From: Beman Dawes (bdawes_at_[hidden])
Date: 2004-01-26 16:58:43
At 04:00 PM 1/26/2004, Robert Ramey wrote:
>Dave Abrahams wrote:
>> I'm actually shocked that we're wedging
>>is_abstract into the release branch.
>Doesn't this kinkd explain why three months can
>elapse between the time a new release is anounced
>and the time it is available.?
>Can't we just say that when a release is anounced any
>new features will have to be added to the next release?
>I believe that the efforts to get features in sooner have
>actually resulted in making the new features available later.
>I also believe that this is common occurance on lots of
Our problem is only partially new features. We also have trouble with:
* Regressions against the prior release, particularly with older compilers.
* Documentation not updated at the time code was originally changed.
* Patches and bug reports not attended to until close to release.
Boost list run by bdawes at acm.org, gregod at cs.rpi.edu, cpdaniel at pacbell.net, john at johnmaddock.co.uk