Guys, I agree, for me a non regression test infrastructure is the most critical thing Tiny should do before attempting any new significant code change. I created a blueprint for such a test environment, interested folks can comment here: https://blueprints.launchpad.net/openobject-server/+spec/openerp-requires-a-test-server
Looking at the past, I think that other issues have been just as critical, for instance: - the public forge infrastructure - the documentation - the performance On all those points Tiny made a great work since one year and I think those points are mostly over now. If they do as much in respect to tests and code quality, next v5.2/v6 could be awesome. Hope they are taking that point seriously, I think 30% of integration costs could be cut if quality were much better. Currently it's always hard to explain to customers they should pay us to test for daily regressions or trivial bugfixing. Not to say that because of this hassle making integrators busy, fewer customers can enjoy to get in production. See you in the irc. Raphaël Valyi ------------------------ http://www.smile.fr -------------------- m2f -------------------- -- http://www.openobject.com/forum/viewtopic.php?p=36223#36223 -------------------- m2f --------------------
_______________________________________________ Tinyerp-users mailing list http://tiny.be/mailman/listinfo/tinyerp-users
