From: Paul Richards <p.richards@elsevier.co.uk>
Message-Id: <199605081122.MAA06028@cadair.elsevier.co.uk>
Subject: Re: require-client functionality
To: new-httpd@hyperreal.com
Date: Wed, 8 May 1996 12:22:48 +0100 (BST)
In-Reply-To: <199605071557.AA090344652@ooo.lanl.gov> from "Rob Hartill" at May
 7, 96 09:57:32 am
Content-Type: text
Content-Length: 2277      
Sender: owner-new-httpd@apache.org
Precedence: bulk
Reply-To: new-httpd@hyperreal.com

In reply to Rob Hartill who said
> 
> > This needs to get sorted out pretty sharpish. The last release was an
> > embarrasment since there were numerous serious bugs in it that had to be
> > promptly fixed.
> 
> Out of curiosity, which release had major bugs that were an embarassment?.
> I thought 1.0.[35] were pretty good.

Umm, well, 1.0.[012]

1.0.0 was a total disaster and bug reports starting arriving within days and
lots of them too.

> Your concerns are perfectly reasonable but you need to consider the
> history of Apache a bit more. The original purpose of Apache was to
> get a httpd that did what we (the developers) wanted it to as well as
> the stuff the end users want.

Yes, I'm well aware of this, my point was that this has changed, it's now
a major internet project. It depends what you want Apache to be, if it's
still just a forum for the developers here to hack their favourite server
then fine, but if you want it to remain the no 1 server in use in the world
then you have to consider your user base and put the project on a more
professional footing (from an organisational viewpoint not a commercial one).

> It'd be a shame to completely loose what remains of the original spirit
> of the project. Keep the rules and protocols, but lets also keep some
> slack which is ultimately kept in check by common sense and peer review.

There are virtually no rules or protocols in this group which isn't a bad
thing. I'm not suggesting any of that change but it's simple good
programming practice to do proper release engineering and that bit isn't
as much fun as simply hacking code. As I said above, if this is just a forum
for a group of hackers to play with their favourite server then release
engineering is irrelevant but I'd make it clear to people that's all it
is because when new versions of code come out of projects like this (as it's
perceived from outside) then they have an expectation that it's been well
tested, otherwise why bother, you might as well not do "releases" and
just grab the snapshots as and when you feel like it.

-- 
  Paul Richards. Originative Solutions Ltd.  (Netcraft Ltd. contractor)
  Elsevier Science TIS online journal project.
  Email: p.richards@elsevier.co.uk
  Phone: 0370 462071 (Mobile), +44 (0)1865 843155