httpd-dev mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From Paul Sutton <>
Subject Re: [PATCH] CGI: script command line and win32
Date Sat, 10 Jan 1998 17:32:35 GMT
On Sat, 10 Jan 1998, Ben Laurie wrote:
> Dean Gaudet wrote:
> > One questions the sense of even voting on these things if you don't test
> > them.  What does this gain us over just committing and testing later?
> > Nothing imnsho.  I know I vote without testing, so I'm not picking on you
> > two, I'm just questioning what voting is even useful for.
> > 
> > On the other hand if you don't vote when you haven't tested then the
> > project stalls.  Whereas if it had just been committed the project
> > continues and some other person (who doesn't necessarily have time to
> > follow every single commit) gets to test the code by grabbing the latest
> > HEAD.
> Exactly. And that's why I vote without testing.

The trouble with voting without testing is that the vote may be for a
variety of reasons, for example

  -- I trust this person, so their patch'll probably work ok
  -- I like this feature and want to see it get into Apache but it'll need
     three +1's so here's mime anyway
  -- This bug should be fixed

I don't think a vote without a code test should be counted as a valid
vote. Whenever I vote on a patch it has been tested on both Unix and
(usually) Win32 and works as described. It also means I've had a look at
the code and don't think it'll impact other functionality. All of this
takes time (especially getting it onto Win32 and testing it) so my votes
tend to be later than others. Of course it would be much easier for me to
+1 the patches based on whether I think the feature is good or whether I
think the bug in question deserves to be fixed, but that isn't the way
that the review process should work.


View raw message