If you know where to look you can find the odd dusty comment posted in the archive of some quiet little corner of the Internet. A whispering, if you will, in the shadows of someone or others blog or website about Blender 3D. These are not the usual "yay another release" exclamatory exaltations (!!!!!) of the 99.9%. They are instead the genuine concerns of the remaining 0.1%; the real people doing real things in the real world with Blender.
It's worded in different ways of course, depending as it is on where it's asked. Blowing away Father Time then reveals a simple request, a magic one even, one that could summon the power of the gods of Bits and Bytes... "... if only the Blender Foundation would change their mindset on how they number and promote each version of Blender, "stable" should mean just that, that it works".
And yet the spell casters at aforementioned institution keep getting it wrong. The release of 2.60 is a prime example. New features are added. Great. Previously working features break. Not so great. That's to be expected. What isn't however, is how some of these breakages get past QA. That's a really big problem for the Foundation because it means the core team are constantly doubling up their workload on each release fixing problems that shouldn't be problems, wasting time on 'stuff' that could be better spent elsewhere. It means 'buggy' software released as being "stable" when it's really not; at most it's a test release based on a previous iteration, a 2.5.10 instead of a 2.60. Reading between the lines it doesn't really seem that important an issue to the Foundation though, problems eventually get fixed, this is open-source software after all.
For the professional user however, those that ironically don't really seem to have much of a voice whilst at the same time being actively pursued to use, switch or otherwise promote Blender, means it's still unreliable, even after all these years - "it's always been this way, the constant changes and updates mean it breaks every-other version if you're lucky. So what's new?".
However, and this is the broader picture the Foundation seem to be ignoring, what professional studio, business or individual can afford to invest in that risk? For game development studios, small start-ups in particular, the problem is especially acute because core import and export features that are the backbone of production, are typically the first on the receiving end of QA being MIA. That shouldn't be happening, open-source or not. The funding is there, the resources are there, the people and the enthusiasm. So what's going wrong? Why does it keep happening?. Do (relatively unimportant) features get lost in "the bigger picture" of adding new bells and whistles that no-one actually does anything with? Be very wary of "Feature Creep" Foundation, it's a monster that can kill.
