GrepLaw |
|
|
This site is a production of the Berkman Center for Internet & Society. Please email if you have questions, contributions, or ideas about improving this site.
|
|
|
|
| |
|
|
Patrik Faltstrom on IESG, IETF and ICANN
|
|
|
|
posted by mpawlo
on Tuesday November 12, @05:31PM
from the brain-picking dept.
|
|
|
|
|
Patrik Faltstrom humbly describes himself as a short person from Sweden interested in communication and information. However, Faltstrom has a lot of insights on IETF, IESG and ICANN. Greplaw has picked Patrik Faltstrom's brain.
|
|
|
|
|
|
# Who is Patrik Faltstrom?
A short person from Sweden interested in communication and information,
and especially the combination of the two -- on a global scale, which
implies the need for internationalization, efficient communication
mechanisms; new and old. Among old ones, sailing and socializing with
the help of food. Among new, Internet based protocols.
# What is IETF and why should we care?
The IETF is a process where rough consensus decide what standards are
to be favored. Nothing is said about what is later used, but the way
Internet works, one only succeed if what one do fit with what other
people do. Cooperation is essential. Proprietary solutions do not
survive. Because of this, implementing what is developed in an open
process according to consensus is a low(er)-risk case than doing
everything on your own.
# How is a new standard implemented?
Two answers:
(a) The standard is specified by publication of it as an RFC. There are
rules for what review the proposal must have got before the status of
the RFC is chosen (by the IESG). An RFC has a number which is never
reused. It can be replaced or updated by later RFC's, but never deleted
from the archives.
(b) A standard is used by having vendors implementing the standards in
their products. That normally happens when customers ask for features
which the standards make possible. Without interest from vendors and/or
customers of vendors, standards will never be implemented -- and there
are many RFC's specifying such.
# Can you explain the decision making process?
A working group have a certain number of tasks. The tasks are agreed
upon between the participants in the wg, and approved by IESG and IAB.
The tasks are listed in a so called "charter" for the working group.
When the working group have consensus for a proposal, they pass it to
their appointed Area Director which in turn take the proposal to the
whole IETF and request comments. The input from this world-wide "last
call" is going to the IESG, the board of the IETF, for decision.
It should also be mentioned that the working group is working on a
mailing list, and not at physical meetings. Even though working groups meet now
and then, it is on the mailing list consensus have to be reached.
# What is the "humming" about?
Sometimes when a working group do have a face to face meeting, it is interesting to see what the view of the audience is. Humming can therefore be requested now and then. The idea is that one can listen to the strength of the hum, without knowing who hums.
# Some say the IETF should be a more democratic institution - do you
agree?
Every organization / process can be improved. We (IESG) is working very
hard on improving the process at the moment. One example is the online
access to the internal IESG decision making process which was made
public early November 2002.
# Is "capture" of the IETF by one major stakeholder likeley? Why / why
not?
No, not as long as the participation is so cheap / easy. What has
happened during the years is though something similar, domination of a
certain kind of people. Early it was "researchers". Later "Internet
Service Providers". Now "Researchers" again. This is a problem, and at
the moment operational experience is something we would like to have
more of. That would help creating standards which are easier to deploy
in the real world.
# What is the Internet Engineering Steering Group, IESG?
The IESG is the group which decides what status a document gets (if at
all) when an RFC is created. See above.
# How did you become an area director and what do an area director do?
I was an active participant, wrote some RFC's and showed interest in
the IETF process itself. So, I was nominated to become an Area Director.
What I do...read email, more email, and then for dessert: email.
An Area Director do a couple of things:
(a) Help working groups in my area reach consensus and create good
standards.
(b) Resolve process issues inside working groups.
(c) Read all I-Ds (from all Areas) and make up my mind what status the
RFC should get (if any).
(d) Liason with external (non-IETF) organizations.
(e) Answer questions from people, do presentations, and do interviews
like this one ;-)
(f) Have a normal day-job.
# How can I participate in IETF work?
By looking at IETF.org and subscribing to a mailing list of
a working group which have a charter which interest you.
# What is the nature of the relationship between IETF and ICANN
IETF is the technical body which is, by creation of RFC's, creating the
technical protocols and rules ICANN (which includes IANA) have to
follow.
# You have been around ICANN for a while. What benefits and problems
do you
see with the current regime?
I see two problems:
(a) ICANN should listen more to people.
(b) ICANN should try to do as little as possible.
Regarding (a) I see many people (including IETF) claiming ICANN do not
listen enough. The later proposals for reform is better, but, as
always, acting slowly make people irritated.
Regarding (b) I think ICANN should try to see what they have to do, and
then start with a scheme for how those things should be done.
As you say, I have been around ICANN, especially before ICANN, in what
was called POC. I know the questions above, (a) and (b) are hard to
find answers to.
But, ICANN just have to.
# Can it be fixed?
Maybe. I think it just have to. It is easy to be pessimistic given the
hard problems facing ICANN.
But, by minimizing the problem space ICANN work with, and for example
maybe only try to look at policy around creations of new top level
domains, success can be reached faster. I do not think ICANN should
become (intentionally or unintentionally) a kitchen-sink for all
Internet-related issues.
# What kind of legal problems do you think the Internet has to deal
with?
(a) Patent issues. As I said before, collaboration and cooperation is
essential for success. If you create a protocol or application which
imply other people have to pay a license to implement or interact with
what you have done, you will loose. The problem is that during that
time, users on the Internet are the one loosing. They implement and use
what you have patent on, and do not see the forest for the big tree
which is standing in the way.
(b) Convergence. When traditional applications and services now are
starting to use the Internet as distribution mechanism, traditional
entities charging for the transmission of information have to change
their business models. For example, we have seen for music problems
with digital copies of CD records, because the music industry is used
to charge for the plastic the CD is made of, and now they have to
change to charge for the music instead. Another example is voice
telephony which traditionally is a service the caller pays for, and
money is moving along the call all the way to the terminating telco. On
the Internet money moves very badly globally, and the charging model
for email is instead that one pay for both incoming and outgoing email.
My view is that first of all Information Want To Be Free. The
information will choose the mechanism to be distributed which is
"cheapest" (which include efficiency, quality etc and not only money).
Secondly, the billing model for voice phone calls will change to be
like the one for email -- which in turn means telephony no longer will
be anything special.
# Should the end-to-end principle be altered in order to address
concerns
of security and copyright?
Not altered, but refined. It is actually something I work on together
with Fred Baker, previously chair of the IETF, and now member of the
IAB as well as chair of ISOC. The idea is to talk about "referential
integrity", and in reality the conclusion is that "end-to-end" means
different things on different layers in the protocol stack.
# You have no beard. That is very unusual for someone this active in
the
IETF. Shouldn't you grew a beard to be respected?
Possibly, but my significant other would not respect me as much as
today.
I am trying to compensate by growing a ponytail.
# Finally, when and why will casting be a part of the Olympic Games?
It is approved by the IOK since around 1956 if I don't remember wrong,
but never taken up on the Olympic Program. It has been present as an
exhibition sport a few times. But, casting is something I stopped doing
around 1986, and I have only participated in some "show" competitions
after that. This after not being able to pay for all the travel which
was necessary in a small sport. I started with badminton and also took
up more serious sailing instead.
Patrik Faltstrom was interviewed by Mikael Pawlo.
|
|
|
|
< Voluntary Tax on Digital Media
| AOL Must Reveal Anonymous Poster's Identity >
| |
|