lucene-dev mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From Grant Ingersoll <>
Subject Re: [Lucene-java Wiki] Trivial Update of "ReleaseTodo" by YonikSeeley
Date Mon, 20 Sep 2010 17:49:17 GMT

On Sep 20, 2010, at 1:07 PM, Yonik Seeley wrote:

> On Mon, Sep 20, 2010 at 1:01 PM, Grant Ingersoll <> wrote:
>> - ''Note: You need committer rights to create a new Lucene release.''
>> + This page is to help a Lucene/Solr committer create a new release (you
>> need committer rights for some of the steps to create an official release).
>> It does not reflect official release policy - many of the items may be
>> optional, or may be modified as necessary.
>> I think putting this up on the wiki is a bad idea.  We should strive to have
>> a repeatable release process.  By saying it is up to the person who happens
>> to be doing the release is just asking for less quality in our releases.  If
>> you don't think you can follow the release process, then you shouldn't be
>> doing the release.  And, if we as a community can't define a repeatable
>> release process, then we shouldn't have a release either.
> Calling something that anyone can go and edit and add their best ideas
> to "official" is silly.

Fine, let's lock it down then.

> It does not list iron-clad requirements - it is there simply to help.

Again, I disagree.  Having done a number of releases, it would simply be impossible without
it, no matter how long the list is.  Unless, of course, all you want is the release to be
the source, but even that is in doubt b/c how would I know where to upload it to?  For instance,
how do you know which Ant target really gets you the right thing to distribute?

> That's pretty obvious by looking at the huge list of content on that
> page.  I'd rather spend my time writing code and improving the
> projects rather than engaging in bureaucratic exercises.

Well, part of an "improved" projects is a release that people can consistently rely on.  If
there is too much chaff in the current release, fine, let's get rid of it or automate it.
 However, to suggest that a written out release process is not needed or is subject to whatever
the RM wants is just plain ludicrous.  Are you really arguing that we, the writers of a massively
used and deployed open source library, should have a release process that is subject to the
whims of whoever happens to be doing it on that given day?  Regardless as to whether you want
to or not, we as a community need to make sure the community can rely on the results of us
writing the code.

To unsubscribe, e-mail:
For additional commands, e-mail:

View raw message