On Wed, Apr 10, 2002 at 12:57:40PM -0400, Alan DeKok wrote:
> Chad Miller <[EMAIL PROTECTED]> wrote:
> > Hi, all.  I've decided to withdraw the radiusd-freeradius* packages from 
> > Debian Woody (the upcoming release).  If history is any indicator, any 
> > snapshot of the tree I take will need a signifigant patch not long after 
> > it's taken.  Debian's standards are too high, and administrative software
> > (and authentication in particular) is too important for me to allow that.

>   Is it possible to verify that a particular version is OK for the
> release?  If so, it should be possible for us to issue a point
> release, and have the Debian release use that.

Though I wish it was possible to include freeradius in woody, I do
understand where Chad is coming from.  How are we to recognize a
particular version as being 'least buggy' ahead of time?  If all the
snapshots have bugs in them of one kind or another, how is any snapshot
going to get enough testing to be confident there aren't any major,
ugly, yet-unseen bugs hiding in the code that will pop out only after a
version of freeradius has been burned to hundreds of thousands of CDs
around the world?

I'd like to see freeradius in woody, but not to the point that I'm 
volunteering to do the heavy lifting on Chad's behalf if bugs show up
after release.

Steve Langasek
postmodern programmer

Attachment: msg04836/pgp00000.pgp
Description: PGP signature

Reply via email to