Knol and Wiki 2

Edmund Lee, English Heritage, wrote about my last posting (read it first!):

"Christopher comments re the errors in the Wikipedia Federer entry

>(Is it fair to extrapolate from this? Other comments suggest it is.)>

"I'd say no. One of the characteristics of an encyclopedia written and edited by many hands is that you can't extrapolate bad (or indeed good) quality by inference from one entry to another entry, just because they happen to both be in Wikipedia (or I expect in 'Knol') any more that you can extrapolate quality from one website to another just because they are both 'on the web'. There is no overarching authority responsible for both which is both the strength and the weakness of the approach. All of which is to say that the trick is to learn how to evaluate multiple authorities before drawing conclusions (but I guess we've been doing that since the year dot). For content providers the trick is to be respected as authorities and find ways of drawing attentiaon to that expertise."

To which I say,

My point wasn't in relation to quality but accuracy. I think you expect at least the objective facts to be correct, and endure quality variations in the authors' opinions. When Encyclopaedia Britannica said something was 23, it usually reliably was, whereas this is demonstrably not so with at least some of the Wikipedia entries. Having said that, I believe a comparison between the two found Encyclopaedia Britannica not that reliable either, so I should concede the point that in general you can't trust such multi-author collections further than you can throw them.

A different but related problem is the fact that Wiki changes all the time - it might be accurate today but not tomorrow, or vice-versa. People keep playing with it, sometimes as some kind of counter-cultural activity, other times for darker reasons. A friend of mine is having to repost every day an entry regarding the German bureaucratic entity for which she works, since it keeps being deleted every day, I'd guess by some disaffected former staff member.

It is important that the limitations of any such enterprise be clearly and dispassionately stated.

Knol and Wiki 1

Google has announced the knol (http://googleblog.blogspot.com/2007/12/encouraging-people-to-contribute.html)

It looks like the knol (I guess, a unit of knowledge, although then it really ought to be a "know") is going to be another sez-who? type of encyclopaedia.

To raise a few concerns: Wikipedia has already been tarred for its basic inaccuracies. To take a trivial, but to me suggestive, example - since it is surely not contentious - the Federer entry provides a few annoyingly wrong tennis statistics. (Is it fair to extrapolate from this? Other comments suggest it is.) Never mind the plagiarism (Wikipedia's main use in the classrooms), what about the wholesale propagation of error? Those who forget their past are doomed to repeat it differently!

Worse are the numerous examples of Orwellian reality manipulation, particularly (but not only) by the great and the good. Can we really leave the truth to whoever feels strongest about it? To the companies and governments with a Chief Wiki Manipulator? To the psychotic obsessives with time and/or money on their hands?

Regarding the advertising-driven model interesting, all advertising is not bad, per se. Just look at non-state-supported TV, where the guy who buys the soap powder or BMW pays for me to watch high-value copyright content for free. Maybe the knol model offers something useful, as long as the advertising doesn't distort the message, like paid-for-by-pharmaceutical-company "scientific" journals.

Has a mechanism to edit/review/improve/disprove knol-edge been offered? It will be needed, as Wikipedia recognizes (but doesn't solve). I think this is an ethical issue and necessity.