Chad Perrin wrote:
In short, though, you'll probably find pretty
quickly that "everyone
here" does NOT in fact agree with that, or that Wikipedia is run that
way, as several responses to your email have already illustrated.
Good point. Looks like I was a bit overambitious in my initial email.
In truth, it
seems to me that what makes Wikipedia work is exactly what
the book you mention claims is invalid: the input of experts.
The book does not claim that the input of experts is invalid, nor am I
claiming the advice of experts is invalid. The book argues that under
the right circumstances groups of both experts and non-experts can make
better decisions than just experts. Whether this is true or not I don't
know, but I'm not claiming that it is gospel truth, only that I think
Wikipedians would enjoy the book. I did make the claim that people
reading this would agree with the claim, but apparently I was wrong on
that one.
I also disagree that the input of experts is what makes Wikipedia work.
I would make an educated guess that the majority of edits that the top
contributors make make are in topics they know nothing about: fixing
grammar, improving the flow, wikifying, placing articles in categories,
removing POV, reverting vandalism, etc. But is it true that the facts of
the article were originally added by domain experts? Not necessarily, a
lot of my favorite articles were written largely a sentence at a time,
where each person knew one fact or only a handful of facts about the
subject. Each of these contributors was most likely ignorant about the
topic as a whole, yet because each added a fact or two relatively good
articles have eventually been created. A good example of this is [[Tea]]
on en.
We are all
experts in some things, and we all know where to find the words of
experts online. Within our realms of knowledge, and within realms of
the knowledge we can find that is contributed to public circulation by
others, we contribute to the aggregate that is Wikipedia.
I don't disagree with this...
There's
nothing "emergent" about that aside from the simple fact that when a lot
of time and effort is put into collecting expert data a nice collection
of expert data emerges.
I agree that there is no emergence at the project level; we are all
trying to create an encyclopedia that is composed of a collection of
articles. But at the same time, Wikipedia isn't supposed to be the sum
of all human data, but rather the sum of all human knowledge. At the
article level it is possible for many people to each add a single fact,
that when combined allow conclusions to be drawn by readers that were
not known by any individual contributor. I would consider this to be
emergence, though if others want to say that this is only a
collaboration and is not emergent then I won't argue because I don't
think it particularly matters.
Meanwhile,
there's nothing distributed about the intelligence involved -- only about the way work
is accomplished.
I don't know enough about what qualifies and what does not qualify as
distributed intelligence (outside of the mostly black and white
definitions used by computer science) to pass judgement on this, but
I'll take your word for it.
In another reply, Stirling Newberry says, "I've read it, it is the "Dow
36,000" for the 90s." I had never heard of this book so I looked it up
in Amazon, and the first review says that the main problem with the book
is that it is written by people who think they are smarter than the
market. This is interesting, considering the Wisdom of Crowds is about
how individuals aren't consistently smarter than the market and about
how to use the intelligence of markets to make good decisions. As a
practical example of how Surowiecki's ideas could be applied by
individuals in WP, consider the process for voting on which articles
should become featured. The task at hand is to separate the excellent
articles from the merely good. According to Surowiecki, when one has to
make a decision through a group then the best decisions are made by
groups which have diversity, independence, and decentralization (to
borrow from Joseph's mindmap).
This means that, according to Surowiecki, the quality of judging of
articles will increase the more diverse the group is in terms of what
they know, their values and beliefs, levels of intelligence, etc. The
takeaway would be that we should get as many people to participate in
choosing articles to go into WP 1.0 in order to make the best decisions
about what stays and goes.
Secondly, each individual should act independently. That means that s/he
should read the article and write down their reaction to it BEFORE
reading other people's comments. If their comment is completely
redundant then they should possibly delete it, but individuals should
not read other people's comments before reading the article.
Thirdly, the process should be decentralized. WP does an excellent job
of this already: there is no one telling people what to do, and everyone
just works on what they are interested in when they feel like it. Going
back to our example about choosing FA/WP 1.0 articles, our
decentralization is an asset because if there were someone giving orders
for people to vote on a certain number of articles per week or else have
their account banned, one could easily imagine a situation of people
voting and commenting without really reading the articles first.
Anyway, I didn't mean for my post to be offensive by putting words in
the mouths of othhers, nor did I mean to imply that I believe all of
what Surowiecki says is gospel truth. However, I did enjoy the book, I
do think that it relates to Wikipedia (even though Jimbo disagrees), and
it got me thinking more about the social dynamics of this project. On
that basis I still recommend the book.
Alex "pHatidic" Krupp
http://www.alexkrupp.com