Re: polarization

[Havoc Pennington <hp redhat com>]

bruce perens com (Bruce Perens) writes: 
> The developers do technical direction. The foundation is mostly concerned
> with nontechnical issues and is representative of the developers, but it
> doesn't do things the developers want or need to concern themselves with
> regularly.
>

The board should be concerned with decisions and issues that require
coordination and vision across all of GNOME, such as GNOME
releases. This is in the charter as its job.
 
> The question is whether we want to be a hacker's tool or the world's
> desktop.  If we're just a hacker tool, no money is fine and we can
> depend on begging for individual items to be donated every time we
> need something. I hope you don't really think that is what
> "independence" means.
> 
> I think we are trying to take GNOME to the next level, and of course not
> everybody is bought into that - many people are content to see the developers
> make the system just for us hackers to run.
>

Believe me, my #1 concern is how to go from hacker tool to world
desktop. I have a lot of thoughts on that. We have advice from
everyone, from GNOME users to CEOs. I do not agree that a money-laden
nonprofit is the answer or the most important part of the answer. That
would wildly misconstrue how open source works and what our advantages
are vis-a-vis proprietary systems such as Windows. 

If we were going to avoid all begging, as I said, you would need a few
hudnred million endowment. That won't happen. Which means that instead
of begging for specific things, we'll be begging for annual membership
fees from companies. Which means no independence.

If companies threaten to withdraw from GNOME today, we lose their
future contributions of developer time, but the project itself keeps
going fine. If they threaten to withdraw and we are paying people's
salaries with their fees, that's a bit more of a problem.
 
We've ended up with a compromise: _small_ annual membership
fees. Which means we can afford some incidentals, hardware, travel
costs, and maybe keep a saving account for legal defense, but we can't
do the kind of spending that would keep the board very busy thinking
about it. And when we can avoid spending on incidentals by asking
people to pay for them, we do that too.

I think the primary thing we have to do is to create the largest
single-vision best-coordinated free software project ever, and use
that to compete with them in technical excellence. Then we should let
companies figure out how to use the technical excellence in
strategic/political/financial ways to grab marketshare from
proprietary desktops. Because that's what companies do, they grab
marketshare. And indeed customers are always going to be buying a
product from a company; the majority of software customers would never
consider simply downloading an unsupported product off the net.

> > The board is not making technical decisions, but it is making sure such
> > decisions get made, and that they get communicated well between relevant
> > developers.
> 
> So we've created a nontechnical management team to make sure the developers
> do their jobs. Nobody's going to resent _that_, will they?
>

You are twisting words. There's a difference between making people do
something (which we can't do) and better-organizing the work
developers already do trying to coordinate the project. We already try
to have release plans, policies (equivalent of the Debian policy you
created), and so on. But people are focused on their own corner of the
world, and lots of this stuff gets skipped or poorly communicated. The
board can help a lot. The interim committee has already been a big
help; just getting a group of people at different companies to talk
every couple weeks is a big help. It's a matter of writing down and
communicating what's going on, that's the core of it. Any company
software team does this, and GNOME isn't magically different. Free
software is not pixie dust.

> > We want to coordinate the technical efforts of the involved individuals
> > and corporations;
> 
> It doesn't sound to me as if the process you intend to fix is broken. Excuse
> me, but I have a sinking feeling about this.
>

It's definitely broken, in the sense that there's lots of room for
improvement. There are problems like this in the project every
day. People have trouble getting involved because they can't figure
out how (GDP has some good solutions here); the bug tracker sucks and
there are several of them; AbiWord and OpenOffice need help knowing
the other exists and what to do about it; people don't know the
release plans (heck, we don't even have very good release plans); etc.

> You will need to be able to defend yourself in court, and that takes money.
> Your coffers aren't available to be taken in a lawsuit if the organization
> is structured properly, but they are your legal defense fund. With all of
> the software and business system patents that are happening, we will need
> to be able to defend GNOME from people who would make it impossible for us
> to use a user-interface element because they've filed a patent on it. We
> can't even fight unenforcible patents if we can't afford to go to court.
> 

This is a crisis situation that could come up, not the primary task of
the board. If the lawsuit involved some crucial element of the
desktop, then of course Sun and Red Hat and so on will have to
contribute; they do have to ship a desktop, and in fact already have
shipped it, and in fact are probably going to be the targets of the
lawsuit for that reason. So if someone sues saying we can't have a
panel or can't use the mouse, there's going to be enough money to
defend or these OS vendors are screwed, because GNOME is their
desktop. I don't think we should stay up nights worrying about the
money to defend. Yeah I'm worried about the lawsuit's possible
existence, but there's nothing much we can do about that, aside from
getting patents ourselves (worth considering); but I'm not worried
about having defense money.


The basic point being: the board should have a lot of developers on
it, and be part of the developer community. It should work to make
sure GNOME runs well, and help the developers decide what they want to
do about companies that ask to help, for example. The foundation might
save incidentals money, or a legal defense fund, but ultimately that
isn't it's primary purpose, or shouldn't be IMO and my impression is
that we didn't create it for that reason.

Havoc