DNSO Mailling lists archives


<<< Chronological Index >>>    <<< Thread Index >>>

[nc-str] Structure - Past, Present and Future

Firstly apologies in advance for the lack of brevity but I wish to
explore in some detail the problems I have with both our current
structure and how the ALSC recommendations will work with that, and
finally some suggestions.

In *theory* the structure of ICANN is excellent.  A three way split
between domain names organisations, IP addresses organisations and
protocols organisations balanced by at large users.  Each SO gets
three board members and the at large nine.

The concept has always been that the grunty policy work will be done
in a consensus seeking manner in each SO and the Board process would
be not quite a rubber stamp but merely a check the proposal is not
repugnant and has followed proper process.

The reality though has been massively different.  In our own area of
the DNSO there has been almost no useful input into domain name
policy.  The Board adopted its own variation of the UDRP, the Names
Council was pretty much ignored on the Verisign variation issues, the
DNSO was told that even stuff such as changing the rules so there is a
registry/registrar split in *.com is not a policy issue for the DNSO,
the GAC gets listened to more than the DNSO on policy about domain
name exclusion etc.  This list is not to beat up on anyone but to
recognise that the status quo is not working at all well, at least in
some eyes.  The Board and the staff are deciding domain name policy to
the near exclusion of the DNSO as a formal body.

This has caused me to relook at the ICANN structure and how the board
is comprised.  The first issue which strikes me is why do the three
current SOs have equal votes on the Board?  It appears to date 95% of
Board time and decisions have been taken up by domain name issues,
around 4% by Internet addressing issues and maybe 1% by protocol
issues.  It appears to me that the DNSO is seriously under represented
on the Board and this could well contribute to why the Board feels
able to ignore the DNSO on so many issues.

Now looking at the DNSO itself one has seven constituencies each with
equal representation and one has to ask why are they equal?  There is
no logical reason to conclude that these seven constituencies fairly
balance out each other and represent the full range of stake holders.
Apart from the obvious missing individual registrants constituency,
there are many other constituencies you could have.

The Business Constituency is about to tighten its rules.  That raises
the possibility that there is room for another business aligned
constituency.  There is IMO an excellent case for gTLD registries to
be split into sponsored and non sponsored as their interests can be
very very different.  Going the other way one can make the case that
the Business, IP and ISP constituencies are all the same group of
people and should be collapsed into one.  In fact of the three current
NC reps from the BC, one works for a leading Intellectual Property
Organisation and the other two for businesses which are or own large
ISPs.  This is not an attempt to suggest any wrong-doing but merely to
highlight that one could credibly have anywhere from four to 12
constituencies and just giving them each three votes on the NC is not
necessarily logical.   

Lastly we have the GA.  I believe we have a chicken and egg situation
here.  The GA has absolutely no power to do anything.  Now many will
says neither should it because (in their view) it is dominated by a
few individuals who are seen as anti-ICANN or worse.

Can I put forward a view that if one did two things - both give the GA
some authority and also had a registrants constituency then the GA
would have a reasonable chance of returning to what I hope was its
purpose - a place where people from all constituencies could gather to
discuss and even come to consensus on issues.  The GA should be the
place in theory where the Registrars, Registrants and Registries talk
about issues such as transfers etc.

There is no incentive at the moment for the GA to work or for people
from constituencies to participate in it.  If you have an individual
registrants constituency much of the present debate could well
transfer to within that constituency and if for example one adopted a
bicameral approach and required policy to be adopted by both the NC
and the GA this would be a large incentive for people to make the GA

Anyway now onto the issues raised more directly by the ALSC and ccSO

The first and massive issue for the DNSO is do we fight to keep the
DNSO as the official main source of policy advice to the Board on
domain name issues or do we recognise this is not even the case at
present and become one of a number of competing voices on domain name

Setting up an ALSO raises the potential for the Board to be able to
choose between competing advice and marginalise the DNSO even more.
However in the absence of individual registrants being represented
within the DNSO then I have to say that this is something to be
supported.  If the DNSO wishes to remain the official source of domain
name policy advice it needs to be fully representative.  

Note I support at large members remaining regardless and retaining the
right to elect half the Board.  It is whether they should formally
constitute a SO which provides direct policy advice to the Board which
is the issue for me.

