On Mon, 2013-07-01 at 09:47 +0200, Bastien Nocera wrote:
> On Fri, 2013-06-28 at 17:46 +0530, Arun Raghavan wrote:
[...]
> > I think it makes sense to use this approach with 3.10 as well - support
> > BlueZ 4 and 5 as a compile-time option for this release, get some more
> > user-testing during this cycle from the more adventurous, and then
> > switch to BlueZ 5 only in 3.12
> 
> Given that we had to break the API to support Bluez5, I really don't
> want to be supporting 2 versions of BlueZ in gnome-bluetooth and
> associated modules. Waiting another 6 months would mean that we'd rely
> on the unsupported BlueZ 4 code for a total of a year and a half.
> 
> I'd rather ship without Bluetooth audio support in 3.10.0 and have the
> support come in in an subsequent point update. This is definitely better
> than relying on unsupported versions of BlueZ.

"BlueZ 4 is no longer supported" is only a tenable argument if it didn't
drop functionality. Put another way, from users' perspective, BlueZ 4
vs. 5 doesn't matter. Having something that worked fine go away, for no
visible gain, does.

I'm not even saying we should be providing active support for BlueZ 4
installs - just that we deal with messiness of retaining 4.x support in
code for one cycle.

Cheers,
Arun

_______________________________________________
desktop-devel-list mailing list
[email protected]
https://mail.gnome.org/mailman/listinfo/desktop-devel-list

Reply via email to