[Wikipedia-l] Re: Extension syntax vote begins

Thu Apr 15 06:39:25 UTC 2004

Brion-
> Please reconcile the above with "The point of these polls is to get a
> reasonably representative *sample* of Wikimedia opinions."

Please reconcile the fact that only 50% of the American population voted  
for either George Bush or Al Gore with the numbnut that is currently in  
office.

Every vote only draws from interested individuals. In our case, that will  
hopefully make for better results, as the people not interested in highly  
technical and insubstantial matters are also the ones who are unlikely to  
have an informed opinion. Furthermore, a less significant vote needs a  
lower minimum turnout. We don't need to ask 1000 people when we get a  
reasonably clear picture from asking 100 people. What reasonably clear is  
depends on the context.

For the comparatively insignificant matter of whether we should type  
<math>2 + 2 = 4</math> or, say, >>>>(XXXX>>>|2+2=4|<<<XXXX)<<<<, it is  
arguable that we already wasted far too much time getting an unreasonably  
*large* sample of community opinion, especially with one option clearly   
dominating all others (20 votes for my proposal, 1 vote for the 2nd place  
proposal, all others negative). The only reason I started a poll at all is  
that people kept bringing up all kinds of wacky ideas and we needed to  
come to a solution.

Who decides what is an important matter that deserves a Wikimedia-wide  
poll with announcements on the Main Pages and a personal message on each  
user talk page, and what can be done simply by announcing it to the  
mailing list and Wikimedia News, or maybe just a talk page? Ultimately,  
whoever sets up the poll, but also those who watch them do so. If you had  
felt the need to bring this to the attention of a larger group of people,  
you had ample opportunity to do so.

The difference between such a poll and a straw poll is that a straw poll  
is entirely unannounced, has no deadline, no prior discussion phase, no  
defined outcome and a very low, if any, minimum turnout. Straw polls  
answer the question: Are we anywhere close to consensus? They do nothing  
else. The Extension Syntax vote was *not* a straw poll by any reasonable  
definition.

Of course, if people just listened to me we wouldn't need any stupid  
polls.

Regards,

Erik