GNOME 2.30: go stable or go cutting edge?

[Willie Walker <walker willie gmail com>]

Hi All:

GNOME 2.30 is coming out on the Ides of March (March 15).  I have one main question for you: do you want it to be stable or do you want it to have more cutting edge stuff?  This question is predicated on the assumption that GNOME 2.30 is the last of the GNOME 2 releases and GNOME 3 is coming out this fall.  It also assumes that we will resolve the harder problems we currently have with AT-SPI/D-Bus very soon.

Here's the background -- GNOME Accessibility has been facing a "perfect storm" for the GNOME 2.30 cycle.  The three major fronts of this storm are: Bonobo/CORBA elimination, WebKit accessibility, and GNOME Shell accessibility.  You can read a lengthy summary of the current state of the work at http://live.gnome.org/Accessibility/GNOME3.

Here's some pros/cons.  Note that the quantity of pros/cons doesn't necessarily mean anything.  They are just talking points, and actually quite simple at that.

PROS/CONS for going with the cutting edge:
==========================================

PRO: GNOME accessibility may be more widely available on smaller/mobile devices -- those devices are happy to have D-Bus but do not want CORBA.

PRO: The cutting edge stuff will likely get more testing coverage for GNOME 3.0, helping improve the GNOME 3.0 accessibility experience.

PRO: We will be able to eliminate a huge portion of deprecated stuff in GNOME.

CON: GNOME 2.30 accessibility could very well be unstable or slow for day-to-day use for doing your job or functioning in life.  Staying on GNOME 2.28.x would be recommended for people who need more stability.

CON: GOK will not work.  OnBoard and an early form of Caribou would be the on screen keyboard solutions.

PROS/CONS for staying stable:
=============================

PRO: Users should still be able to use GNOME 2.30 with the same stability and reliability they get with GNOME 2.28.x.

PRO: GOK will work.

CON: The testing of cutting edge stuff may not be as broad, so GNOME 3.0 may go out without as much testing as it needs.

CON: GNOME will need to continue to carry Bonobo/CORBA around.

CON: GNOME accessibility will remain unavailable on smaller/mobile devices that do not ship Bonobo/CORBA.

My first concern is the end user.  As a result, I tend to be more conservative and lean towards stability.  That is, making sure GNOME provides a compelling accessible desktop for reliable and usable day-to-day activity goes a long way to addressing the needs of the user.  With this, we're likely to say GNOME 3.0 will be more wrinkled in terms of accessibility and we could look to GNOME 3.2 and 3.4 to iron things out.

However, given where we are with proximity to GNOME 3, I'm also tempted by the notion of getting the new stuff out there sooner.  This would potentially forsake the accessibility of the last (or one of the last) releases of the GNOME 2 series while helping set us up for an earlier accessibility success for GNOME 3.

Please, speak up with your thoughts.  The collective opinion of our group matters and it will help shape what recommendations we will make to the release team for GNOME 2.30.

Will