[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
Re: How do we want DNSO to be?
- Date: Thu, 26 Nov 1998 11:29:13 -0800
- From: Kent Crispin <email@example.com>
- Subject: Re: How do we want DNSO to be?
On Thu, Nov 26, 1998 at 04:20:29PM +0100, Ivan Pope wrote:
> >> How do the attendees break themselves out by constituency?
> >> Registry: 18
> >> Registrars: 3
> >> ISP and infrastructures: 7
> >> Businesses: 1
> >> TM: 2
> >> At large: 3
> >> (from the Monterrey meeting notes)
> >> If businesses equate with the Domain Name Owners, we must all agree that
> >> they were pretty badly underrepresented in the meeting.
> I have read the DNSO Monterrey meeting notes.
> It seems to me that we are right back in the madness of trying to construct
> a representative democracy in which all interests are finely balanced, in
> which no party can 'capture' the organisation, in which we write all the
> rules and regulations and byelaws before the thing exists.
> I would state clearly that this way lies madness.
I agree that it is nuts to try to write all the rules and
regulations etc in advance. But I don't agree that the concern
about capture is groundless -- capture is a serious concern, and if
you look at the behavior of the ccTLD registries, you see the danger
> What is happening here is that the issues that are important to those who
> are discussing them are what is driving the discussion. There is no thought
> as to what will actually happen once the DNSO is formed, or even to what
> its role and function will be.
Actually, there has been a great deal of thought about those things.
> There is just an obsession that no-one be
> 'left out', or that we all are allowed to stake our little claims.
There is, in addition, the obsession you describe.
> The preposterousness of splitting by constituency is obvious. So, 18
> attendees at Monterrey were 'Registry's. Well, how do we define a Registry?
Actually, defining a registry is about the *simplest* problem we have.
> It is fairly clear that this is a self-selecting group.
> Is Network
> Solutions a Registry or a Registrar?
The categories are not mutually exclusive. You will note from the
notes that it is expressly considered that many entities may wish to
participate in more than one constituency. One important restriction
is that if you participate in any other constituency you can't
participate in the At large constituency.
> What about Nominet? What about AOL?
> Oh, are AOL an ISP? Or maybe they are a business? No, they are an
> Infrastructure? Do Thompson and Thompson provide Internet Services? Are
> they then in the ISP group? Or are they in the TM?
> I could easily, without blinking, put NetNames in all categories.
Probably so. Feel free to do so (except of course the At Large).
We discussed this *exact* issue at some length.
> So what is the value of being in one category or another? Do these
> categories have sole rights to discuss only 'their' subjects? Do we have to
> pass any recommendations to all other categories? Can there be a veto?
The primary purpose of constituencies, from my perspective, is to
prevent capture. Capture is a very real problem, if you only have a
single level general membership. I believe that the constituency
model essentially eliminates that as a concern, *provided* that there
are enough constituencies, that they are sufficiently different, and
that none of them have excess representation (like the registries
> Come on, we are building a nonsense nightmare. And worse than that, it is a
> nonsense nightmare designed only to satisfy the immediate needs of those
> currently engaged.
> Firstly, we don't have to replicate the ICANN structure.
Actually, you have it backward. ICANN doesn't have to replicate the
> We do have to
> build a strong advocate for Domain Name issues. We have to have a fair and
> open organisation that allows all voices to be heard and which can make
> efficient decisions. We should pull together, not pull apart at this stage.
That's what we all have in mind, I believe, except maybe the part
about "efficient decsions" :-)
> The DNSO should be a broad membership organisation, but there should be
> limits on that membership.
It *is* a broad membership organization.
> For example, we should not try to build a
> multi-million membership organisation. We need to impose some restraints on
> who can directly join, or it will be madness and we will fail in our brief.
> And remember, failure leads right back to the USG controlling it.
> I propose that we set out some idea of who we would like to join. In my
> view this should be any organisation (or individual) who is prepared to pay
> a joining fee and sign up.
Great minds think alike. That is *precisely* the basic requirement
for membership that has already been established.
> I propose that we forget at this stage notions of categories or how many
> representatives they can elect or what those representatives can and can't
This suggestion sounds good, but it's not in tune with the political
reality that exists, in that there are some already-existing
groupings, and they already know that they want to be considered a
> The general membership should elect a board of directors.
You are repeating all the logic we went through, you know...
The DNSO "board of directors" is the "Names Council", in our
interpretation. You have to realize that there are constraints that
come from the ICANN bylaws, and those requirements drive a lot of
what we do here. One of them says:
"The Domain Name Supporting Organization shall create a Names
Council to make recommendations regarding TLDs, including
operation, assignment and management of the domain name system and
other related subjects"
It doesn't say to whom the recommendations are to be made --
presumably the ICANN board.
> The board of
> directors should expand notions of accountability, mission and
> consultation. There are millions of organisations out there who do this, so
> its not rocket science.
> Until DNSO is up and running we won't have the faintest idea of how the
> membership want it to run or what they want it to do.
That is no where near correct. We have a great deal of information
about how the membership will want to run things, because every
single participant in the DNSO process *will* be a member.
> Any attempt now to
> settle this is an absolute definition of 'capture' in my book. Capture by
> I propose that we find some general purpose bylaws that retain flexibility
> to evolve the organisation.
Bear in mind that we don't have bylaws, because the DNSO is not a
separate organization from ICANN.
> That we invite members to sign up and pay their fees on this basis. I
> suggest a fee of $1000 and an aim to recruit 1000 members.
It is my strong impression that such a scheme Won't Fly with ICANN.
We need to allow individual members somewhere, and the minute we do
that we have to worry about capture.
> That we ask ICANN to recognise that this organisation is aiming to become
> the DNSO.
My impression is that ICANN won't recognize an organization that
doesn't allow for individuals somewhere. Period. But it's moot,
because the sentiment for individual memberships is far to strong to
> The membership will soon work out who stands for what, members with
> different ambitions and aims and interests will find that they group
> together, vote for candidates, join working and sub groups etc etc. Life
> will start.
> There, its as simple as that.
Been there, done that, won't work. Next idea?
Kent Crispin, PAB Chair "No reason to get excited",
firstname.lastname@example.org the thief he kindly spoke...
PGP fingerprint: B1 8B 72 ED 55 21 5E 44 61 F4 58 0F 72 10 65 55