In <412DFD96(_dot_)104(_at_)solidmatrix(_dot_)com> Yakov Shafranovich
<research(_at_)solidmatrix(_dot_)com> writes:
The following two messages are from Eben Moglen of FSF and Larry Rosen
of OSI. I have received permission from both of them to forward these
messages to the list. And yes, they are both lawyers.
Ok, in light of opinions express by these two knowlegable lawyers that
the SenderID license is not compatible with open source licenses, the
GPL, AFL and OSL in particular, I now feel comfortable saying that I
strongly oppose the advancement of the marid-core and marid-pra I-Ds.
They can not be deployed in too much of the existing email
infrastructure on the Internet. I also oppose the advancement of
marid-submitter on the grounds that it is irrelevant without the other
two I-Ds.
In <868C8B09-F772-11D8-9196-000A95B3BA44(_at_)hxr(_dot_)us> Andrew Newton
<andy(_at_)hxr(_dot_)us> writes:
On Aug 26, 2004, at 8:54 AM, wayne wrote:
Instead of addressing [the suggestion of an early decision on the
SenderID license], PHB has tried to reframe the discussion.
If you are admonishing Phil, your energy is misplaced. It is not your
place to decide the schedules of this working group nor how consensus
is arrived at in this working group.
You are right, it is not my place, it is *your* place. And you didn't
answer the question in your reply, I ask you again:
Can there be an early decision on the SenderID license?
While I think license issue is very important to discuss, the volume
is such that it is drowning out most other discussions about the
I-Ds.
In <44B4AB1E-F765-11D8-BD7E-000A95BC6A7E(_at_)margaretolson(_dot_)com> Margaret
Olson <margaret(_at_)margaretolson(_dot_)com> writes:
I largely agree with Phill, but let me try to present the issue a
slightly different way. I use several MTA implementations, both
commercial and open source. I've concluded that the current license
will allow me to continue to have a good selection of open source and
commercial choices with Sender ID implementations. Open source is open
source - Sender ID is sufficiently valuable that I'm convinced that
viable Sender ID variants will emerge, even if they are not
distributed in quite the same way that the code is today. Both Eric
Allman and Phill have discussed ways to approach this problem. I am
speaking as someone who intends to implement receive side Sender ID
as soon as possible.
I would be interested in learning on what basis you think that open
source developers will risk violating the open source licenses in
order to support SenderID.
Other than Eric Allman, I know of no other open source developer who
is willing to even do the work needed to support SenderID with the
current license. Who do you think will do the required work? Are you
offering to pay for the development they way Meng offered for SPF?
(Mind you, Meng didn't end up having to pay much since, unlike
SenderID with the current license, many open source developers
volunteered.)
-wayne