A ccSO raises in my opinion (Note conflict of interest I am on the
Council of a ccTLD manager) far fewer issues because a fairly neat
dividing line can be drawn between gTLD issues and ccTLD issues.  Very
few ccTLD issues can be dealt with globally - they tend to be local
solutions to local problems and I believe a ccSO is an effective way
of dealing with these.  However some in the GA are strongly against a
ccSO and unlike the debate on at large board numbers in the GA (which
clearly favours nine from my observations), I can not detect at this
stage whether there is either clear support for or against a ccSO.  I
believe proposed bylaws for an ccSO will need to be viewed before
asking the GA to vote.

My thinking is that ICANN needs to fundamentally decide whether it is
abandoning its model of distinct semi-autonomous SOs with no policy
overlap, or to recognise the reality that power is with the Board (and
staff) and then focus on how that Board should be comprised.

There is a recurring issue about massive influence by ICANN staff.  My
observation is that this is not unique to ICANN but happens in almost
all voluntary organisation.  What is different about ICANN though is
that the staff basically service the Board only and not the SOs.  If
one really wants the SOs more involved in policy areas one would need
to either develop your own staffing base (beyond admin) or have ICANN
treat SOs as fundamental parts of ICANN and have them directly service
SOs.  For example Louis Touton just made some useful observations on
the .org issue.  Wouldn't it have been more useful though if a staff
member had been assigned to work with the Taskforce so these issues
could have been considered as part of their work?  

I know of few other organisations where staff only service the Board
and not what is in effect sub-committees of the Board (SOs are at the
end of the day glorified sub-committees of ICANN).

Having got my introductory comments out of the way (yes Virginia there
is an end to this), I'll comment on Philip's evaluation.  There has
been little substantive comment to it on the GA list but I will post a
copy of this post there which may generate some.


I agree that an IDNH constituency is preferable if policy making stays
within the DNSO.  In fact there is a need for both - an IDNH to have
an effective voice within the DNSO on domain name issues and an ALSO
which would concern itself with wider ICANN issues.  It is not
necessarily a case of one or the other but preferably both - with well
defined areas of responsibility.

If the DNSO becomes a providers SO and the ALSO a users SO I agree
with Philip that there will be a very high overlap of issues.  Note
however that while the DNSO remains without an IDNH constituency this
will be seen by many as a positive thing.  If the DNSO becomes truly
representative of all stake holders then it would be less desirable.

Philip correctly identifies that there is no mechanism for cross-SO
communication.  If an ALSO is set up as recommended then I suggest
there will be some sort of need for an ICANN General Assembly which
SOs can get together and talk.  This need not be identical to the DNSO
GA but it is possible that an ICANN GA could replace the DNSO GA if
domain name issues are no longer the theoretical exclusive preserve of
the DNSO.


Philip identifies the possibility of the Board receiving contradictory
advice from different SOs.  I agree this is likely but as the DNSO
already has such small influence (as a formal body, some individuals
are very influential) on domain name issues the actual difference may
be little.  As I said previously the DNSO needs to sharpen its act up
if it wants to retain the preserve of domain name policy.

It is also a good thing that the ALSO will allow individuals input
into any issues arising out of the ASO and PSO.  While there are not
many of these issues, there is also no effective way for individual's
to have a direct say on these issues except through public forums.

I disagree with Phillip's evaluation that the real test of
individuals' interest in an at large SO will be when members are asked
to pay because it all depends on what the fee is.  I can not see an
ALSO functioning for less than US$200,000 as it has to cover the costs
of staffing, website, online voting, organisation to have any chance
of being able to compete with other SOs in having input into Board

Now that is asking 2,000 members to pay US$100 each.  Well of course
heaps are not going to.  If the ALSO is asked to self-fund itself it
will never amount to more than a larger version of the GA and will be
doomed to ineffectiveness.

What is needed IMO is a partnership between ICANN and an ALSO.  First
of all one has to recognise that individuals will always be at a
massive disadvantage compared to businesses when it comes to funding
an organisation.  100 businesses paying $1,000 each (for example) is
far easier to organise than 10,000 individuals paying $10 each.

A fee should be set for the ALSO in the US$5 - US$10 range.  This is
enough that people will be showing some commitment to ICANN.  I can
not see any alternative except ICANN itself fund surplus costs of the
ALSO as an expense of democracy basically.  And yes this is registries
and registrars and businesses subsidising individuals but I would
suggest that said individuals directly or indirectly fund the
registries, registrars and businesses :-)

The last part of Philip's evaluation I also disagree with - strongly.
The ALSO proposal does not in any way increase the at large
representation from five to six.  It decreases it from nine to six.  

Once again apologies for the length of the post.  Hopefully some
issues for consideration and debate.

ICQ 29964527

<<< Chronological Index >>>    <<< Thread Index >>>