In <792DE28E91F6EA42B4663AE761C41C2A02BEBDB2(_at_)cliff(_dot_)bai(_dot_)org>
"Ryan Malayter" <rmalayter(_at_)bai(_dot_)org> writes:
[wayne]
The possibility that requiring every single user of qmail to get a
signed license from Microsoft is a reason for concern.
IANAL, but it is quite clear to me that the MS license requires nothing
of the sort:
I think it would be really cool if you are right. A qmail patch is
basically taking a hunk of source code and installing it into another
program, possibliy modifying the patch code in order to make it work
with the modified program. So, if I, as a small ISP or whatever, want
to add SenderID functionality to a specialized program that I have,
all I would need to do is take the qmail patch, modify it as needed to
make it work with my specialized program, and not have to worry about
a license. Or, at least not have to worry unless I decide to
redistribute it.
However, my gut feel says that MS has not intended such a thing.
So, a qmail pactch consisting of source code containing the PRA is a
source code implementation. Creators of such soruce code implementations
will need to sign the MS license to include PRA, but *users* do not.
Can a qmail patch be a licensed implementation? I don't think so.
In particular, take a look at section 2.2 again:
2.2 _Source_Code_Distribution_ You also have [...], a [...]
license to distribute or otherwise disclose source code copies of
such Licensed Implementation licensed in Section 2.1 only if You
[....]
This seems to say that a source code distribution can only be a
Licensed Implementation if you can take the source code, compile it
into object code and have a Licensed Implementation as per section
2.1.
The trademark requirement for what is considered a "licensed
implementation" seems a bit onerous. But it says NOTHING about it being
a "registered trademark", and mentions no jurisdictional requirements.
So make up a brand name for your qmail patch, and you have a trademark.
Yeah, I agree, trademarks, are granted automatically. I raised the
issue because it again points to my belief that MS is intending this
license to be applyed to a classic software product, that functions
and does something useful, not just random snip-its of code.
I'm not even certain that a library that contains the PRA algorithm
would qualify as a Licenced Product because you must implement all
required parts of the SenderID specification. I believe that would
include things like the SUBMITTER functionality, which may mean that
you can only have MTAs as Licensed Products. I haven't thought about
that part very hard though, take it with a grain of salt.
-wayne