Re: [tlug] Re: Why Vista Sucks (was: linux: it's becoming ubiquitous)

["Stephen J. Turnbull" <stephen@example.com>]

Attila Kinali writes:
 > On Sat, 29 Mar 2008 04:23:40 +0900
 > "Stephen J. Turnbull" <stephen@example.com> wrote:
 > 
 > > It's not about the creators.  Even Jefferson knew that creative people
 > > do it because creating is what they do.  They'd die without it.  What
 > > is well-known but understood by almost no open source advocates
 > > without a PhD in economics is that IP is all about the capitalists.
 > 
 > Hmm.. i don't have a PhD in economics, heck, even worse, i'm
 > an engineer! So, could you please explain this a bit further,
 > so that people like me get an idea why everything we do is in vain? ;-)
 >  
 > > The only way to get VCs interested in open source development is going
 > > to be building a successful business around it first.  That's a lot
 > > harder for all concerned than going to them with a patent or a
 > > copyrighted program.
 > 
 > That i already know. OGP is basically blocked by this.
 > We need money to build a working business, but cannot get
 > anyone even lending some without a working business.
 > (see also "dead lock")

I explained what you asked, you say you already know it, and you still
don't get it?

The point is that this deadlock gets broken all the time, in fields
where there are property rights in what's being sold.  Free software
is giving away the store, without replacing it with anything a sane
3rd party would want to invest in.

 > > As Mark Shuttleworth pointed out bluntly at BALUG last week, whatever
 > > else you say about Microsoft, there's one thing you can't take away
 > > from them: they made software cheap for the masses.  This is not true
 > > of open source, not yet, and maybe never.
 > 
 > What is your definition of "cheap software"? Respectively, how
 > do you measure the cost of software?

"Cheap" is a price at which people are willing to pay for on the order
of a billion units.  Bill Gates did not become a multibillionaire (and
make a few others only an order of magnitude less money) on margins of
$1,000,000 a unit, you know.

 > Yes, i know i will sound like a OSS zealot, but if it doesn't
 > get implemented in OSS, then the interest in this particular
 > feature isn't big enough. If the interest would be big enough
 > someone would actualy do it, either because he needs it or
 > because someone pays him for it)

That's a reasonable point of view for the developer, but it's just
plain wrong if you've got a million people willing to pay $1 for a 1
man-year project, but no way to collect from more than 10,000 of them.

If you want to fund projects on one transaction per project, go into
real estate.  In software you have to sell lots of units at small
margin over media costs just to cover development costs.  If you need
to sell lots of units, it's an insane business plan to set things up
so that you have a million competitors giving it away, driving both
your margins and your unit sales down.

 > >  > Of course, under the condition that the bussines model works.
 > > 
 > > Which is an extremely strong condition.  Again, at BALUG Mark S said,
 > > "we don't know how to design working business models for open source
 > > yet."
 > 
 > Yes, but it's a strong condition no matter for what kind of
 > thing you try to make a working business model.

Sure.  But compared to developing and selling proprietary software,
developing and selling open source is the equivalent of tying one hand
behind your back and then using the gun in your free hand to shoot a
hole in each kneecap.

There are known business models based on open source.  There just
aren't any yet known in which lots of people get paid for developing
it.  Eg, you'd think that if anybody deserved to get paid for his
development work it would be Uli Drepper.  But when he gave a talk in
Tokyo a few years back, he claimed that all he got from Red Hat to
support his glibc maintainership was free coffee; he still had to show
up at work in the morning to do his paid job.