incubator-general mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From Rob Weir <>
Subject Re: [VOTE] Apache OpenOffice Community Graduation Vote
Date Tue, 21 Aug 2012 00:14:36 GMT
On Mon, Aug 20, 2012 at 8:01 PM, Marvin Humphrey <> wrote:
> On Mon, Aug 20, 2012 at 3:03 PM, drew <> wrote:
>> Well, for myself, I don't have a problem with the AOO project not having
>> official binary releases - in such a circumstance I would strongly
>> prefer no binary release at all.
> I wonder who might step into the breach to provide binaries for such a
> package...
>> On the other hand if there is a binary release from the AOO project then
>> I believe it should be treated as a fully endorsed action.
> At the ASF, the source release is canonical.  I have never seen anyone assert
> that the source release is not offical and endorsed by the ASF.

What would suggest is the concrete distinction between an "official"
binary and an "unofficial' binary?

I'd assert all binaries that I've seen a project release have these qualities:


2) Are build from the canonical source

3) Can use other 3rd party components per policy

4) Are voted on by the PMC's

5) Have hashes and detached digital signatures

6) Are distributed via the Apache mirrors

7) Are linked to on websites and announcements

8) Are used by and appreciated by users

9) Are for the public good

Which of these do would you say are not qualities of an "unofficial
binary"?  Or would you suggest another?

Unless ASF or IPMC policy defines a distinction here, I think we're
just arguing about what color the bike shed is for angels dancing on a
head of pin.  It is a distinction without a difference, or at least
not one that has been stated,


> There has been disagreement about whether binaries should be official or not.
> To the best of my knowledge, every time the matter has come up, the debate has
> been resolved with a compromise: that while binary releases are not endorsed
> by the ASF, they may be provided in addition to the source release for the
> "convenience" of users.
> What is different with AOO is that the compromise does not seem to satisfy
> an element within the PPMC and thus the matter is being forced.
> It would be a lot of hard, time-consuming work for the ASF to build the
> institutions necessary to provide binary releases that approach the standards
> our source releases set.  (As illustrated by e.g. the challenges of setting up
> the code signing service.)  Not all of us are convinced that it is for the
> best, either.
> Marvin Humphrey
> ---------------------------------------------------------------------
> To unsubscribe, e-mail:
> For additional commands, e-mail:

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

View raw message