On 10/12/06, Noel J. Bergman <[EMAIL PROTECTED]> wrote:
Roy T. Fielding wrote:
> > robert burrell donkin wrote:
> > the source distributions unpacks to the same directory as the binary.
> > this is inconvenient for users. it's better to unpack the source to
> > incubator-activemq-4.0.2-src.
> I disagree with that.
> I don't think there is a generally accepted Apache way of separating
> these things, since the really important part is the source tree.
> I think it is important to make it clear that the binaries are just
> a convenience layer and not a separate distribution (even if they are
> distributed separately).
Can we agree that regardless of which style one might prefer the packaging,
there are multiple valid approaches, and that this level of difference
should not be a release criteria for the Incubator?
the only critieria we have is three +1's
The Mentors can and should engage the community on best practices. When the
Incubator PMC is presented with a release to approve, we ought to focus on
actual requirements, such as:
Licensing
Notification
Signing (if we choose to enforce this)
...
the reason i didn't +1 wasn't anything to do with the unpacking but
the fact that there are a lot of files without license headers and so
of dubious original.
And what those actual requirements are should be documented so that the
projects aren't surprised when submitting their request. If we decide to
add requirements, we should agree to add them to the release requirements
document.
we don't have a requirements document. we don't have a requirements
process. it's a simple vote.
when i review a release, i post additional feedback after critical
issues in a section called notes. the comment about source directory
was in that section. perhaps i'll change the format so that it's a
little more obvious that these are comments intended as feedback.
unfotunately, all the context was cut so it's not clear to people
jumping onto the thread.
i've never heard anyone defend intentionally unpacking into the same
directory before. i was genuinely interested that people do this for a
reason. this happens quite often by mistake so i try to pick it up as
feedback.
Agreed?
nope :-)
i'm not willing to +1 a release that i'm not personally happy with.
the only reasons why my refusal to +1 a release makes any difference
is that too few PMCers review releases. if PMCers feel strongly about
this issue then they should devote the hour or so that each release
takes to review.
- robert
---------------------------------------------------------------------
To unsubscribe, e-mail: [EMAIL PROTECTED]
For additional commands, e-mail: [EMAIL PROTECTED]
