What a great discussion!
To wit:
>
But I don't think Evan
is actually talking about the mechanics of how patches get in, although
maybe he is. I think he's probably talking about a more diffuse set of
topics like how does this fit in with standardization, code quality, code
style, scheduling, testing, priority assessment, managing scope creep,
funding, feature planning, avoiding micromanagement, product release
cycle, and, apparently, whether there exist uniform practices here. Ie,
all the complicated issues no true blood programmer likes to talk about.
But that's just my guess; I'm sure Evan can help clarify that.
We could also ask Evan if he's thinking more release or merely in source
control, and if a release which branch(s) of distribution, (maybe also
if he intends to transplant patches elsewhere.) There is also that
fascinating question, why he wants to know anyways (otherwise known as
"what is the right answer to give him?")
-Marcus Watts
>
Generally, what Marcus said.
I totally understand the Goodness that are code revision and branching systems.
I'm a big fan. That being said, what I was trying to get at were the more
upper-bounded questions around "how do we know something's ready and able to go
into Stable/1.4/Mainline/Whatever We're Calling It This Week," and "how do we
know something's ready and able to go in Development/1.5/Experimental/Whatever
We're Calling It This Week," and "what's the relationship between
Development/1.5... and Planned 1.6, if not everything in 1.5 is going to be in
1.6, how does that call get made?" and "if everything in 1.5 is going to be in
1.6, how does inclusion in 1.5 get decided?" Right now these answers are
somewhat unclear for folks who are not able to go to each hackathon.
I know this is a very difficult question, going to the heart of "What Is
OpenAFS?" I don't want to create a huge debate, I only want to get some
guidelines to help me (and others!) plan our work going forward.
I want to thank Derrick, Simon and Jeff for all giving me good pieces of the
answer over the past few days. Derrick pointed me to some historically
discussion on other lists, Jeff pointed me to the roadmap. Simon's discussion
was especially useful as an overarching review of the current thinking from
those making the decisions.
Going off on a small tangent, I believe all critical DAFS issues that have been
reported to OpenAFS.org have been fixed or have fixes publically available. I
know that was important, and it has been a main priority for us.
At this point, I think I know what I needed to know regarding the process. Many
thanks to everyone on list for helping me out.
Sincerely,
Evan Macbeth
