Sublicensing isn't an merely because MS is big company with lots of
lawyers. The IBM Public License (see below) that they use for code
that they release to open source projects is sublicensable.
It is critical that a code license be sublicensable because otherwise
there is no way to modify the code and to pass the modified code
on together with the original. The need to sublicense is a direct
consequence of the need to merge two work products.
This is not a code license, it is a patent license. Confusing the
two is mistaken.
With a patent license there is no way for a third party to extend
the patent scope. If a fresh patent issues then the result is a
separate patent with separate patent licensing terms. The work
products are distinct.
So why is
it critical here? The reason you don't allow sublicenses of a free
license is to preserve your ability to change the terms in the future.
I'm not weaving conspiracy theories here, I'm just stating the
obvious.
This has no effect on the ability to change the terms. The reason that
sublicensing is prohibited is that this could be used to avoid the
granting of reciprocal rights.
So the question is whether we're willing to use a license that
Microsoft has told us that they're considering changing.
Where do you find that statement? I don't see it.
Without
attributing any evil intent, I find that much too risky simply because
I don't know what situation Microsoft will be in three or five years
from now, and neither does anyone else.
Once the license is granted it is granted.
As someone who is constantly
writting twiddles to mail software and exchanging them with other
people (qmail patches, mostly) this presents a huge spanner that could
be chucked into the works at any time.
Given that the courts have in the past revoked patents after the
holders failed to disclose them under 'note well' type rubric
it hardly seems likely that Microsoft can tell the working group
that it has IPR, that it is available on license terms X and then
subsequently change the terms to anything that is less favorable
(in the Pareto sense) than X.
The same issue is at work here, detrimental reliance.