FOP compatriots:
I shared more than a little of Joerg's frustration in his recent posting
about trying to get going in trunk development again. I have been able to
spend some quality time walking through code with my debugger, and my work
on the doc has helped some of the design concepts sink in better as well, so
I am getting more confident about being able to jump in and help for real.
Here are some issues for discussion.
Issue 1 -- Project Focus
I think I jumped into this project just as serious efforts were being made
to discourage new development on the maintenance branch. This was rightly
perceived as a dilution of resources. It has taken me nearly six months to
see another dilution of resources -- new development on the trunk! The only
thing that the redesign is really supposed to touch is the layout. So my
thoughts about working on fonts are misplaced. Fonts and Avalonization might
need to be done before 1.0, but AFAIK, they don't need to be done before
reunification. This difference is so important to this project, that I offer
it as a proposition to be voted upon:
Resolved: Rather than focusing on a 1.0 release, our efforts should be
focused on a 0.21 release from the trunk that is as good as 0.20.5. New
features, whether in the maintenance branch or the trunk, whether for the
user's benefits or for our's, should be abandoned unless they have an
immediate payback toward a 0.21 release. Only after stabilizing 0.21 so that
our users can safely move to it should we start adding features. This
implies that, whenever possible, we should be working on trunk layout code.
Issue 2 -- Unstable Code
We know that we are working with unstable code, which presents some risks.
First, with stable code, if a hacker like me messes it up, it will be
obvious almost immediately, and can easily be rolled back. That won't be
true in our situation. Second, unstable implies that some of the code might
be working well or finished. Most of us don't understand what is going on
the layout code as well as Keiron. I've spent a little time in there, and I
frankly don't have a clear idea of what works and what doesn't, what is
stable & what is not. So I really have two choices: 1) wade in and start
hacking away, perhaps at cross-purposes of the big picture, or 2) have a
"manager" pass a task off to me that is reasonably well-defined, and that
fits well into the big picture, then, when that is done, give another, etc.
I can do either one, but the second one makes more sense to me. I would
naturally look to Keiron for this, if he is willing. I am not looking for
something like "get line justification working", but more like "take class
Xyz.java and make method abc() return the correct word-spacing to justify
the input line -- here is a test file."
The good news is that it is probable that after a couple of cycles of this,
I would be able to find the next task on my own.
I guess what I am saying is that I am willing to follow if someone is
willing to lead.
Victor Mote (mailto:vic@xxxxxxxxxxx)
2025 Eddington Way
Colorado Springs, Colorado 80916
Voice +1 (719) 622-0650
Fax +1 (720) 293-0044
|
