Re: Dave Crocker's Appeal - goal and request for positions

[Bob Hinden <[e-mail]>]

Christian,

Thank you for the opportunity to participate in the review of the
complaint regarding the ONCRPC working group.  I have been involved in
this matter while I worked for Sun Microsystems and think I could
provide some insight into what went wrong.

Overall I think that it is very important that the IETF be able to
accept technology developed outside of the IETF.  This ranges from
complete protocols such as XDR and RPC and to specific techniques
which are patented.  If the IETF is unable to accept outside
technology it will be doomed to try to recreate everything.  An
official "not invented here" policy would be very unfortunate.  In the
case of patented technology given the rate that industry is patenting
network related technology, the inability of the IETF to create a
reasonable agreement with a patent holder could stop all forward
progress in all of the IETF working groups.

There are a number of issues which I believe caused the agreement
between the IETF and Sun Microsystems to get into trouble.  They all raise
issue with our current process.  These include:

1) The text in RFC1622 which defines the conditions under which the
ISOC can accept outside technology is not workable.  It was written to
completely protect the ISOC and is in my opinion unacceptable to all
commercial organizations.  A donor is required to assume blanket
liability for the ISOC.  This goes well beyond what any donor would
accept.

Until the RPC/XDR agreement was written, the procedures covering this
topic in RFC1622 had never been used.  The RPC/XDR agreement is the
first implementation.  I think that the text in 1622 should be revised
based on what was learned in the implementation process.  Assuming the
agreement with Sun is completed, it should be used as input to this
process.

2) There is not clear owner of an agreement in the IETF/IESG/ISOC who
has the authority to negotiate with a technology donor.  In the case
of RPC/XDR, Sun Microsystems was told at different times to negotiate
with a number of different people.  This included the IESG executive
director, ISOC lawyer, and IESG Chair.  In each case an agreement was
written but was later rejected by someone else for different reasons.
I was personally told by the IESG chair on two separate occasions
(each with a new draft agreement) that he would sign the document.
This was later retracted.  Each time a new person got involved new
negotiations started.  This process seems to be continuing.  Now
Sun is negotiating with Vint Cerf, the ISOC president.  Certainly we can
not expect every agreement to be signed by the ISOC president.

An agreement needs to have an owner who is empowered to sign it.
It could be one of the above or someone else such as the
appropriate IESG area director (this would be my choice).  It needs
to be one person and a person who is willing to sign the agreement.
This person needs to have the time available to do this work.

3) The IESG as an organization did not want to take responsibility for
this agreement even after it had chartered the RPC/XDR working group.
They could not come to any consensus to sign an agreement or who
should sign.  The IESG chair also did not want to sign the agreement.
Everyone (except for Marshall Rose who did offer to sign for the IESG,
but the IESG rejected his offer) wanted someone else to sign it.
There seemed to me to be a great fear in the IESG about someone taking
legal action against them as an organization or as individuals.

This fear in the IESG is ironic given their non-action may make them
just a liable as if they had signed an agreement.  Sun Microsystems
might have cause to claim damages.

The ISOC was supposed to have provided liability insurance for the
IESG.  This has not happened.  Either this needs to be done quickly or
the relationship between the IETF and the ISOC needs to be reviewed.
It is unclear to me what value the ISOC is providing to the IETF.

My overall conclusion in this matter is that the formal complaint
filed by Dave Crocker is valid.  The IESG did mishandle the Sun RPC/XDR
agreement.  The IESG did not fulfill its contract with the community it
made when it chartered the RPC/XDR working group.  This includes the
IESG as a whole, the area director who was responsible for RPC/XDR
working group, and the IETF chair.  The IESG needs to be held responsible
for its actions (and inactions).

Robert Hinden