I'm trying to recapitulate on what was said. Please correct me if I got
something wrong.
cdebconf
Joey Hess approved unblocking cdebconf 0.101. Then, Colin Watson noticed
this thread after uploading 0.102 and said he was sorry. I discussed
this with him on IRC:
<cheal> [...] first, what are the implications of the cdebconf 0.102
upload? will it still be possible to unfreeze it soon?
<Kamion> I guess it will take another couple of days, unfortunately
<Kamion> unless it was already unfrozen in the britney run last night
So, Colin only said he was sorry because 0.102 needed usual testing and
it will be possible to unblock it.
libd-i
Joey Hess:
Unblocking libdebian-installer 0.97 might cause problems due to this change:
* archdetect: Rename sb1-swarm-bn to sb1-bcm91250a for consistency.
Then Martin Michlmayr clarified (at least for me) that the problem was
risking to break a MIPS subarch getting 1 to 3 insts. on popcon for
beta2 installs.
Bastian Blank replied to Martin's mail:
We will glue the libdi transition to the newt transition.
I failed to get a clarification from him, but from an IRC discussion I
had with him the day before that post, I'm pretty sure he meant that an
upcoming libdi transition and the newt transition would be tied *if*
libdi doesn't transition to testing now.
In summary, the only problem which is expected from unblocking libd-i is
to break beta2 installs on a rare subarch. I can't see how this would be
worse than risking to delay the newt transition, but if some people
aren't convinced, what's the alternative? If it is waiting for beta3,
does that mean the upcoming libd-i transition Bastian warned about will
be tied? If so, what should this libd-i transition look like?
--
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
