incubator-general mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From Ross Gardler <>
Subject Re: Apache Incubator Proposal
Date Wed, 01 Jun 2011 22:03:09 GMT
On 01/06/2011 19:24, wrote:
> Ross Gardler<>  wrote on 06/01/2011 12:21:23 PM:
>> There are only two initial committers identified in the proposal. Why
>> only two for such a large codebase?
> We could have put a much longer list of IBM names on this list, developers
> familiar with the code base via their work on Lotus Symphony (which is our
> OpenOffice based project).  But then we could have been criticized for the
> proposal being too dominated by IBM.  It is clearly our intent to grow
> this project, both from our corporate developers, but also by recruiting
> new members to the project, including developers from related open source
> projects (see my previous note)

So my "optimist interpretation" earlier in the thread was accurate. I 
think this is a sensible move. Normally we don't care about projects 
heavily influenced by a single company as long as the community is 
balanced. The incubator is here to bring that balance. However, I 
understand that in this case there are other considerations.

It might be worth making this decision explicit in the proposal though. 
Personally I see it as a strength of the proposal. I suggest something like:

"In order to help facilitate the creation of a broad and varied project 
built upon merit as required of an Apache project we have not loaded the 
initial committer list with contributors from a single company. Our 
intention is for the initial committer list to be representative of the 
various users of OOo code."

I realise that this might slow down entry into the incubator, but I feel 
that (if its an accurate representation of your intention) it will serve 
as an olive branch to "members of related open source projects".

>  From a practical perspective it would have been impossible to do all of
> that recruitment without this proposal becoming public prematurely. So the
> majority of the recruitment will occur during incubation.  We obviously
> don't graduate from incubation with only two.  But it should be enough to
> get the ball rolling.

Yes, I think it is now that we have your explanation (even without my 
proposal to expand the list during proposal stage).

> I am "robweir", committer (inactive) for Apache Xalan.

Excellent, thanks Rob. Glad to see you back at the ASF.

>> There is a statement that "Oracle will assist in the transition and
>> migration from", I am probably reading too much into it,
>> but why is there not a statement that Oracle intend to continue
>> development once the transition is complete?
> Companies don't write code.  People do.  The intent is to get the best
> developers we can to continue working on this project, regardless of the
> former or current affiliations.

Fair comment

[ASIDE: I sheepishly admit to being upset about having to make this very 
point in a proposal I was involved with just a couple of months ago ;-)

>> As you will no doubt know, the incubator is not a place for code dumps
>> and I expect that recent events will make plenty of people worry that
>> this is, in fact, a code dump. By answering these questions I hope you
>> can start to address these concerns for the Incubator PMC.
> Is there any feasible way that I can prove, in advance, that a project
> will be successful?  Is there any concrete step I can take now to prevent
> people from worrying?  A little skepticism is warranted.  But my
> understanding is that this is why we have the Incubator, for projects to
> prove themselves.

There is no way of proving it in advance, no. The concrete steps you can 
take are the ones you are taking, answer the concerns of those on the 
Incubator Project Management Committee.

The Incubator is not here for projects to "prove themselves", it is here 
to ensure that ASF Top Level Projects are viable. This means a viable 
community. There is little point in entering the incubator if it has no 
chance of being viable.

You know, as well as we do, that "build it and they will come" does not 
work. As the proposal was written it sounded like a "build it and they 
will come". My questions were targetted at the areas I felt needed to be 
addressed in order to remove this concern.

That being said, the answers given so far have, to a large extent, 
satisfied my concerns about entry to the incubator (I'm not ready to 
vote yet). Successful exit is a different issue of course ;-)


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

View raw message