-----Original Message----- From: wikitech-l-bounces@Wikipedia.org [mailto:wikitech-l- bounces@Wikipedia.org] On Behalf Of Erik Moeller Sent: Thursday, December 11, 2003 5:26 PM To: wikitech-l@wikipedia.org Subject: Re: [Wikitech-l] Categories simplified
Jakob-
Ah - that?s the fault! The link [[Category:Stuff]] should refer to
the
article "Stuff" and not to some special-category-only article. [[Category:Biology]] *is* [[Biology]]! Please have a look at the
actual
article [[en:Biology]]: What does ==Fields of study in biology==
contain
but a listing of subtopics? Why introduce a new artificial namespace
when
you can better name the article for instance "Fields of study in
biology"?
Because this will standardize these lists and automate them at the
same
time. Currently we have "List of xy topics", with lots of Wikipedia- related meta information, floating around in the regular namespace.
These
are not really articles and they should not be counted as such. Having [[Category:Mathematics]], [[Category:Biology]] etc. is much more
economic
and only shows the information to those who care about it.
Possible problem: Does "Related changes" still work on category pages?
We
might need to modify that.
Why do we need [[Category:Mathematics]] and [[Category:Biology]] when we already have [[List of xy topics]]?
Why can't we just add software to recognize that "List of" articles are categorization articles, and use that to automatically categorize things?
E.g. anyone listed on [[List of actors]] we know is an actor. Anyone listed on [[List of American actors]] is an American actor. Anything listed on [[List of mathematical topics]] is a mathematical topic.
Somewhere on the page we could have "This entry appears in the following lists:"
And we could also have a function
"Add this entry to a list" which would pull up the names of lists and could slap the entry into the list.
I guess I'm on the side of Jakob here that instead of forcing people to learn new behaviors and adding complexity to their tasks--something that's acceptable if the only users of a system are to be experts, but not acceptable if you're designing for general use--we should try to make the software smarter and make only minor adjustments to people's behavior.
--tc