incubator-ooo-dev mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From Louis Suárez-Potts <lui...@gmail.com>
Subject Re: questions about the "porting" project
Date Wed, 16 May 2012 20:51:05 GMT

On 2012-05-16, at 16:07 , Kay Schenk wrote:

> well I am not quite THAT bored at the moment. ;)
> 
> Thanks for all this. Yes, it did help. Our current situation, as with any open source
project, is that you can only *build* what you can sustain.

More or less, yes: the no part being, I would maintain, that as long as one is honest about
what is being done and can be done, then what counts as "sustainable" can, arguably, be relaxed.
After all, the original Mac OS X builds maintained by the community members numbered two and
had day jobs. The effort, fuelled by coffee and IM, was in a way not sustainable at all; but
it inspired and proved a point, and so in the end, became sustainable. A key reason? Not the
execs at Sun who liked Macs—nearly all—but the actuality of a noisy market. That is to
say: marketing can help bridge the gulf between what is feasible by the resources at hand
and what can be done, given the needed resources. (A 'resource' is a person, here, whose salary,
in this case, is in effect, a debt paid back by the users and those who supply them with services
at a cost.)
> 
> Mostly I was asking about this to try to get a feel for what we should include as "official"
builds vs not.

It's a difficult question, and I do wonder: do we need "official' or simply a limit on the
size that can be held? When I set up the mirror system, I stratified it by "stable" and "contributor"
(or the like) builds. The "stable" would map to "official," but the point was that it related
more to builds that were *ready* to go than to builds that demanded privileged treatment because
they were "official." A ready-to-go build could be ready simply because it attracted the right
level of interest among the right sort of people, not because it had been deemed "official."
Yes, there will be a degree of competition.

There may also be—would be—confusion among users, esp. the big ones, like governments.
In this case, I'd suggest we have more of an argument to insist that they actually put their
money where their code is.

As to corporate contributors: they have their own agenda, and it probably is pretty much most
everyone else's. The point is that they will need to do what their clients want, no matter
what.

> 
> Considering Maho and Pedro (with FreeBSD) and Dario (OS/2) are involved with the project
as committers, why wouldn't we include these builds on the mirrors? And, we have a Solaris
participant as well now.
> 
I think that if their builds are ready, great. We *could* instate a simple requirement, that
Build A must have a roadmap leading to Build A.n+2, if not B. That is, two post-A releases,
but that is probably not necessary. It's only put there, as a suggestion, to give users and
contributors a sense of where to allocate their own energies. 


> A further discussion I think. I would think any "ports" by AOO committers would at some
point, be part of the official builds. But more to follow...


Good; yes, this is a worthy discussion, and I wish we could have had these at the old OOo
community council. Certainly, many of us wanted that. But [redacted].

cheers
Louis
Mime
View raw message