Cyberspace Law and Policy Centre, University of New South Wales
Unlocking IP  |  About this blog  |  Contact us  |  Disclaimer  |  Copyright & licencsing  |  Privacy

Wednesday, November 08, 2006


Citizendium v. Wikipedia

[This is a guest post, written by Roger Clarke. -- Ben]

Wikipedia is the most prominent of the new-age collaborative information sources. But even its champions acknowledge that there are challenges, and choices to be made.

Larry Sanger, one of Wikipedia's co-founders, has long been dissatisfied with some aspects of its management. He announced on 17 October 2006 his intention to spawn a fork, or republished version, of Wikipedia that is intended to progressively develop higher-quality, more reliable articles.

Sanger envisages the core difference about Citizendium as being a set of editors, with interleaved scope, who will take responsibility for approving articles and amendments to articles. There will be rules that are rather less loose than Wikipedia's (e.g. contributors must declare their 'real names' - whatever that means), 'constables' who will enforce the rules, and a process for appointing and controlling editors and constables. Sanger intends that the appointment process will have collaborative features, but the proposal at this stage is sketchy.

The essence of the debate is whether and how to quality-assure the content of collaborative information sources. The orthodoxy within the open movement is the 'many eyes' principle: errors will come to attention and be addressed, because of the sheer volume of people who are looking and who are empowered to do something about them. The risk of pollution is high, and anarchy looms; but believers say it can be avoided.

Some people are nervous about pollution and anarchy, and uncomfortable with constructive looseness. They prefer layers of controls, and trust in a few rather than trust in the 'great unwashed hordes'. They point to the increasing incidence of Wikipedia pages being frozen for short periods, to let tempers cool. (As this was being written, the Wikipedia entry for 'Wikipedia' was locked, with the explanation "Because of recent vandalism or other disruption, editing of this article by unregistered or newly registered users is currently disabled. Such users may discuss changes, request unprotection, or create an account.").

The distinctions between the two approaches might be seen this way:

QA Principle'Many eyes''A few good men'
QA StyleOpen collaboration among manyAn inner clique of guardians, possibly self-perpetuating
QA ProcessInformal review, by genuine 'peers' as in 'equals'Formal review, by an approved set of 'peers' as in 'peers of the realm'?
Editorial StyleSelf-organising and/or AnarchicHierarchical command and control, but with a collaborative appointment process?

There are many aspects of Citizendium that cast doubt on its ability to survive any longer than its predecessor Nupedia, let alone thrive. Will the elite few prove to be as energetic as the egalitarian hordes? Will the bureaucracy of editorial committees cause even the first few score pages to miss their window of opportunity? Will any of the pages ever score high enough on Google rankings to be noticed? Will the quality difference matter to people, or will the 'good enough' of Wikipedia trump the new approach, just as Microsoft's Encarta, by using some of Funk & Wagnall's middle-brow encyclopaedia, trumped Britannica? Will the inevitable re-branding as something trendier like 'Zendi' be enough to revive interest?

Ultimately, the community will vote with its feet, or consumers will determine what the market wants by paying with their clicks and eyeballs (choose your preferred metaphor). Perhaps the venture's greatest contribution will be to help us learn about quality assurance of open content.

[This was a guest post, written by Roger Clarke. It is available from Roger Clarke's website under either an AEShareNet Free for Education licence or a Creative Commons 'Some Rights Reserved' licence. -- Ben]

Labels: , ,

Anonymous Larry Sanger said:
The table comparing Wikipedia with the Citizendium belies any rudimentary understanding of what's going on with the latter.

Wikipedia is said to rely on 'Many eyes' while Citizendium has 'A few good men'. Utter rubbish. Like Wikipedia, the Citizendium will be an open wiki, with editing and commenting open to many non-experts. It will have "many eyes." It's just that many more of the eyes will be expert eyes.

And, as unhip as it might be to say so, that is, of course, a difference that makes a difference.

