Bennett Haselton wrote
The disadvantage of making people register under a presumed real name, much less following up and trying to verify their identity and credentials, are of course that you will have fewer users that way. On the other hand, the advantage is that you can give articles the stamp of reliability if it's been signed off on by, say, a professor whose .edu address has been verified.
Could you get the best of both worlds by (a) allowing unverified users to build up the meat of an article, but then (b) verifying the credentials of certain users (Citizendium calls them "editors"), and having those users sign off on the contents of a given article once it's reached a stable state? (And then future edits to that article have to go through them?)
One can. One could set up one's own wiki and import any WP articles that one wanted to 'bless' into it.
The WP way has been to be free-wheeling and pick up the pieces afterwards. That's obvious. Something less obvious to most people, apparently, is that _discussing WP article by article_ misses the basic point that we write hypertext, not self-enclosed essays. You don't really 'improve' a knowledge network by stabilising it. I hear almost nothing from this angle.
Charles
----------------------------------------- Email sent from www.virginmedia.com/email Virus-checked using McAfee(R) Software and scanned for spam
charles.r.matthews@ntlworld.com wrote:
Bennett Haselton wrote
The disadvantage of making people register under a presumed real name, much less following up and trying to verify their identity and credentials, are of course that you will have fewer users that way. On the other hand, the advantage is that you can give articles the stamp of reliability if it's been signed off on by, say, a professor whose .edu address has been verified.
Could you get the best of both worlds by (a) allowing unverified users to build up the meat of an article, but then (b) verifying the credentials of certain users (Citizendium calls them "editors"), and having those users sign off on the contents of a given article once it's reached a stable state? (And then future edits to that article have to go through them?)
One can. One could set up one's own wiki and import any WP articles that one wanted to 'bless' into it.
The WP way has been to be free-wheeling and pick up the pieces afterwards. That's obvious. Something less obvious to most people, apparently, is that _discussing WP article by article_ misses the basic point that we write hypertext, not self-enclosed essays. You don't really 'improve' a knowledge network by stabilising it.
This is an important point that is difficult for anyone married to a hierarchical knowledge system to understand. Hierarchy requires stability to function. The "stamp of reliability" is an artifact of a hierarchical saystem. It ensures the survival of received wisdom, and leaves little room for alternatives.
I've tended to look at wikis as a fractal element of a paradigm shift.
Fractal systems incorporate a lot of randomness. Tiny new structures keep popping up all over the place. Most of them, like so many of our non-notable articles are not viable. Left alone they are eventually resorbed into the growing system. Very few become major growth points. We cannot effectively apply a hierarchical system to predict which will grow. Order is statistically imposed. Radioactive half-lives operate statistically in that we know that within a given time (about) one-half of the atoms will have decayed, but we cannot predict which ones. Is there any basis for thinking that it would be any different in a knowledge network? Belief that a hierarchy can impose its will on a statistical reality is tantamount to belief in a form of interlligent design where the hierarchy is substituted for God.
Major paradigm shifts are extremely rare. The present situation which has put into everyone's hands to means of bilateral communications is what puts us there now. The old hierarchies worked because the abilities of the hieresiarchs and the masses to communicate were grossly unequal. As Thomas Kuhn said: "There are losses as well as gains in scientific revolutions, and scientists tend to be peculiarly blind to the former." With the widening of the base of uploaders in a collaborative knowledge system the loss of respect for established experts is bound to suffer. The experts will just have to learn to live with that cruel fact. The reliability upon which those experts relied may not be as reliable as they imagined.
I do not believe that reliability should be or even can be one of our goals. I also believe that teachers and professors are perfectly justified when they criticize their students for using Wikipedia as a reference. I would also hope that Wikipedia becomes so reliable as to allow them to do otherwise. This does not mean that we should stop striving for accuracy; it means that if we succeed too well we will have emasculated that drive from our readers. We want readers who are empowered to question not only what they see in Wikipedia, but what they encounter in real life.
Ec
On 3/20/07, charles.r.matthews@ntlworld.com charles.r.matthews@ntlworld.com wrote:
The WP way has been to be free-wheeling and pick up the pieces afterwards. That's obvious. Something less obvious to most people, apparently, is that _discussing WP article by article_ misses the basic point that we write hypertext, not self-enclosed essays. You don't really 'improve' a knowledge network by stabilising it. I hear almost nothing from this angle.
The value to end users of a knowledge network is a combination of the value of each piece on a standalone basis, and of the network interlink count and quality and quality of the linked information. Article quality shows up in both terms.
Different users will have different use patterns - some will come in via Google or another external search, or a very direct Wikipedia search, to a single article, and then be done or leave. Others will spend all day hopping around wikilinks expanding their horizons.
We mean too many things by quality. Among other things, we don't really have a quantatative description for the overall value of Wikipedia, to be able to look at usage patterns and say "Oh, we need more links over here" or "This section of Wikipedia is mostly accessed one article at a time, we need to encourage article quality growth on a standalone basis".
I think that we have widespread agreement on long-term direction, but we're essentially random-walking all the individual components vaguely in that direction.
Based on articles I pay attention to, I would have to say that vandalism and spamming are more of a problem than "bad edits". I see a very few good intentioned edits that I want to edit back in a weeks' time. I see several hundred vandalisms. Stable versions would clearly help with vandalism problems. It would remove incentive for people to do it, and would make managing that which did happen much easier. I am guessing that some form of vandalism fighting takes up between a quarter and a half of all senior user time, including admins and arbitrators and staff. How much worth do we get from getting all that time back for other useful pursuits within the project?