trinity-devel@lists.pearsoncomputing.net

Message: previous - next
Month: February 2014

Re: [trinity-devel] R: [trinity-devel] TDE kftpgrabber and kasablanca FTP clients - connect, but no file list ever retrieved

From: "David C. Rankin" <drankinatty@...>
Date: Wed, 19 Feb 2014 15:08:52 -0600
On 02/19/2014 11:58 AM, Darrell Anderson wrote:
> Ready? Seems all I have done the past few weeks is file bug 
> reports. Regressions, sloppy coding, stdout/stderr problems, build 
> issues. We have 66 build issue bugs filed, 48 stdout/stderr bugs, 
> and 18 regressions.
> 
> As I keep saying, we get only one chance to get R14 right.
> 
> Darrell

Well, we will ultimately find a balance. Personally, I agree with you 100%. I
also recall having this exact same conversation in February 2011 and again in
April 2012,, etc... Then it was wait till Serghei finishes with cmake, then it
was gcc4.5, then 4.6, then 4.7, then it was port to Qt4, then it was 3.5.13-sru,
then it was don't port to Qt4, update Qt3/TQt3, then it was TDEHWLIB, then it
was..... The point being, there will always be things that will need to be fixed
or improved, the *Question* is when is it good enough for an outstanding R14
release?

As I said, except for the critical/blocker bugs, I think we are getting close to
having TDE in shape where we can feel good about a RC1 freeze. Take the ftp
apps, the other smaller ones, I don't think those warrant manpower for R14RC1. I
think we come to a consensus on what bugs we want to see fixed so that we can
all say "R14 was a darn good release!", fix those bugs, and then identify those
that are not working in the "Release Notes" and slate fixes for the minor apps
for 14.0.1.

That makes the critical question, "what of the *current bugs* that are
critical/blocker do we consider mandatory to fix before RC1 freeze? I think we
take that list, focus our manpower on fixing those, (and if another
critical/blocker arises -- add it to the fix list), get them fixed, draw a line
in the sand and ask "Everone OK with this being RC1?, Any other gotta haves?",
then plan on a 2/3 majority "yes" target (knowing we will never make everyone
100% happy - me included), freeze RC1 and footnote the rest in Release Notes.

That's how I see this having to play out in order to get to RC1 -- otherwise you
and I will be having this conversation again in 2015 :)

-- 
David C. Rankin, J.D.,P.E.