So, in the article it's claimed that Wales wanted to do a study using this method, and (implied) if appropriate, revise his conclusion.<p>This is four years ago, anyone knows if that happened?<p>Edited to add: One thing that has been annoying me a little, is when articles, regardless of their quality, are deleted because they lack notoriety. If these deletes adhere to the conclusions in this article (written by an outsider, then deleted by an insider, rather than both written and deleted by insiders), this strikes me as an example of a policy that should be changed in face of this evidence, since capturing the knowledge of these drive-by contributors seems more important than "saving space".
As a moderator of a wiki-project, I can confirm Aaron Swartz's results. After we had gained enough popularity, most of the articles were written by outsiders and core community was there only to maintain -- categorize, wikify, create interlinks here and there, delete garbage and, last but not least, have flamewars in the discussion pages.
A key takeaway for me is that you can't just take crude metrics, make some inferences and run with them. Jimmy Wales has been running his organization under a wide reaching and <i>completely wrong</i> understanding. That's because he equates number of edits with importance of contribution.<p>It's a basic psychology 101 concept, but one thats easily missed: don't equate your operationalized variables with the phenomena you are trying to measure.
Felipe Ortega wrote a Ph.D. thesis devoted to this topic. "Wikipedia: A Quantiative Analysis", available at <a href="http://libresoft.es/Members/jfelipe/phd-thesis" rel="nofollow">http://libresoft.es/Members/jfelipe/phd-thesis</a> His conclusions were featured on a Washington Post article, "Volunteers Log Off as Wikipedia Ages", discussed here <a href="http://news.cnet.com/8301-1023_3-10403467-93.html" rel="nofollow">http://news.cnet.com/8301-1023_3-10403467-93.html</a> and here <a href="http://news.slashdot.org/story/09/11/25/160236/Contributors-Leaving-Wikipedia-In-Record-Numbers" rel="nofollow">http://news.slashdot.org/story/09/11/25/160236/Contributors-...</a>
The article's dismissal of an editor by complaining that "[his] edits were all deleting things and moving things around" is a perfect illustration of why "bytes added" is just as bad a contribution metric as "edit count". We're not in 2001 anymore. The Internet is not short of bytes about Alan Alda or Anacondas, and nor is Wikipedia short of people who add bytes about them.<p>Newly-added bytes may be true or false. They may be useful or not useful even if true --- readers do not want every byte about a given topic, they want a few tens of thousands of the most useful ones. (This is entirely orthogonal to the inclusionist-deletionist debate about what <i>topics</i> should be included. Some bytes may be useless in the context of a main topical article about Alan Alda himself, but they would be very relevant to a subtopical article about Alan Alda's dental health).<p>For a popular topic, you'll have dozens of people adding bytes of varying quality. Insiders subtract the false or useless bytes (an action easily captured in statistics and then maligned on the internet by pundits), but also look at the true and useful bytes, fact-check them, and then <i>leave them in place</i>. This contribution --- curation --- is not captured in any statistics, but it is an important part of the mechanism by which you can have 1 expert and two enthusiastic amateurs stop by every few weeks on their lunch break to expand an article with no centralised notice or approval, without having have the place turned into a mess by the 97 vandals and well-meaning incompetents who came by in the meantime.<p>The real problem Wikipedia faces is in the long tail of topics, where there is only one person adding the bytes, and that person is either grinding an axe, self-promoting, or afflicted with incurable "nerdview". The well-meaning, harried, underinformed Wikipedia insiders inevitably screw up when they try to distinguish useless vs. useful bytes on these topics, but I wouldn't call the outsiders who added the bytes in the first place "experts". Unfortunately both sides' conduct may be scaring away the people who <i>are</i> actual experts on those long-tail topics ...
Out of curiosity, how many HNers write on Wikipedia? I do it every so once in awhile, but only only on a few topics I'm intimately familiar with, or to fix obvious vandalism.
Fascinating analysis, using a different metric and coming to a completely different conclusion than Wales did. If he's right, it'll be interesting to go back and reread what various people have written about Wikipedia. Four years later, has anybody run the larger-scale analysis he suggests here?
I found the followup at the bottom of the page equally interesting:
<a href="http://www.aaronsw.com/weblog/writefp" rel="nofollow">http://www.aaronsw.com/weblog/writefp</a>
The problem is; neither method alone is a good measure of value.<p>Plenty of articles I come across have "drive by additions" of some length that are unreadable, repetitive and disturbingly worded. There are contributors whose sole purpose is to go round and copyedit that content, which they may do in a number of edits... but change (by comparison) very few words. It can still take hours of effort.<p>A Wiki article is the sum of all those edits.
One of the primary reasons that Wikipedia is written by such a tiny percentage of its users is that the process of writing and editing is arcane and shrouded in one of the most insular online cultures in existence.<p>I don't know if greater access to its priestly class would mean a better or worse site, honestly. I'm sure there are thousands of people who are experts in their fields and could improve the content of the site greatly - but I can imagine that opening the floodgates and working to democratize authorship and editorship could also drag it further into the gutter.
Here is a dump I made sometime ago but just posted. I am torn on what to think. These companies know more about their products than the general public, and thus are more qualified to talk about them, but there's that little cynic in me saying they shouldn't. Take it with a grain of salt because some of them are legit.<p><a href="http://www.christopherwoodall.com/blog/?x=entry:entry101221-024331;comments:1" rel="nofollow">http://www.christopherwoodall.com/blog/?x=entry:entry101221-...</a>
Even though I disagree with his views on just about every political issue, Aaron is an inspiration for me. I would love to be in a position where I don't have to worry about money and invest significant amount of time in things that interest me, without worrying about money.<p>It seems like ever since I started working for myself, I don't have time for hobbies at all - everything I do has to go through a prism of <i>how will this affect my bottom line</i>...