An on-line copy of these and other IAB minutes are available from http://www.iab.org/documents/IABmins.
This meeting was devoted to an open investigation into Dave Crocker's appeal of activities, or lack of activities, by the IESG in relation to the work of the ONCRPC Working Group of the IETF. At the start, Christian Huitema described the scope of the inquiry -- in particular, that the technical merits of ONC/RPC or other technologies is not being addressed here. He then described the format, which included 10-minute statements from the involved parties who had been invited to submit a written statement in advance followed by questions from IAB members. After all of the statements, comments were taken from the floor, limited to two minutes and without repeat speakers. Christian also announced that the IAB would present its finding at the open IAB meeting scheduled for 5 April.
Dave Crocker, the initiator of the appeal, went first. See his slides included with these minutes. He closed with three recommendations:
Questions from the IAB:
Robert Elz: How should the IESG consult the IETF, when should it have happened
in this case, and what effect would it have had on RFC 1602? Answer: When
process hit a brick wall, the IESG should have come to the community via
e-mail or an open meeting at IETF. The community should have fixed RFC 1602
on the fly -- it was a broken specification.
Bob Hinden, who was employed by Sun till January of 1995 and was the negotiator on Sun's side until then, spoke next. See his slides. Bob suggested that we need to fix the specification (RFC 1602) regarding Intellectual Property procedures. He also asked where was the relevant area director during the negotiations and why didn't ISOC provide liability insurance in a timely fashion?
Questions from the IAB:
Robert Elz: What might the IESG have done to move this along? Answer: The
problem was a lack of ownership on the IESG side and too much handing around
of the problem. Also, whoever is doing the negotiation really needs to also
have the signing authority.
Christian Huitema: Should Sun simply have ``abandoned'' it via a public statement? Answer: Perhaps; Sun has been giving this technology away for a long time.
Steve Crocker: It is open ended as to why there needs to be a signed agreement. What did Sun need the agreement for? Answer: To meet procedures of RFC 1602, and also to transfer rights to the trademarks.
Next, Steve Nahm from Sun who took over the negotiations from Bob Hinden was invited to speak. He declined the opportunity to present a formal talk, and reported that the agreement is likely to be signed this week.
Question from the IAB:
Christian Huitema: Is Sun trying to maintain Intellectual Property Rights
(IPR)? Answer: Sun retains the IPR but licenses to ISOC the exclusive right
to make it an Internet standard, to evolve it. Sun will license the technology
for free once it gets to Proposed Standard status.
Dave Sincoskie: Would Sun have been happy to send a letter regarding IPR, rather than negotiating a contract with ISOC? Answer: Sun would not like to totally give away the technology; it would prefer to license it.
Dave Sincoskie: Would there have been a way to do it without anyone from ISOC signing anything, but instead simply sending a letter to the standards body? Answer: Sun was following RFC 1602, but would definitely have looked at some other way of doing it.
Steve Crocker: The essence is where has the time gone. How much time has been consumed by Sun, early on and more recently? Answer: The time gap between original suggestion and working group formation was Sun's and the Area Director's problem. The working group first met at the March 1994 IETF and was done with their work in May 1994. Since then, the time has been spent in trying to figure out what process to go through to complete it.
Raj Srinivisian, who works for SunSoft and was chair of the ONCRPC Working Group declined to present a formal statement. He commented that the ``problem word'' in the IPR section of RFC 1602 is ``warrant.''
Question from the IAB:
Brian Carpenter: The question has been raised whether two competing technologies
can both be IETF standards -- what does Raj think about this? Answer: There
is a place for competing standards. His hope is that there will be a boilerplate
contract that can make the process fast.
Vint Cerf, ISOC President, was not present. His prepared statement was read aloud.
The final presenter was Paul Mockapetris, IESG chair since March 1994. He observed that each legal iteration takes about three months, and outlined the time line for the ONCRPC effort. See his slides. He also observed that everyone is one ``rule exception'' away from happiness, but the question is who decides.
