On Sat, May 28, 2005 at 10:03:17AM -1000, Warren Togami wrote:
> A good argument against this would be if it takes far too much work.

With 3.0, there are 3 mass-check runs (~1 week each to get participation and
such), then the score generation takes ~1 day total.  So in short, it's
roughly a month with testing to get new scores done.  In theory, with 3.1
that'll be closer to 10-14 days.

> It is true that I don't know how much work it would entail.  If that is the 
> case, then I would propose scheduling a 3.0.5 with more backported fixes 
> and rescoring for a few months after 3.1.0.  That would give me time to 
> learn how things work here so I can better contribute time and 
> especially testing resources.

I think that once 3.1 comes out, it'll be much easier to keep a stable older
version going than with 3.0, due to the availability of sa-update.  Of course,
I don't think we have enough time to actually do a lot with that for a while.
:(

-- 
Randomly Generated Tagline:
"I try to avoid doing the Linux vs. Microsoft thing . . . it's not as
 if I would be unbiased anyway. Asking me about Linux is like asking the
 Pope about the existence of God."  - Linus Torvalds

Attachment: pgpCs6mttVkxR.pgp
Description: PGP signature

Reply via email to