On Sunday 15 March 2009 14:10:13 Martin Schlander wrote:
This discussion started when someone said that openSUSE is what we make of it, but both 10.1 and 11.1 show that at least every 4th openSUSE release (the x.1s) are very much so at the mercy of structures, planning and decisions made by some invisible hand that I usally refer to as "Novell". I think it's difficult to grow when every 4th release automatically sucks.
It would of course help if fixing bugs wouldn't automatically be the task of "Novell". And what you say is wrong: openSUSE and SLE had feature freeze on the very same day. But e.g. SLE had more time to test and fix strings and translations and it showed - we wouldn't have changed that many strings (that were arguably very often broken or not translatable) without SLE.
11.1 wasn't automatically doomed because of SLE though, it had more to do with the short release cycle, the general fact that Novell developers were busy with features (and very often these features had no influence on the quality of openSUSE), the change to build within the build service and to a small part too the fact that we had a break in managing openSUSE releases due to several people being responsible over the course of 11.1. That surely didn't help either.
But again, I think the biggest seed of growth of openSUSE are people from outside wanting to drive things - not just vote for a feature someone else has to do. It works pretty well e.g. with translations where I see a huge responsibility people take - and taking regressions as personal offense. There "Novell" has a very weak position and this is what it should be like I think :)
There are many other examples (your oxygen icon set comes to my mind), but we're not there yet not to rely on "Novell" - and so Novell's products will have an impact - if they are on the same code base as openSUSE releases or not.