I guess it really depends on how critical you view the problem then. I'm certainly not advocating a bi-weekly release schedule (I found out first-hand how much of a pain it is to prepare), but likewise, I don't see it being problematic to issue a quick fix to a critical problem. Arguably, it's something that should have been caught in testing for the 3.0M3 release.
If it's not that critical, then waiting until the next release cycle shouldn't be a problem. As for me, I think the 4 - 6 week is a sweetspot. As is, it takes us 2 - 3 weeks to get a release out the door. So waiting upwards of 10 + 3 seems like a really long time to me. Once again, I have really high hopes that the CI server and/or Cayenne zone (any update on this?) will help streamline the process. -- Kevin On 2/12/08 12:32 PM, "Andrus Adamchik" <[EMAIL PROTECTED]> wrote: > I agree except that posting releases every week will create a huge > overhead for the the PMC to test it before the vote... There is a > difference between a release and a nightly build in the expected > quality, so we can't just let it out without some thorough testing. I > suggest to start doing releases more often, but not *that* often. We > can aim for a release every 6-10 weeks? > > Andrus > > > On Feb 12, 2008, at 12:17 PM, Kevin Menard wrote: > >> I know I'm stating the obvious, but there's nothing wrong with us >> pushing >> out a 3.0M4. This would have the benefit of working for everyone, >> rather >> than just those that happened to see your post. >> >> While not a ton has changed, there were some fairly important bug >> fixes. I >> think CAY-574, for example, could help out a lot of ROP users. >> >> -- >> Kevin >> >> On 2/12/08 12:05 PM, "Andrus Adamchik" <[EMAIL PROTECTED]> wrote: >> >>> FYI: I just fixed a nasty Modeler bug below that affected >>> CayenneModeler 3.0M3. You can get a Mac and Windows versions of the >>> patched Modeler following this link (consider it an unofficial >>> nightly >>> build) : >>> >>> http://people.apache.org/~aadamchik/nightly/02122008/ >>> >>> Andrus >>> >>> >>> Begin forwarded message: >>> >>>> From: "Andrus Adamchik (JIRA)" <[email protected]> >>>> Date: February 12, 2008 11:27:33 AM EST >>>> To: [email protected] >>>> Subject: [JIRA] Created: (CAY-984) Runtime relationships leak into >>>> CayenneModeler >>>> Reply-To: [email protected] >>>> >>>> Runtime relationships leak into CayenneModeler >>>> ---------------------------------------------- >>>> >>>> Key: CAY-984 >>>> URL: https://issues.apache.org/cayenne/browse/CAY-984 >>>> Project: Cayenne >>>> Issue Type: Bug >>>> Components: CayenneModeler GUI >>>> Affects Versions: 3.0 >>>> Reporter: Andrus Adamchik >>>> Assignee: Andrus Adamchik >>>> Priority: Critical >>>> Fix For: 3.0 >>>> >>>> >>>> This affects 3.0 M3 and causes lots of user confusion and potential >>>> modeling errors.... In *runtime* Cayenne creates missing reverse >>>> relationships to have a consistent mapping graph internally. This >>>> should be invisible to the Modeler... however it is not, as when a >>>> project is loaded, the "defaults" are applied and runtime >>>> relationships are shown to the user. Luckily they are not saved to >>>> XML , but the whole things is very confusing. >>>> >>>> -- >>>> This message is automatically generated by JIRA. >>>> - >>>> You can reply to this email to add a comment to the issue online. >>>> >>>> >>> >> >> >
