Re: Progress on AdaOS (Was: Re: How Ada could have prevented the Red

["Warren W. Gay VE3WWG" <ve3wwg@home.com>]

Marin David Condic wrote:
> "Warren W. Gay VE3WWG" <ve3wwg@home.com> wrote in message
> news:3B82789B.8D195045@home.com...
> >
> > Money would help, but unless some white knight shows up with a keen
> interest
> > in the project, I think you have to look elsewhere for motivation. I
> believe
> > that many Ada people would be willing to contribute to any AdaOS project
> that
> > is showing significant progress. You need achievable milestones, otherwise
> > the project just languishes.
> >
> I doubt there is going to be a White Knight who is going to say "I have this
> spare $100m lying around - would you guys build me an OS?" 

Agreed, very doubtful.

> However, if there
> was some carrot held out in the way of a future reward, maybe some people
> would see it as an opportunity to make a few $$$ on the side and devote more
> time to it.

It is easy to agree with this general principle, since none of us hard
working types (on our own time at least!) would deny ourselves some sort
of a reward, even if it were nothing more than to upgrade equipment and
software.  However, I think that in practice, this is going to prove
very difficult to achieve, depending upon the project goals.

>  Putting the OS out under the GPL (or some version thereof) might
> not hold out much hope of money. If it was being developed under some more
> "restrictive" license such that the developers might be able to make some
> $$$ from eventual sales, it might get viewed as a business venture with the
> developers investing in sweat equity.

The attraction of GPL (or licenses similar to it) is simply that your
efforts, and your fellow contributors efforts will always be available
to you for free. I can easily imagine a situation where I might want to
start wwgOS, and have others contribute to it, and myself retire from the
project someday (to work on other things naturally).  I'd still want to
be able to enjoy the efforts of the ongoing efforts, even though mine have
stopped (or moved somewhere else).

If you use a $$ model, then everything becomes more complicated. First of
all, there has to be a fair system of attribution worked out. With a
lot of contributors, this could be a flammable liquid looking for a
spark. $ has a way of changing the way people interact ;-)

> Or if it *must* be GPL (because of
> some religious zeal?) then doing the work under some corporate structure
> wherein contributors are given shares - that is holding out some hope of
> future financial rewards. (Think of the ACT model of selling support. First
> you have to have a product to support!)

I suspect that part of the reason Linux has taken off is due
to it's "free wheeling nature". The GPL is not bad when applied to
an operating system, since your proprietary application is not bound
by the GPL that the O/S might be (although IANAL). The GPL ensures
that all concerned, including the original contributors, have full
access to future versions of it.

It is tempting to say that corporations should pay for it, while making it
available to everyone else for free. Unfortunately, this kills most chances
of it sneaking into the corporation by the back door (as is often done for
FreeBSD/OpenBSD/Linux etc.)

So while I like the idea of $ rewards, I don't believe it will be compatible
with a goal like "AdaOS world domination" ;-)  A better approach might be
to win "world domination" first and then make money from the books 
that you can author and conference talks (since you'll be the 
expert on the subject.) Who knows, you might then later be able to join
a new company like "AdaMeta" ;-)

> > You then have the compiler you need, and you don't even have to fuss with
> > booting issues at this point. The issue at hand, as pointed out above, is
> > to get _something_ working.
> >
> Even if it doesn't have a boot loader and is just run as an "App" at some
> level, you then have at least something that is working. Even under Win2k -
> it could be a kind of "Simulator" program for a while until enough of it got
> built to make it worth doing a boot loader and having it take over the
> machine.

Indeed, you can work out the higher level concepts this way, and obviously
enjoy a more hospitable development environment.

> > I think the "leadership" issue will take care of itself, if someone would
> > just get started on it. But again, as a hobby, I cannot be critical of
> > someone who _wants_ to do a compiler. However, if the goal of this hobby
> > is the OS, then I think that the compiler is just an unnecessary
> diversion.
> > This could be done -after- the OS is getting somewhere. After all, the
> > compiler may need to be different anyway, if the OS is revolutionary
> > (ie. non POSIX in nature).
> >
> Well the "leadership" almost by definition is going to come from whoever it
> is that grabs the bull by the horns and gets something built. That's what
> will set the direction of the project and that individual is who everyone
> else will look to for further direction.

Absolutely.
-- 
Warren W. Gay VE3WWG
http://members.home.net/ve3wwg