On 06/06/2007, at 2:06 PM, Jason van Zyl wrote:
Is this different/related to the design paper you were working on?
As I've said before, I'm interested in collaborating on this, and
would like to see it posted somewhere.
No, he started on this prior to what I started. I have the
underlying logical mechanism. No final API.
Maybe we're getting our wires crossed here, but I thought in both
cases you were currently working on something. I can understand not
getting heavy into implementation until we have our process nailed -
but if there is anything in progress I'd like to see it.
Before I'm willing to talk about anything related to 2.1 this is my
list:
- Apply as many patches as possible (6 left to look at)
- Clean up JIRA
- Flush out the duplicates
- Close what issues have been dealt with
- Delete the issues with incomprehensible explanations
- I think if 10 of us when in there for a day we could reduce
what cruft there is and discuss how to manage the great information
that's in there which we generally don't look at
- Release 2.0.7
- Document making patches so nothing will be accepted without at
least working example project, with a preference for unit tests and
integration projects
- Make an archetype for test cases
- Make an archetype for ITs
- Document running the ITs
- Dealing with the IT issues I've listed previously
- Separate the design documents from the Taxonomy
- Taxonomy
- The start being the actual home page of the Maven space
- Align JIRA to Taxonomy
- Align Wiki to Taxonomy
Yep, this is all goodness. I've always seen JIRA, testing
requirements and the organisation of design documents as the
important things before 'getting started' on 2.1 in earnest.
I would like to champion the re-initiation of the development process
we started some time back. I think we need to have a point where we
say "from this time on, only a certain level of quality is
acceptable" - and be prepared to reject changes that don't come with
tests, etc. I think it ties in to what you're doing, so I'll get
stuck into that document again.
The only caution I'd urge is that we don't over-reach in these things
and have them fail. I'd rather draw a line, get started, and
gradually improve (the taxonomy springs to mind here). I think we're
on the right track at present.
Until these things are done we'll continue moving along slower then
we could be. I've managed to get releases out in a reasonable time
frame and it will only get better by finishing these items first
and I think we need to pull together and deal with these first
before any meaningful discussion can ensue.
Should we have these listed as tasks that can be assigned to people?
- Brett
---------------------------------------------------------------------
To unsubscribe, e-mail: [EMAIL PROTECTED]
For additional commands, e-mail: [EMAIL PROTECTED]
