AbiWord 2.7.0 plans

Marc and I have been discussing a possible 2.7.0 release on #abiword,
and we're looking for your input on how to proceed. It seems like we
have two options, as of now. We can release 2.7.0 fairly soon and
push cairo integration off until 2.9.0 (to give adequate time for
testing) or we can delay 2.7.0 until the cairo work is ready. As I
see it, this is how the two options break down ('+' for a benefit, '-'
for a drawback):

Releasing soon:
+ A shorter release cycle with enough new features -- annotations,
smart quotes, and S5 exporter -- to "justify" a release
+ The prospect of a subsequent release with cairo integration and any
other improvements from the upcoming Summer of Code
+ More concentrated testing of the annotations and smart quotes features (?)
+ Annotations support on the OLPC sooner (?)
- We keep our existing rendering code and libgnomeprint dependency
- GOffice plug-in languishes

Waiting:
+ The benefits of cairo integration and the ability to use libgtkprint
+ Removal of our libgnomeprint dependency
+ GOffice plug-in can depend on newer versions of libgoffice
- Longer release cycle, which could get out of control like 2.5.x
- More new features to try to get ready before a .0 release

I'm sure I missed a few benefits and drawbacks, so feel free to
mention them. I guess another option is to release now and integrate
cairo further into the 2.7 series, but I'm not sure it would be a good
idea to introduce a new feature in the middle of a release cycle. I
personally think a quick 2.7.0 release is the best option, since it
would establish a precedence for quicker development release cycles
and it would probably result in a more polished 2.8.0 release. Let us
know what you think.
Received on Sat May 10 01:38:15 2008