[WikiEN-l] How is validation supposed to work?

[stevertigo]

--- Magnus Manske <magnus.manske at web.de> wrote:
> stevertigo wrote:
> 
> >IMHO, the 'lets mine this turnip and
> >hope we find gold' approach won't work. It's a
> >Murphy's Law, or something like it. Testing to see
> >what crap floats is indeed only useful for
> Calvinball.
> >
> Yeah, just putting new toys online to see what
> happens is really a
> stupid idea.
> 
> Thankfully, we never tried such nonsense at Nupedia.
> Who knows what we
> would have ended up with, instead of the twenty-plus
> high-quality
> articles we now have after years of hard labor.

Point taken (and thats certainly my kind of sarcasm
:]) but you also make My point:
The use /discovery /application of the wiki software
to the purpose of creating an encyclopedia was a
relative fluke. You can't expect that kind of
breakthrough to happen again in the same manner to the
same general group of people in a newly developed
application toward the same idea. 

In fact the conditions are greatly different than
Nupedia, as Wikipedia was a separate experimental fork
which was started from absolute scratch. Im sure
your'e far more aware than I how Wikipedia is not a
'lets throw it up and see what it does' kind of
environment anymore. Im sure thats not the way many
other non-coders see it, who contribute largely
through the established "normal" wiki processes, and
see the value of Wikipedia in the hard work put into
it by countless people.

The AVF idea is about drastically changing the way
that Wikipedia articles are edited (ideally in a way
which conforms to and mirrors the egalitarian and
meritocratic openness of Wikipedia that has brough
success thus far). There's a philosophical base and a
history of discussion about the idea which newbies and
even oldbies need to understand, in order for it to
really work IMHO, but I understand if ATP your'e just
looking at it from the basic 'will this code work for
something' point of view.

> All of the above *might* happen. Which is why we
have a test phase. If
> the voting trolls really turn out to be a problem,
we could restrict
> viewing of the individual votes to admins (to check
for bot abuse and
> the like). Or each user could chose to hide his/her
votes individually.
>
> In the test setup, everything goes. Even anons (or
whatever they're
> called now :-) can vote. This might or might not be
a good idea in the
> long run. But there's only one way to find out.

No, all of the above can be *expected* to happen in a
'test phase' for an app which in fact does not yet
work and for whose immediate results are explained to
be meaningless. If written to solicit help, the T'Pol
wiki will probably give you pretty much the same kind
of results. All that said, Ive read Timwi's
explanation of what to do with the first batch of
cookies and Im somewhat more hopeful. Yes, baby steps
-- I understand.

Stevertigo
 


	
		
__________________________________ 
Yahoo! Mail - PC Magazine Editors' Choice 2005 
http://mail.yahoo.com