Hey all,
I think it's probably time to start planning the next release, or atleast decide when it's going to be.
The GNOME folks rave about time based releases, and think they are the best way to do Open Source software. I think this mostly comes from giving everyone a timeline to get features into the repository.
But, Bryce has done a really good job of making Milestones also - and I think that if we set "Milestone 2" as when we're going to do the next release that would be reasonable too.
Thoughts? Ideas?
--Ted
On Nov 11, 2003, at 10:48 AM, Ted Gould wrote:
Hey all,
I think it's probably time to start planning the next release, or atleast decide when it's going to be.
The sooner the better :)
The GNOME folks rave about time based releases, and think they are the best way to do Open Source software. I think this mostly comes from giving everyone a timeline to get features into the repository.
I should think that overall gnome has more contributors than Inkscape. Using a time-based deadline or target has a tendency to quantify the goal too much. This can put undue pressure on everyone, as well as force a false sense of accomplishment. If we were herding kittens, then a stopwatch would probably be best.
But, Bryce has done a really good job of making Milestones also - and I think that if we set "Milestone 2" as when we're going to do the next release that would be reasonable too.
We're already halfway through the #2 notch, and I assume that this will continue to be the trend. We will keep overlapping until we have some extension into a +2,+3 milestone lead. This keeps the project moving forward, using core devs to complete current milestones, and new folks can jump on bugs, or documenting future milestone waymarkers.
But really those are just my thoughts and ideas:)
Thomas
On 11 Nov 2003, Ted Gould wrote:
Hey all,
I think it's probably time to start planning the next release, or atleast decide when it's going to be.
Agreed.
On the high-level roadmap at http://www.inkscape.org/roadmap.html we have 'New Features' as principle objective right now, and that may be the best focus for us for the next few releases.
There were several reasons why this was set as the next step in our roadmap:
* New features help attract users, and we need users right now
* It gets us into the habit of making features early in the project
* It's generally easier to figure out what to do for a specific feature, than to come to agreement on and properly prototype for major architectural changes.
* We can get a few stable releases under our belt before embarking on larger changes that could introduce significant risk of bugs.
* Features can be implemented by one or two people so don't require a lot of coordination.
By 'Feature' I mainly mean some small enhancement such as implementing a small aspect of the SVG spec, improving how a dialog box works, things that improve usability, etc.
I think the choice of what features to make should be left to the person or people working on them. Mental is interested in working on Layers, Nathan on the grid system, myself on arrowheads, and Bulia on keyboard bindings. Folks who want to help can either offer to lend a hand on one of these or announce their own feature they'll be working on.
We'll see where we stand in a few weeks and if enough of the new features are ready to go, we'll cut 0.36.
The GNOME folks rave about time based releases, and think they are the best way to do Open Source software. I think this mostly comes from giving everyone a timeline to get features into the repository.
But, Bryce has done a really good job of making Milestones also - and I think that if we set "Milestone 2" as when we're going to do the next release that would be reasonable too.
Thoughts? Ideas?
--Ted
I tend to prefer the milestone approach more than the timeline approach with projects because it can be so tough to predict how long things are going to take to do. That said, it would be great if we could strive to make a release every few weeks.
We still have some important items like the web look & feel to finish up before we can cross off Milestone 1, but with the release yesterday we've accomplished the hardest part.
Milestone 2 needs to be reviewed and revised to make sure the tasks are all things that can be done in a reasonable amount of time. Please take a look at it and put in tasks that you think can and should be achieved for this milestone. I'll work with folks to make sure we have a good plan and post it after M1 is complete.
http://www.inkscape.org/cgi-bin/wiki.pl?DevelopmentRoadmap
Bryce
participants (3)
-
Bryce Harrington -
T Ingham -
Ted Gould