On Thu, 2002-10-10 at 06:16, Max Horn wrote:

> So while I still think Variants are somewhat cool, I wonder if they 
> are really worth the effort, or at least if they are worth the effort 
> right now. Maybe they are. Maybe not. But I think before we storm 
> ahead and somebody "just implements it", we should have a rock solid 
> design of it. That said, of course certain things can only be found 
> out by implementing and evaluating them - if somebody wants to open a 
> CVS *branch* I am fine to it, but please stay away from HEAD CVS for 
> now.

Yeah, I'm inclined to believe that the only way this is worth it *right
now* is if it's accompanied by a rewrite of the dependency loop.  That
whole section of the Fink code has obviously been growing without
bounds, accumulating warts of special-cases, it really needs to be
cleaned up.

> And I definitly see no reason to switch to an XML format. The current 
> format has proven with splitoffs that it is powerful enough to handle 
> what we need. While I don't quite agree with the syntax of the 
> example given on this thread, its general outline is what we are 
> looking for I think.

I'll second this.  I'm all for XML where it's useful (primarily
machine-generated and machine-parsed configuration or data that needs to
still be hand-editable), but for something that's mostly hand-edited by
a human, key:value is perfectly sufficient.



-------------------------------------------------------
This sf.net email is sponsored by:ThinkGeek
Welcome to geek heaven.
http://thinkgeek.com/sf
_______________________________________________
Fink-devel mailing list
[EMAIL PROTECTED]
https://lists.sourceforge.net/lists/listinfo/fink-devel

Reply via email to