When releasing FOSS software a 0.0.1 alpha release qualifies itself as being suitable for public beta testing. You should expect no major changes to the underpinnings, just a few bug fixes.
Likewise, when the project hits beta it’s actually a release candidate and should be treated as such. As a user, you should rely heavily on it for completing your work. As a developer you should code against it, assuming nothing will change between now and the final release.
Furthermore, when the project hits release candidate (assuming the project even has one) you should treat it as a final release. Tell all your friends and relatives to download it; Grandma won’t have any bugs to deal with at all.
Eventually, if the project doesn’t die, it will hit a final release. At this point claim it’s an outdated piece of garbage and focus on the the next alpha.
Comments
You're not thinking about Ubuntu 10.04 LTS at all, are you?
Actually, I'd welcome Loon clarity on the concept of LTS. Releasing precisely the same alpha-quality cack and promising precisely the same cack level of support for three to five years ... well, that's one hell of a marketing proposition.
What's the betting that, in six months' time (I have no idea where I pulled that number from), they'll issue a Service Pack -- ie a complete reinstall -- and boast about their support levels?
You think that? You really do?
I wonder where this comes from, considering that there is no 'release' grade software ( http://www.gnu.org/software/gss/ ) that would depend on something in alpha grade for many years now ( http://www.gnu.org/software/shishi/ and 6 years )
On the other hand there is no outdated 'release' software either ( http://www.hpl.hp.com/personal/Hans_Boehm/gc/gc_source/ <- GC 7.1 doesn't work with about any application requiring GC, but 7.2 ALPHA is just fine ).
I wonder what you might be talking about.
You must be signed in to leave comments.