[wtp-dev] 1.0 Planning and Beyond

[David M Williams <david_williams@xxxxxxxxxx>]

I'd like committers and component leads
to consider the following while finalizing your 1.0 plans. 
At this week's dev. status meeting we
can discuss if there's comments or concerns. 

First, a reminder, that for 1.0 (which
will be on 3.1 Eclipse base), I'd advise you to 
be as conservative about what you can
deliver, without any chance 
of any big "last minute" efforts,
rushes and instability. 
Beyond all the usual reasons we should
be doing that anyway!  
we now have an additional reason: At
some point, 
during 1.0 development (near M9) we
will need to also begin
development for 1.5, on Eclipse 3.2
stream!

Second, the reason for the "dual
stream" effort above is because as a PMC we have agreed to 
"join the Eclipse Train" and
have a more synchronized release next year with several other 
major Eclipse projects... the base platform,
GEF, EMF, TPTP, and DTP .  And part of this 
"synchronized release" process
is "synchronized milestones". While in theory there might be

ways of doing this with different parts
of the train having milestones of different durations, 
in practice, it will be * much * easier
if we in WTP also plan to have (roughly) 6 week 
milestone periods like the base Eclipse
teams do.  This will take some adjustment in our 
thinking since I know we are all still
prone to "over promise" in our planning, so this will be all
the more 
reason to be conservative in our plans.
(In general, a 6 week milestone means 4 weeks 
of development, 1 week of planning and
1 week of milestone level testing). 

While the exact milestone dates and
durations are being worked out by Tim Wagner 
(to be coordinated with the planners
of other projects). I'm expecting ours to be around 
6 weeks (some might be 6.5, some 7?)
but wanted to send this note out now to see 
if anyone had concerns with this general
shorting of our milestones. 

I personally thinks its achievable and
desireable, but didn't want to speak for everyone, 
in case there were issues I was overlooking.


Thanks,