Re: [mizar] Mizar Twiki

[Josef Urban <urban@ktilinux.ms.mff.cuni.cz>]

Dear prof. Rudnicki,

I hoped to avoid this "meta-discussion" by giving pointers to the 
appropriate twiki pages, where e.g. the general topic "TWiki vs. 
Mailing-lists" is sometimes discussed, and by mentioning some 
Mizar-specific reasons explicitly in the MizarTwiki topic. Most of all, I 
really think the border between twiki and mailing-lists is a matter of 
experiments and adjustments, and as noted in the presentation at 
http://twiki.org/p/pub/Main/TWikiPlatformPresentation/Slide40.GIF, there 
is a place for both.

I agree very much with you, that if mizar-forum worked well, the need for 
things like twiki would be much smaller. However, people generally refuse 
to post "stupid" questions to mizar-forum, although we all know that there 
are _many_ questions (sometimes "showstoppers", if you do not know the 
answer), when you want to learn Mizar and work with it. 

To repeat here some of the general arguments, I think the twiki is much 
better suited for things like incremental creation of documentation - e.g. 
when I find a good example of the 103 error, I will just copy&paste it to 
the appropriate twiki place, while I would probably decide against posting 
it to mizar-forum. There is a lot of outdated documentation on Mizar - if 
such documentation was on twiki, anyone could update it as early as he 
detects some outdated part. Generally, any "closed" discussion can be 
"refactored" later, to contain just clear information.

Anyway, I just tried to make the documentation process as easy for the 
users as possible. If there is no will to document Mizar, no tools will 
help, be it twiki or mail-lists. It seems to me, that Mizar is coming to 
the point, where the choice will have to be made, whether it will become a 
not-completely-user-unfriendly standard system for working mathematicians 
who want to formalize their results, or continue to be a secret knowledge 
obtained from great masters only in the oral form ;-).

Regards,
Josef