incubator-general mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From Stephen McConnell <>
Subject Re: incubator, exit and publication
Date Fri, 26 Sep 2003 08:02:11 GMT

Berin Lautenbach wrote:

>You will see from my last e-mail that I've lost
>the plot somewhere, but some thoughts because I
>can't resist.


>I was looking at the same thing yesty, but from a slightly different angle.
>I thought (maybe wrong???) that Incubation was
>not about "how good is this code base/product".
>If something is accepted into incubation, then
>we (or rather the ASF) thinks that the basic idea
>is a good one.

Sure - we the ASF may consider it the best thing since sliced bread.  
There is no problem with the notion of ASF opinions.

>Incubation is about "The Apache Way".  Are we
>doing things the way they need to be done, and are
>we building the community the way it needs to be
>to thrive within the Foundation.

One could also argue (rightly or wrongly) that incubaton is equivalent 
to procedural masterbation.  Personally I think there is some value 
added though incubation - namely - pracice makes perfect.

>It is also about protecting the legal back-side of
>the foundation.  Have we done everything we need
>to to ensure that the code being released is now
>owned by Apache etc.

Disagee - this "point of order" canot be closed until the 
project/podling in question is no longer ander the authority of the 
Incubator PMC.  So long as the podling is a podling - it has not cleared 
the criteria.  Therfore the ASF is at fault if it publishes content 
under its license relative to a project that has not exited incuabator.

>Take the second first.  That (it would appear) is
>done fairly early in the incubation process.  I
>think with XMLBeans it's pretty much done.  All
>code has been transferred to Apache, and all CLAs
>have been signed.  So there should be no legal
>reason not to do a release.

Not according to everything w have said about incubation.  If the 
incubation process were to have an intermidiate step of 
*legal-acceptance* - then yes.  But the incubation process does not have 
this step.  Without this step - any action by the Incuabator PMC to 
publish an artifact would IMO be legally irresponsible.

>However I would absolutely argue that we should
>not be allowing *any* form of release, whether
>informal or formal or CVS without having sign off
>from the Shepherd (and possibly the Incubator PMC)
>indicating that all legal issues are covered off.

But the Shepard is only providing a recommendation.  Publication (as 
distict from a release) is a PMC responibility and we are discussing the 
Incuabtopr PMC and as such we are discussing projects within the 
Incubator and as such we are implicity discussing non-Apache content - 
irrespective of best intentions.

>Now the first - community.  Does it really matter
>if we have a good community there or not for the
>purposes of a release? 

No.  We are discussing an act of product publication.
The community behind the publication is academic because the act of 
publication involves that transfer of responsibility from the committers 
to the Incuabator PMC as an working entity of the Board.

> Apache is already standing
>beside this code in some form - it's in the CVS
>and it has an Apache License stuck to the front of
>every file.

I disagree.
The incubator is specifically an entry obsticle.  If at some point the 
Incubator PMC made a decision that the content of a project was suitable 
for publication - one must ask the question - why is the project in the 
incubator?  If the answer is that they have a nice community and we want 
to help them out - then one must restate the question - why are they in 
the incubator?

>Sure we might not be sure that the code base is a
>"great set of code", but are we sure of that for
>any project in Apache?  

Irrespective - the content in Apache is content under the duristiction 
of a particular PMC.  Siad PMC have taken a decision and is 
accountable.  It is not up to you or I to judge that.

>We rely on the local
>community of a sub-project/project to make that
>call.  And anyway - that's not the purpose of 
>I dunno.  I'm not the expert here, and I've seen
>good arguments both ways.  It just seems to me
>that the only absolute argument that should be
>placed in policy (which I'm now working on :>)
>should be that no release in any form may be
>undertaken until there is sign of that all legal
>ownership issues have been covered off.

For me this is insuficient.  Iff the project is go - then release it 
from incubation.  If not then it is still subject to rejection.  If its 
the subject of a potential rejection then it has bnot entered into 
Apache - therefore - donc - the Incubator PMC would be at fault if it 
published anything other than internal process, procedures, related 
documentation, sucess stories, etc.

It is Sponsoring Entity that is (IMVHO) responsible for publication - 
only because it can override the opinions of the Incuabtor PMC as to 
readiness of a technical artifact - as distinct from the readiness of a 
community baking a artifact.

Cheers, Steve.

>Everything else could almost be case by case.
>    Berin
>>From: Stephen McConnell <>
>>Subject: incubator, exit and publication
>>Date: 26/09/2003 16:54:29
>>I'm been following the messages concerning exit criteria and releases 
>>and I see a conflict.  If a project is under incubation, then it is not 
>>accepted into Apache and therfore the content that is generating is 
>>simply content under observation - not Apache content.  As such, how can 
>>a such a project release any content under an Apache license  and 
>>presume the endorcement and protection of the ASF?  My personal 
>>conclusion is that anything under incubation cannot do anything in 
>>relationship to Apache concerning publication, and that the Incubator 
>>PMC would be at fault if it were to endorce the publication of any 
>>artifact related to an incubated project.
>>Can anyone explain to me the error implied by these assumptions?
>>Stephen J. McConnell
>>To unsubscribe, e-mail:
>>For additional commands, e-mail:
>This message was sent through MyMail
>To unsubscribe, e-mail:
>For additional commands, e-mail:


Stephen J. McConnell

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

View raw message