“Judge Jackson’s relevant market is therefore precisely the
market Sun et al don’t think will be relevant in a couple of years
time. If they wanted Windows source access they’d want it largely
for use in products that are outside the relevant market, so from
the point of view of injecting competition into the relevant market
the judge should conclude they’re not relevant. Got that?”
“And although our friends in the open source movement could
be in the relevant market, they’re also outside it. By defining
Linux as largely a server OS Judge Jackson defines it as largely
outside, and understates its capabilities as a future desktop
competitor. But Linus Torvalds himself has been taking about a
future of lower footprint, appliance-style implementations. He
seems to think Linux Everywhere will happen to a great extent via
new platforms that obsolete the old Wintel model, and he’s hardly
alone in taking that view.”
“If open source’s future also lies outside the relevant market,
then it’s not a neat solution to Jackson’s highly focussed version
of the problem either. Is it possible to maybe promote a couple of
Linux distributions in order to build competition in the PC client
market in the real term? How could you do that without letting the
whole thing escape, and turning into something much bigger than the
narrow problem you meant to solve? What Linux distributors would be
willing to co-operate in such an exercise anyway?”