httpd-dev mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From d...@ast.cam.ac.uk (David Robinson)
Subject Re: 64.auth_type
Date Tue, 19 Dec 1995 15:16:00 GMT
>> The concern that the latest *stable* (non-beta) release is functionally
>> complete is going to lead us down the path of constantly retreading
>> 1.0.1, 1.0.2, 1.0.3 for every desirable but non-critical patch that
>> comes along...
>> 
>>       [64.auth_type may be fixing *really* broken behaviour, I'm not
>>       leveling the observation at this specific patch]
>> 
>> ...it's clearly a grey area about which David and I disagree; I'm quite
>> happy to be re-educated, it's just a little confusing at the moment.
>> 
>> Ay.
>
>I think the criteria for 1.0.x patches is the same as for 0.8.x patches - they
>must be bug fixes and not features. I think every bug fix should be considered
>as a candidate for 1.0.x, and only left out if it doesn't work. Leaving them
>to drift until they bite someone badly will just cause even more 1.0.x
>versions, not less. I think the "show-stopper" only policy is misguided.

That's what I'd prefer; we (notionally) fix bugs in 1.0 as patches become
available, but a public release occurs only when a 'showstopper' bug is
fixed, or when we think we have a 'final' version. Obviously we would
optimise the process by batching up the minor patches for consideration only
just before a publically released version.

 David.

Mime
View raw message