> Ok, use revision 31190. Everyone happy now? Look, there's two sane approaches to this. Either: - you generate official releases (ideally with binaries), and yes that means you are going to be asked to support them regardless of whether a fix is already checked into source control for the user's problem, or - you continue with the current model in which you can ask people to use "latest SVN", but "latest SVN" might sometimes not work and is in any case not be available to a lot of people without a frankly unreasonable amount of messing about. Neither is ideal, but I think the content of this thread is ample evidence that the current approach is just winding people up, both developers and users. Nobody's really winning here. At the end of the day this is why commercial developers have alphas and betas and RCs and releases, and yes it is annoying and yes it is time consuming and yes nobody really cares to do it. At some point though presumably people work on FOSS "in the hope that it will be useful", and I think it's worth spending some time on this stuff or you just cripple the usefulness of it for everyone. I don't care what the solution is, I'd just like to see one that works. If there's any way I can help, as a non-software-engineer, then let me know. P
