I think there are different definitions of "release" that are confusing things.
One is a release from the perspective of ASF which is concerned with the process (votes on the MLs, etc.) and certain legal requirements. I think Gianugo's last email expresses this focus on this consensus. What is released (alpha release, beta release, RC, etc.) is here not the focus. The other definition of release is more associated with the development status of the software in question. For me, the focus was more on the ASF definition of the release. My main goal was to use the release to meet the legal requirements (ESME-47 + other concerns) necessary as well as learn "how to release code" which is one of the main criterion for becoming a TLP. I also see a release as a sign of a project's maturity. Obviously, the current release isn't perfect - that is one of the reasons why I like the idea of tagging it as a RC rather than our official release. If you take a look at the the releases of the other projects in the incubator, you will also see a variety of release names: http://www.apache.org/dist/incubator/ D. On Tue, Mar 2, 2010 at 12:26 AM, Anne Kathrine Petterøe <[email protected]> wrote: > Glad we are having the discussion now, because I also got a bit confused by > the RC "acting" like a full Apache release :) > > Ethan, I agree with your definition of RC vs full release. > At this point I cannot see why we should go through a full release process to > see if we are ready to make a release. It feels like a slight overkill, at > least for the time being when only 3-4 people (afaik) on the dev list use the > RC for testing purposes. > > - anne > > > On 2. mars 2010, at 00.14, Ethan Jewett wrote: > >> How is publishing a tarball on a public site that is clearly labelled >> as a trial release and *not* for general consumption significantly >> different than making available a public SVN repository? >> >> Ethan >> >> On Mon, Mar 1, 2010 at 6:06 PM, Gianugo Rabellino >> <[email protected]> wrote: >>> On Mon, Mar 1, 2010 at 11:57 PM, Ethan Jewett <[email protected]> wrote: >>>> If we are cutting a tag into a trial release in order to help >>>> ourselves figure out if we are ready to make a release, then I think >>>> we should call it a "release candidate" and we should not go through >>>> the full Apache release process and we should not "release" it. We >>>> should just put the RC tarball out on the public site for people to >>>> evaluate, >>> >>> Which at great length fits the ASF definition of a release... :) >>> >>> "Releases are, by definition, anything that is published beyond the >>> group that owns it. In our case, that means any publication outside >>> the group of people on the product dev list. If the general public is >>> being instructed to download a package, then that package has been >>> released." (http://apache.org/dev/release.html) >>> >>> -- >>> Gianugo Rabellino >>> M: +44 779 5364 932 / +39 389 44 26 846 >>> Sourcesense - making sense of Open Source: http://www.sourcesense.com >>> > >
