On Tue, Jul 29, 2003 at 11:34:48PM +0800, b schewek wrote:

...

...

3. The synthesis of the first two concerns: what happens when

there are errors in the stable version? Mistakes can be corrected in the current version, but only sysops will be able to change which version is listed on the locked Brilliant Prose page.

Sifter would have the exact same problem. IMHO the best solution is to give sysops some leeway to change the pointer to a new revision in case of obvious corrections (linkfixes, spelling, uncontroversial factual changes as per talk etc.), and to otherwise require a re-listing on the candidates page.

What about a 'web-of-trust', whereby /all/ registered users can 'approve' or 'reject' a version of an article, and every registered user can select any number of 'trusted registered users', and thereby articles appear 'approved', 'rejected' or 'undetermined', according to the settings of 'trusted users'.

Then it is up to every individual to select the 'sifting agency', so to speak.

Approval is not transitive (I can think that X is sane, but maybe he's opinion about Y being sane is different from mine) and is field-specific (I can trust X about chemistry but not at all about anything related to, say, history).

Note* People, perhaps it would be a good idea to change the subject header to something intelligible when responding to/from a bulk-digest email. Thanks. -----

Schewek What about a 'web-of-trust'?

I kinda like the general idea of this, but have rarely seen it implemented to its full potential. Either its black and white or too superficial to mean anything.

This approach would probably best manifest itself as a "temperature rating" for articles that is set by users as they read (not edit) articles. This rating can then be used as a variable in a search/tour so that users can get to what they want quicker. The ratings could be sliders set for various factors, most prominently -- "quality, usefulness, completeness, favorite" etc. (Now it becomes clear how *subjective these determinations are)

And since "perfection" is only the permanent and unacheivable goal -- the God infinity which we use as a reference point in our calculations and factorizations -- we can stipulate (in our thinking) that this "perfection" is the common goal -- and that true perfection is not to be achieved by being classist about the approach. The more open the better, and the is the more beautiful approach is the one that is most minimal and least obstructive. Wiki101.

-S-

__________________________________ Do you Yahoo!? Yahoo! SiteBuilder - Free, easy-to-use web site design software http://sitebuilder.yahoo.com