Brian E Carpenter <brc at zurich.ibm.com> writes:
Did they think of *talking* to the IETF before these changes?
You must be kidding. When you manage something as complicated as an
operating system, you have other things to do than to engage in
lengthy legal discussions (not to mention checking everything with an
attorney) with people when you have already read their legal
texts.
As you may realise, companies managing something as complicated
as a proprietary o/s do exactly this because they have to, in order
to stay out of court or to succeed when litigation occurs. I can see
that this is much harder for an open source project, but I really
find it hard to believe that FreeBSD has no lawyers.
Many vendors derive their getaddrinfo man page from the RFC. I
compared getaddrinfo(3) on my Solaris 5.9 machine with RFC 2133, and
it re-uses a considerably amount of text. The man page also modify
the text in several places.
I think there is agreement that the vendors have no legal rights to do
this. At least nobody has been able to disprove this so far in this
discussion.
The FreeBSD people noticed this legal issue, read RFC 3667 and then
contacted IETF to determine if their interpretation was correct, and
it was suggested that it was. With permission, I'm including portions
of the e-mail exchange that occurred on the IESG or IAB list last year
below.
I believe the e-mail below prove that people DID contact the IETF
about the problem with the copying conditions on the getaddrinfo man
page.
I think your view of the situation (as quoted above), given the facts,
is rather twisted. FreeBSD did the license analysis correctly, and
other vendors, including big companies, dodged the issue.
Thanks,
Simon
Date: Mon, 27 Dec 2004 14:56:43 -0500
From: Rob Austein <sra at hactrn.net>
At Mon, 27 Dec 2004 19:16:58 +0900 (JST), Jun-ichiro itojun Hagino wrote:
quick question.
it seems that many vendors reuse RFC text as manpages (such as KAME
(fixing), Sun Solaris). is it permitted? i see no text that
permitting reuse in ISOC copyright boilerplate, so (taking "default
is not allowed" logic) i think it is not permitted.
i think you are correct, it is not permitted.
see rfc 3667 for details, but the short version is
a) man pages would be a derivative work;
b) in general (see caveats in 3667) the ietf and isoc get a
perpetual irrevocable worldwide non-exclusive right to create
deriviative works; but
c) nowhere do the ietf rules say or imply that the ietf/isoc rights
(b) are transferable to or apply to any other party.
so ietf/isoc is allowed to create derivative works, but anybody else
being allowed to do so is a separate matter beyond the scope of the
ietf rules. since the ietf/isoc rights (b) are non-exclusive, the
original authors can grant similar rights to other parties should they
choose to do so, but the ietf doesn't require them to do it.