No offense taken. In fact it is probably a good idea for us to be clearer about exactly what we are voting for. Certainly we don't all have time to test every release in enormous detail. How much testing does each release get?
* verify all signatures ( I do that sometimes but not every time)
* run all unit tests on the source (I used to do that, but now I'd like to entrust that to Hudson)
* make sure the Modeler runs and opens some models I have (usually, but only OSX, never Windows)
* put the jar into my own project and see that it still works and passes tests (sometimes, not this time)
* test it against every database Cayenne supports (well, that never happens...)
My vote here is not to say that this release is perfect and I've tested everything. More that I've been watching all the commits and know what has gone into this release and I believe that now is a good time to cut RC1. I've had a look to see that Andrus didn't accidentally upload pictures of his cat instead of the release, and I like that the jars folder is now tidier without the extraneous files. I'm happy to vote that this is a really good time for the project to release RC1 and that the release (with 15 minutes of investigation) look fine and that the commits since the last release are all non-intrusive and unlikely to cause major issues.
Cheers
Ari
On 1/01/10 4:37 AM, Kevin Menard wrote:
> I hope I don't come off as a jerk with this, but you're voting on more than
> just a release in the grander sense. You're voting that the binaries work
> as advertised, that there aren't any packaging issues, that the gpg
> signatures match, etc. It's more of a quality control step at this point.
> If you don't have time to test them, that's fine, but it probably shouldn't
> be a +1 vote at that stage, IMHO.
>
----------------------------> Aristedes Maniatis GPG fingerprint CBFB 84B4 738D 4E87 5E5C 5EFA EF6A 7D2E 3E49 102A
This archive was generated by hypermail 2.0.0 : Thu Dec 31 2009 - 18:58:15 EST