[SharedKnowing] Distribution of contributors

[William Pietri]

Ben Kovitz wrote:
> Oops, I was wrong!  I read more of the Aaron Swartz essay at 
> http://www.aaronsw.com/weblog/whowriteswikipedia, and he says that the 
> heavy editors mostly just make small formatting changes, and that most 
> of the substantive content is entered by people who make very few 
> edits.
>
> This suggests that raising the bar even a little bit against "go live 
> right now" could be very costly.
>   

I think there's another reason for that as well. Small contributions and 
edits are the gateway drug.

In my day job, I work a lot with community-focused internet companies. 
In thinking about the business, we'll often draw what's known as "the 
funnel". At the narrow end of the funnel are our core contributors, 
those heavy editors. At the wide end is anybody who touches our site. In 
between are the various intermediate stages of participation. Naturally, 
you want the mouth to be as big as possible, but the rest of the shape 
is important, too.

One of the key things we look at is how much it narrows stage, and what 
drives that narrowing. It's my firm belief that for these sorts of 
applications, you should follow the same principles as designing a video 
game. Although the barriers to advancement at the narrow end of the 
funnel can be quite high, they must be very low at the wide end. The 
novice must be able to easily and quickly take some action and have 
quick positive feedback.

Take my experience on Wikipedia as an example.

Well before I created an account, I was reading articles. On a couple of 
occasions I noticed a minor error, and made quick copyedits. They were 
instantly live, and I couldn't have been more pleased. The total 
overhead was almost zero (find edit button, click, find misspelling, 
fix, click save), and I got an immediate positive reward. In game terms, 
it was equivalent to shooting that first, easy zombie, or guiding 
Pac-Man onto the first glowing flashy dot that lets you chase the ghosts.

At some point later, after a walk by Golden Gate Park's bison paddock, I 
wondered what the normally quiet bison's call sounded like. Wikipedia 
didn't have the answer, but I found the sound on some government 
website. Uploading that to
Wikipedia was more work and worry: I clearly remember being scared that 
I had screwed things up somehow. But it was a success, and I was 
delighted. Level 2, achieved!

Now, four years later, I somehow ended up an (albeit not a hugely 
involved one) and I just spent several hours last night and this morning 
doing what, if I'm honest with myself, sure looks a lot like work. But 
without those first, high value-for-effort interactions, I'm sure I 
would have done the sensible, responsible thing and not gotten involved 
at all.

So going back to the funnel, I think Wikipedia did some great things at 
the wide end:

   1. With every URL durable and every page easily linkable, they
      maximize incoming readers.
   2. Since the idea of editing is prominent (e.g., "everyone can edit")
      they get readers to consider the possibility of editing.
   3. By making editing quick and easy (registration not mandatory,
      "edit this page" an obvious tab at the top, simple interface and
      simple editing flow), they convert a lot of those considering
      editing into actually editing.
   4. Since every saved edit is immediately displayed, they give instant
      feedback, which is generally positive when the user has improved
      things. That makes it much more likely that people will try
      editing again in the future.
   5. If mistakes go live, that can be ok, as they often provide
      something for some other user to fix, increasing the number of
      people who cross the line from reader to editor.


A lot of the attempts I've seen at community sites fail because the 
desire for control or fear of user action leads them to squeeze the 
funnel drastically at the wide end, providing later stages with too 
little input to build a thriving community or a useful product. Not that 
bad content isn't a worry, but Wikipedia reduced the impact of that by 
making reversion of recent contributions easy. Unlike the real world, 
cleaning up vandalism on Wikipedia is generally easier than vandalizing, 
which makes a lot of real-world intuitions incorrect.

Not that "go live right away" is appropriate for every situation, but 
when you can to it, it's a relatively easy way of converting potential 
contributors to actual ones, and occasional contributors into regular ones.

William

-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mail.citizendium.org/pipermail/sharedknowing/attachments/20080329/2a5f1acb/attachment-0001.html