httpd-dev mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From Harrie Hazewinkel <har...@lisanza.net>
Subject Re: [PATCH] mgmt_get_vars hook
Date Thu, 14 Jun 2001 14:53:41 GMT
Hi,

A reply like this is so disappointing.

Greg Stein wrote:
> 
> Three +1 votes are not required for commits like this. That is really only
> needed when you have "big" functionality, or you make large (design) changes
> to the system. Most everything else is commit-then-review.

WHy is this not done with some patches I sent out a while ago??

> 
> I committed the code because it looked just right to me. With it in source
> control, we can refine it if necessary. But even better: we can build on it
> now.

So why is code which looks right to me not commited?? I can tell you the
answer, "I don't have commit access". 

> 
> Discussion is never "closed" on any of the code. If you have a better way,
> then we can continue discussing. But I have to say, I believe the hook is
> the absolute right answer (it frees all endpoints from knowing about the
> others), and I think human-recognizable strings for names is appropriate. I
> also liked the set of types available for each item (integer, string, or a
> (sub) hash table)

This is what I think is freightening. Someone like you thinks it is the
absolute right answer and therefore it must be done that way.
I will see it this way, you are having commit access and I don't.
This is regardless which is the right answer to some problem.

Oh, and please do me a favour. Stop argueing about this.
It comes to the point where it is not productive, but 
unfortenate it needs to be said that there is indeed an
Apache 2.0 code police.

Harrie
-- 
address: Covalent Technologies, 645 Howard St, San Francisco, CA - 94105
phone: +1-415-536-5221                               fax:+1-415-536-5210
personal website: http://www.lisanza.net/

Mime
View raw message