Ian Lynch wrote:
On Thu, 2008-04-24 at 14:50 +0100, Alex Hudson wrote:
argument though, either. If you're selling a service
rather than software, the software immediately becomes almost
irrelevant,
That depends on volume. If I sell 2 million laptops and save $5 each on
Windows licenses it's still 10 million and the margin on these machines
is going to get smaller and smaller.
Well, you've hit the nail right on the head there as far as OLPC is
concerned: they're not selling millions of machines. The $5 saving of
Windows tax is illusory until you actually sell the same volume: if I
make $10 a unit and sell 2 million, and you make $15/unit but only sell
500k, I'm still ahead and I have four times more customers.
I think what OLPC is discovering is that there are sufficient customers
who want Windows that the dent in their costs is likely to be more than
offset by the increase in volume they can ship. That's speculation,
though, I admit.
On that scale I might well be
better to design and source my own machines. I think Mandriva is
profitable and it does Linux desktops. Not sure about Canonical but they
have enough reserves to keep going for quite a while. While there is
Linux desktop development I might as well tap into it.
Well, I agree with that (to the extent that you realise you're dependent
on the investment of others ;), but again, it doesn't matter how cheap
you can get it unless customers want it, and I think that's the problem
OLPC have bumped into.
and that's basically what has happened to the OLPC project,
except that there 'education' supplants software, to seed the kind of
self-learning that 8-bit micros did. They've just realised that you
don't need free software to do that.
You don't if the software license costs come down to zero or very close
to it. Free software is likely to be instrumental in making that happen
though and then as stuff moves to the web the desktop OS will become
irrelevant anyway. I think that is probably worse for MS than for anyone
else.
Possibly. I'm not sure; I think that would require a much greater degree
of commoditization than currently exists and is likely to exist in the
next 10-15 years. I don't think price is the issue at this point; you
basically can't give a free desktop away at the moment.
Cheers,
Alex.
_______________________________________________
Fsfe-uk mailing list
Fsfe-uk@gnu.org
http://lists.gnu.org/mailman/listinfo/fsfe-uk
