incubator-general mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
Subject Re: Blondie's Parallel Lines...
Date Thu, 02 Jun 2011 19:34:51 GMT wrote on 06/02/2011 02:42:11 PM:

> No Rob, I don't question your credentials, have not done that, will 
> done that. Both of us know better than having that kind of talk, both of 
> have worked together for years now, at the OASIS and elsewhere. What I'm
> questioning is the ability to have two projects, OpenOffice and 
> with so much overlap and only a vaguely defined reason to have two 
> projects (the reason being, that some contributors -IBM- might prefer 
> Apache licensing). What I'm concerned is the fuzziness around the 
> who would contribute to the codebase. For someone who has
> repeatedly explained that the LibreOffice developers were not that many, 
> think that betting on a sustainable project here is a 
> leap of faith.

Hi Charles,

Maybe this will make it a little more plausible. 

As you know IBM develops Lotus Symphony, which is essentially a fork of 
OpenOffice.  IBM has experience in how many developers are required to 
code, test, translate, document, support, etc., a project of this size. 
We've been doing it for several years.  It does not require 400 
developers.  It does not require 200 developers.  It does not require 100 
or even 50 developers.  If you claim to have 200 developers working on LO 
then I suspect this is with a very low level of engagement. 

When I check the commit logs for LibreOffice and apply the Apache criteria 
for what defines an "active" participant (a commit within the last 6 
months), I see only 54 names.  And most of those names are making very 
sporadic, but I'm sure very valuable, contributions.  Notably the top 20 
contributors were making 90% of the commits and of those the majority are 
Novell employees.

So it is clear that even with LO, a small number of core developers, even 
just 20, do almost all the core coding. This observation is consistent 
with what I know about the development of Symphony.

So I believe that a reasonable goal for Apache OpenOffice, for graduation 
from incubation, is to have a set of at least 20 active committers.  That 
should be sufficient, as a bare minimum, to be the developer nucleus of a 
respectable project.

Now is it plausible to get to that number?  I think so.  But let's not set 
some bogus target of 400 developers or whatever.  There is no intent to 
dump the code with no developers.  But I don't think we want to crowd 
source the project either.  I think we want a core group of dedicated 
committers who can facilitate the review and integration of patches from a 
larger number of less-engaged developers.  That is the kind of 
distribution I think we'll want.  But our target metric should be the 
around active committers. 

The "halo" of additional developers is important as well.  But their 
effectiveness is entirely dependent on the ability of the core committers 
to review and integrate their work.  So we need to grow the project from 
the inside out.  That's my opinion, in any case.  But LO is really no 
different.  Its core is developers transplanted from the Novell Edition of 
OpenOffice.  Surely, there is nothing that prevents other companies with 
OpenOffice forks from doing exactly the same thing.

> I am certainly not going to enter a debate on licensing, and I think 
> wants that here. But I just think that there are other ways to cooperate
> than pretending the elephant in the room (LibreOffice, the Document
> Foundation) does not exist or does not de facto embody the largest part 
> the OpenOffice community (yes, I know, there are a few exceptions).

Please, Charles, stop saying that anyone is saying that LibreOffice does 
not exist.  You are here, on the Apache list, at the invitation of Apache. 
 I'm happy to stipulate that you exist, I exist, exists, 
Apache exists and that TDF/LO exists.



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

View raw message