Hi Eric, and other OpenOffice.org and Mac afficionados,

> Ed can be reached via this mailinglist or via private Mail.
> But since in the past there was a lot of miscommunication between
> NeoOffice/J and OOo and the last piece in this was our german PR
> which asked for developers and which made you want to join the
> Mac OS X port of OOo. But to show our good will it may be a good
> start to share experiences and to have a person without a past
> to mediate.

Well, someone nominated me. But whoever it becomes, they will need a vote of 
confidence from both communities.

I personally think this is a very good moment to consider cooperation plans. 
OpenOffice 2.0 and NeoOffice 1.1 are both nearing completion and release. New 
people have joined in the effort who don't have bad feelings from what has 
happened before and who have displayed great energy and enthusiasm. Now, both 
from an engineering and a process perspective, the stars are right.

The Mac is (according to Apple) the largest Un*x OS in use, with about a 5% 
market share (AFAIK, give or take a %). While I don't know hard figures, I'm 
pretty confident it is a lot bigger than Solaris. So there is a small but 
significant chunk of advanced desktops at stake.

> But there is one problem all the NeoOffice/J code is GPL and
> therefore it isn't easy to integrate this code in official OOo
> sources which are dual licecened LPGL/SISSL and need a signed
> JCA. The NeoOffice/J coders needs to give his permission that
> we are allowed to use the code unter the JCA conditions.

OK, the perennial sticking point...

Now, I am not a lawyer, nor a programmer (though I have at one time written 
assembly and I have read volumes of legal texts in my time), so please bear 
with me. I'll give this one a shot:

If you take away the egos and the principles, what exactly is the problem with 
including GPL code in the OpenOffice tree? As far as I understand the licenses, 
OpenOffice.org could use GPL code and publish NeoOffice here without any 
problem.

I seem to recall the problem comes when a company wants to use the GPL code and 
earn money from it. Then they would have to ask Patrick and Ed for permission, 
and Patrick and Ed will probably ask to be compensated for the 3000 manhours 
and 4000 dollars they've poured in so far. While the other 99% is OpenOffice 
code, nearly all of that has been written by paid employees, who have therefore 
already been compensated.

Considering this is not Linux, there will not be a large number of companies 
trying to do this. In fact, I would not be surprised at all if Sun were the 
only one. Which would mean Sun pays a wee bit (for them) of money to Patrick 
and Ed, and off goes StarOffice for the Mac...

Because Patrick and Ed would like to see their work used widely, I would be 
surprised if they tried to choke Sun with an excessive amount. Because Sun 
would get a lot of code, fairly ready to go and would need them for extra 
support and still turns a profit, I would be seriously disappointed if they 
tried to cheat Ed and Patrick.

Now I see only one snag here: someone else contributes and they have to split 
the money or negotiate together. I know a lot of people have contributed time 
to this product and some might also want their due. This would be tricky, and I 
have no sane way of determining how contributions should be rated. Currently, 
Patrick is solliciting grants to implement certain features (which is slowly 
lifting off). That might work here too. But we are now considering worst-case 
scenarios. Someone with better entrepeneurial or legal skills than me will 
think of something..

Again, I might be wrong here. But if everyone displays a modicum of trust, I 
don't see the problem for OpenOffice.org releases. Even without being able to 
spin a commercial derivative, having a pretty Mac version (something I once 
referred to as a duck) would boost adoption of OpenOffice on other platforms. 
This simply due to inclusiveness because its formats and interface could be 
easily used on another platform.

REMINDER: Please flame in a civilised manner. I repeat that I am neither a 
lawyer, nor a programmer and I am trying to be as bluntly unemotional about the 
whole thing as I can be.

Bringing NeoOffice and OpenOffice back together will also be greatly helped if 
OSX becomes a tier 1 platform. Patrick has complained vehemently about this in 
the past and cited it as one of the reasons to fork: all sorts of things change 
constantly and break OSX builds. I saw complaints here as recently as this 
month that OSX patches still need to be fixed between milestone builds. This 
puts and awful strain on the time and resources of volunteers.

Given the new enthousiasm for Mac porting (which has made me and everyone at 
NeoOffice very happy - NeoOffice is nothing without the X11 version and full 
aquafication will take many bright hands), I would like to ask OpenOffice.org 
staff to reward these efforts and ensure that change to code be checked for 
breaking OSX builds. This would probably boost morale for the volunteers 
working on OpenOffice.org. Of course, adding paid Mac   staff would be even 
nicer (hey - a guy can dream..).

In this area I am slightly worried that good intentions are being chained down 
by lack of resources. Specifically, issue 11004 (meta key awareness) has been 
sitting and waiting for two years. This would facilitate Mac keyboard support, 
but has apparently been dropped for 2.0. This stands in contrast to the upbeat 
sounds I keep hearing. I haven't seen the engineering explanation, so I an not 
going to condemn that decision, but it is a disappointing sign.

This bit is not intended as criticism of the hard won gains in OpenOffice 2.0. 
I am just asking for consideration for the volunteers' efforts.

As I write this, I am wondering if a sort of development plan should be 
conceived, one that both communities (hopefully soon unified) can work towards. 
I would never presume to tell anyone what to code, but a shared goal might help.

>From an engineering perspective, the experts tell me that a hybrid of Java and 
>Cocoa code would mesh relatively well with both the VCL code and the Mac 
>frameworks. I am curious to see who would back this and be interested to 
>develop, build and/or test this. A lot of stuff is actually already present, 
>if issues mentioned above (and a few others) can be worked out. If someone 
>feels this is not the way to go, I challenge them to show how they would make 
>it work instead.

>From a process perspective, some sort of cohabitation agreement would make 
>life easier. Unless someone wants to go through a messy divorce and rebuild 
>their codebase, NeoOffice and OpenOffice are joined at the hip (please recall 
>those figures of 3000 man hours and 99% before arrogantly dismissing this 
>argument).

To finish the awfully long post: At this point, all of the work on both 
NeoOffice and OpenOffice.org for the Mac seems to be done by volunteers (with 
some support by committing patches from OpenOffice.org staff, AFAICT). I would 
like to hear from them first.

I hope there will be no replies staing blunt disagreement without offering a 
better solution. We are all intelligent people with one general goal. Let us 
work towards it. But feel free to point any errors in my line of reasoning :).

Best wishes,
Oscar van Vliet

(who is musing that the entry into force of the Kyoto Protocol
should demonstrate that common ground can be found almost anywhere)

__________________________________________________________________
Switch to Netscape Internet Service.
As low as $9.95 a month -- Sign up today at http://isp.netscape.com/register

Netscape. Just the Net You Need.

New! Netscape Toolbar for Internet Explorer
Search from anywhere on the Web and block those annoying pop-ups.
Download now at http://channels.netscape.com/ns/search/install.jsp

---------------------------------------------------------------------
To unsubscribe, e-mail: [EMAIL PROTECTED]
For additional commands, e-mail: [EMAIL PROTECTED]

Reply via email to