The "QA Style" comparison is also not only incorrect, it's insulting--denying that CZ features "open collaboration among many." Both WP and CZ will feature "open collaboration among many" in order to assure quality. It's just that CZ will have, in addition, individual editors (or groups of editors) who approve particular versions of articles. Why that should imply an "inner clique of guardians, possible self-perpetuating," represents merely unjustified prejudice on the part of the blogger here. The statement is not made with anything like evidence.

Finally, the "Editorial Style" line of Wikipedia is said to be "Self-organising and/or Anarchic" while CZ's is "Hierarchical command and control, but with a collaborative appointment process?" This is completely misleading. Since CZ is a wiki, very many editorial decisions will be made not by editors but by rank-and-file authors. Calling the wiki-based after-the-fact oversight provided by editors "command and control" shows ignorance either of the functions of CZ editors, or of the meaning of the phrase "command and control."

In short, Roger Clarke seems to have the Citizendium confused with, say, the Britannica, or some other traditional, top-down, non-collaborative encyclopedia. He seems not to have spent any time on the Citizendium website before writing about it. In fact, he evinces no understanding that the project is, like Wikipedia, a wiki. He could have found that out, however, by clicking through to the pilot project wiki.

I view Clarke's knee-jerk reaction to our endeavor as a perfect nutshell example of how people are incapable of thinking outside of a false dilemma: either you believe in radical egalitarianism, strong collaboration, and bottom-up production, or you believe in elitism, individual authorship, and top-down "command and control." There's nothing in between and no intermixing of the categories. The idea of a bottom-up peer-to-peer production that makes a special role for experts sounds self-contradictory. Thus, so many people can't even conceive that Web 2.0 might make room for experts.

Well, five years ago, no one was able to conceive how an open wiki could produce an encyclopedia, either. Sometimes, the only way to persuade people is to show them.
Anonymous Roger Clarke said:
In reply ...

When I wrote the original post, my sense of the CZ project wasn't what Larry describes.

Larry, I think you need to put together a clearer rendition, in order to stop me and many other commentators mis-interpreting your intention.

I now interpret you to be saying that, contrary to the impressions many people have:
(1) CZ is to be an open wiki
(2) all articles will *initially* be open to any contributor (under Wikipedia rules? or under different Citizendium rules?)
(3) some (progressively all?) articles will be edited by editor(s)
(4) their status will then be changed to 'approved'
(5) thereafter contributions to that page will be moderated, also by
that/those editor(s)

- will approval apply to an article, or to a version of an article?
- if whole articles are approved, then the 'wiki-ness' will progressively decline
- if, as your post suggests, a *version* is approved, then the living, open-access article continues to be available

But in practice how does that work, e.g.
- the approved article (on, say, 'Montypython') takes precedence
- but prominent (or not) text in the header (or footer) points to the live, open-for-amendment one?
- vice versa?

The value-add of editorial teams is clear, as is the downside of editorial teams. I don't see any reason to resile from what I said in the first place about that.

But, if Citizendium is meant to attract 'many eyes' across from Wikipedia, that has consequences - somewhat akin to Montypython's warring factions (the Judaean Liberation Front vs. the Liberation Front of Judaea, if I remember correctly).

If I'm now reading Larry's posting better than I originally read his article and the reports on it, then ...

CZ seems to be intended as a direct competitor to Wikipedia:
(a) for the attention of readers (and of intermediaries such as search-engines), but also quite critically
(b) for the key resource - the 'many eyes'.
(The terms 'competition' and 'resources' imply an economic rather than a social analysis, but that's what's called for here).

Worse, the forking of the complete collection divides the effort. Public enhancements to an open CZ article wouldn't automatically become enhancements to the corresponding Wikipedia article; and the reverse also holds.

Isn't a more elegant design available than this?

Can't CZ be conceived as 'the home of quality-assured Wikipedia pages', by forking individual pages only, and only once the editor has done the value-add?

The relevant, living Wikipedia page would then remain, at all times, on the Wikipedia site, but enhanced with text stating something like 'A quality-assured, moderated CZ version of this page is available at *link*'.
Anonymous Jonny Cache said:
A discussion of Clarke's post and Sanger's response is under way at The Wikipedia Review.
Post a Comment

Links to this post:

Create a Link

<< Home

This page is powered by Blogger. Isn't yours?