Question from the IAB:
Christian Huitema: Do you want to be forced to abide by RFC 1602? Does RFC
1602 document rules, guidelines or history? Answer: We need to make a template
agreement that will be acceptable to both sides in the future. The final
Sun agreement might provide a model for this. We also need ways to deal
with patent owners. Finally, we need someone chartered to do negotiations
based on these.
Elise Gerich: Do we need some sort of documents other than Informational RFCs to describe binding rules? Answer: Too much corner cutting was a problem.
Dave Sincoskie: Would a unilateral statement be enough? Answer: Yes, as long as the working group can use its judgment and we have a form that sets precedent.
Christian Huitema: Was the existence of competing technologies slowing things down? Answer: No. In fact, Paul was involved in no technical aspects of this matter.
There followed a series of statements from the floor:
Bill Noichian, editor of the original Informational RFCs about Sun RPC: At that time, Sun pretty much left it on the street corner. We need Sun RPC on the standards track so can base other work on it.
William Simpson listed the conclusions he would like to see:
John Stewart:
It is a bad model for the IETF to own anything, and for ISOC to be a holding
company for this. We need a clear definition of the rules.
John Tavs, addressing Mobile IP and patent issues:
``Fair and nondiscriminatory'' is difficult to describe. Cookie cutter agreements
are not going to be sufficient because different technologies have different
values and every case is different.
The IAB has been asked to review and respond on the timeliness of the procedures followed for forwarding RPC/XDR to the standards track.
The IAB invited representatives of the IESG, SUN, and ISOC to submit statements documenting their perception of what transpired. Then, during an open session at the IETF, the representatives were invited to present their positions to the IAB and the IETF membership. In addition, the IETF membership was invited to address this topic after the representatives had presented their positions.
After reading the submissions and listening to the presentations and comments made at the open meeting, it seems like the fundamental question being asked is:
(To quote D. Crocker)
``Does the process represent reasonable prudence or institutional paralysis?''
The conclusion of the IAB, and apparently of all the speakers, is that the defined process led to institutional paralysis. The text of RFC 1602, which defines how the IETF/IESG should operate, has a procedure for adopting external technologies which is unimplementable.
The process defined in RFC 1602 has raised deeper issues which warrant consideration. RFC 1602 is intended to specify the process to be followed by the IETF and related bodies. On the other hand, it was assumed that the procedures might require modification over time.
The working group process is the appropriate place to propose modifications to the procedures, with revisions appearing as Internet-Drafts, undergoing the Last Call procedure, and finally being adopted. This is necessarily a slow process and as such may result in an unnecessary obstruction of IETF business.
There must be a mechanism by which the IESG can obtain dispensation from the rules stated in RFC 1602 (or the next version) when documented procedures introduce unnecessary obstruction of business.
The IAB supports the recommendations of many of the speakers who proposed revisions to RFC 1602 which would:
The IAB also supports the recommendation to the POISED group, that the working group review other standards' bodies' procedures prior to revising those of the IETF.
Since the specific reason for this appeal appears to be resolved, it is unnecessary for the IAB to request any action by the IESG or any other bodies.
| Autonomous Networks | Deborah Estrin (USC) |
| End-to-End Services | Bob Braden (ISI) |
| Resource Discovery | Mike Schwartz (Colorado) |
| Privacy and Security | Steve Kent (BBN) |
| Libraries | Cliff Lynch (UCOP) |
April 4, 1995
7/93 ONC BOF Amsterdam
9/93 last update to 1602 ID
3/94 ONCRPC chartered
1602 published
4/94 Latest IESG / OSF legacy
CNRI agreement rejected by IESG
9/94 PVM task ed to deal with Bob Hinden (Sun)
12/9 4ISOC BOT zeros IETF $ ,
insurance scam 1
1/95 IESG Chair finishes with Hinden,
PVM delivers draft to ISOC,
Secretariat
2/95 Marshall/ Crocker social contract
Vint begins last edits
4/95 Today
Insurance policy still unavailable,
insurance supposedly there.