[Qgis-developer] Merging of incompatible changes

[haubourg]

 +1 for a spring or mid 2013 release of 2.0 with API break + plugin task
force. I was afraid that API change could lead to end 2013. 

I would be please to support such an effort. I agree we need multithreading
enabled, and a polished API. But I remember the amount of effort needed to
release a version without blocker like in 1.7.1-1.7.3 period.
Of course we have a test environnement now, but, 1.9 have so many new
features... and still very blocking issues without a clue to solve them
(crash in composer for high resolution exports, shp network performances, 
It makes me think we should'nt underestimate work to have a stable 2.0.

I also agree with Radim saying an API must last +/- 5 years.

-1 for keeping a legacy API call method, because I estimate actual API
really need improvements. If we wait, plugin repositories will have more and
more plugin bases on old API, and migration effort will be even higher.


Then I insist on the fact that users in professionnals environnements need
regular stable releases so that master is not to far from stable branch. (I
don't want to pay for special releases for my own use... this is wasted
money and dev time)

Régis 



--
View this message in context: http://osgeo-org.1560.n6.nabble.com/Merging-of-incompatible-changes-tp5010325p5011109.html
Sent from the Quantum GIS - Developer mailing list archive at Nabble.com.