional direct links and by the naturally-arising web of softlinks that evolve as people travel about, linking knowledge, stories, and emotions together in a useful and sometimes-surprising web. The intermeshing purposes of the content, both educating, entertaining, and exploring, entangle the people creating those links, and hence you find a community intertwined within the writing, making the content and the people a bit hard to separate. Which, I know for some of the more clinical types on the site, is a problem, because it means emotional turmoil affects the health of the site and potentially means loss of material. It also makes material written by noders who have moved on sometimes weirdly hollow. Like rummaging through the attic of a recently-past relative and wishing you could talk with their ghost...
But, I wasn't intending to get too far into the matter of E2 itself, the article largely argues that Wikipedia succeeded due to being based on a familiar trope, an encyclopedia. But that's wrong. If you're like me, an active user of E2 who's talked to people IRL about the site, you've said something like "Everything2 is this community writing site, kind of like Wikipedia..." Wikipedia made a relatively new model of content creation popular enough that it is now a useful way to describe other things. And so I found it ironic that the article writer decided to say that a familiar model was what sold WP. The idea of writing words that might get mangled by five more people only minutes after you've made them is freaking unsettling at first. But it's shown undeniably useful results, mainly because the proper community is present to encourage good behavior.
But arguing that people understand what an encyclopedia is and even that they understand how Wikipedia matches an encyclopedia, I think, is wrong. Most people still don't entirely understand what encyclopedic writing is about. You can tell this because people regularly reference Wikipedia pages as if they were a primary source, and don't even bother to reference a particular revision to ensure referential integrity. You could tell this even before Wikipedia because people would look to the encyclopedia text rather than to its bibliography for research. You can tell the lack of understanding of even some of Wikipedia's own community by the bitter debates in their Talk pages. Look at the edit wars and raging about adequately presenting points of view on subjects such as Russian occupation and conflict in Chechnya; or the events of the Bosnian War; or even any revered, living monarch.
They did hit on one of the far more important lessons, however, which is the low barrier to entry. Wikipedia lets anyone edit most of its articles, even now, despite the great rise of semi-protection of popular articles in the last couple of years. A better place to look for this lesson is to look to the slow death of LiveJournal versus the massive growth of Tumblr. LiveJournal has a pretty obvious metaphor which is universally understood: the journal, shared privately or publicly. What's more, it provides far more control over the exposure of your content than Tumblr presently provides. But it has seen declining numbers while Tumblr, whose name seemingly means nothing, is usurping it as the go-to place for the quick, semi-public introspective rambling.
Why these simultaneous trends? For one, Tumblr is young, so it retains a buzz about being the hot, new thing, but that's not all. If it were, Wikipedia would be in a death spiral for being just about as ancient as LJ. Signing up to Tumblr has always been easy and very fast and the posting interface is spartan and obvious. LiveJournal has since copied some of this, most notably allowing sign on from Facebook and other external accounts. But its greater set of feature makes it seemingly harder to use, and hence not as likely to grab somebody.
If you'll pardon a return to navel-gazing at the end of my ramble, E2 has the same issues as LJ here. Our sign up page doesn't have AJAX controls built in to say "that username is already taken", "those passwords don't match", and so on that all modern sites do. We don't let people use their Facebook/LJ/Twitter/Tumblr/etc. or other OAuth account be their account here (provided the username is free). And if you click the sign up button on a page down where it says "If you had an account, you could write something here.", you don't get taken to a page, post sign up, that lets you start writing your thing right away. Beyond that, it'd be helpful to make our WYSIWIG editor actually show links as they appear, rather than [showing them like this], making the whole linking thing less obvious than it should be.
These are idle thoughts as a personal matter, not intended as policy or things directly on my docket to work on as staff. I'm not especially in a place where I can be very productive for the site right now, sadly.