Pierre van Rooden <[EMAIL PROTECTED]> wrote:
> Michiel Meeuwissen wrote:
> >So, then I don't know it any more. There seems to arise a complete new rule
> >here. I think any improvement of implementation should be acceptable 
> >always,
> >in a hack or in other code as well.
> 
> We need to avoid situations where a Hack is proposed, and the three 
> weeks of work are spent on getting it in line with everyone's wishes.
> The solution is very simple:
> Post the code, asking for comments. If comments are forthcoming (if 
> teher are any, they will in a few days), change teh implementation is 
> needed.

Where excactly do we spare time then? If we spend three weeks of work (which
seems exaggerated), why would it cost less in this way?

> Finally ask for a vote.
> This is not a 'project'. A project is approved by the MMC and has a 
> projectleader and maintainer. It deals with a concept and not one piece 
> of code.

A hack involving new code should principally also have a maintainer, because
any code needs maintaining.

> MMBase is OS and as such there needs to be some consensus regarding new 
> code. Discussing code beforehand gives people the chance on input 
> without confusing the heck out of people during the vote.

I guess that at least half of the voters only very superficielly regards the
code, if at all, which is quite sensible, because code is for a large part
also a matter of taste.  

Michiel


-- 
Michiel Meeuwissen       |
Mediapark C101 Hilversum | 
+31 (0)35 6772979        |  I hate computers
nl_NL eo_XX en_US        |
mihxil'                  |
 [] ()                   |

Reply via email to