Re: [Libreoffice] Proposal to join Apache OpenOffice

Hi Allen, *,

On Sat, Jun 4, 2011 at 3:08 PM, Allen Pulsifer <pulsifer@openoffice.org> wrote:
> [...]
> I don't know what vision IBM has for the project.  I don't know what code
> contribution they are going to make--I'm certain they will make some, but I
> don't know what they will be.  I don't know what contributions members of
> the LibreOffice community will or will not want to make.

Given that they had 35 people working on it according to their press
releases, that was ended up in OOo was  basically nonexistent. As
you've been with the OOo project for a couple of years you can
probably understand that people that were part of OOo project before
switching over to TDF/LibreOffice don't have much trust in IBM's lip
service.

The few times they did contribute, it was code-dumping, far from
contributing in a collaborative manner. The accessibility stuff that
Rob just mentioned on the apache list has been promised since 2007 and
he correctly stated that is is still (considerable) amount of /work/
needed to get it integrated. They dumped it instead of contributing
it. To me that's still a difference. The code is against an obsolete
branch (OOo 1.1.5 codeline (!))
http://wiki.services.openoffice.org/wiki/Accessibility/IAccessible2_support

> I do know this however.  There is currently an open invitation for us to get
> involved.  If we get involved, we can have a say in with direction of the
> project.

Not really, as you first have to "surrender" to the Apache's licence
terms. And that alone is reason for me not to join the effort.

>  We can ensure that direction of the project provides the maximum
> benefit for LibreOffice, which includes any contributions from IBM.
> Basically, we can get IBM working for us.

I really doubt it. What would change for them now, with the permissive
licence, that did prevent them in the last 5 years from contributing?
They (according to their press release) had massive manpower working
on it (35 people), but what ended up in OOo is two code dumps to
ancient codeline, one of which being lotuswordprofilter, the other the
abovementioned accessibility dump.

But before you say: It's not only IBM in the foundation. Then let me
ask: Who else is? Oracle is gone for good. The few  individual
contributers that have enlisted themselves as initial contributers on
the apache wiki are to a big extent non-coder. (Not to say that the
non-code contributors are not important, that's far from being my
point)
I currently find 5 people in that list of whom I'd say the have /some/
idea of the code. And two of those already have a focus on a
side-project/fork of OOo.

So if you ask me who is on the Apache project who is not engaged in
TDF/LO, then the only answer is: IBM.
(But I'm also well aware that the proposal is new, and there might be
more to come, and I'm also aware that to the apache-voting the big
picture doesn't matter, they don't care whether it is considered a
good idea or not. If there are enough people to run the podling and if
IBM can convince them that it is possible to get rid of all the
thirdparty stuff that doesn't comply with the strict licencing terms,
they will approve it as an incubator project)

And I don't really see a point in shifting this perception "now that
nobody cares who enlists".
IMHO you only should enlist yourself if you're really convinced that
the Apache Foundation along with its restrictions/limitations and
rules, esp. regarding licencing are a good idea, when you actually
support the move.

If you do, then go ahead and add yourself, I won't question your decision.

The only "reason" on why the TDF should contribute is to why neooffice
did "join" go-oo at the time: To make grabbing their code easier. But
that is a very, very weak reason in my opinion.

> So what I would like to see is an many LibreOffice people at the table as
> possible.  If possible, I would like to see LibreOffice people dominating
> the Apache OpenOffice community to get as much out of the project as we can.

What is the point? If it is run by LO people, what is the benefit of
creating another entity instead of letting OOo be what it is (or
better was), and instead focusing only on LibreOffice?

ciao
Christian