Re: Representation on PAB

John Charles Broomfield (jbroom@outremer.com)
Mon, 02 Feb 1998 15:03:03 -0400


> > Voting and "rough consensus", you mean?
>
> No. I meant that there is no logical connection between "One person, one
> vote" and "the rough consensus mandate". If one person represents many
> parties, that must be taken into consideration when deciding whether
> rough consensus has been reached.

Before anything, I think I was the dissident when the vote was taken on
if multiple votes should be allowed from the same person representing
different companies. I'm allowed to change my mind though. ;)

It is organisations that sign the gTLD-MoU, however it is (as Kent has
pointed
out) individuals that discuss on PAB. PAB is *not* generally a voting
body,
but (as has also been pointed out), one where rough consensus is to be
achieved.
There is a great wealth of benefit to be gained by having many REAL
voices
on PAB (as opposed to representations etc...).
As you know, there are many countries in the world where setting up an
organisation or society is a trivial matter, and in these places,
multiple
versions of what is in fact the same companies happen. There are no
strict
controls or limits as to what companies/organisations may sign the
gTLD-MoU,
so if it were accepted that one single individual votes on behalf of all
those companies, you could have a single person hijacking the whole
procedure.
I could with hardly any problem whatsoever go out tomorrow and register
2000
different companies, sign on behalf of them all (despite them being de
facto
non-entities, but de jure they exist), and (as you propose), demand 2000
votes.
Not really a good way to go I would say.

> ISPA UK has a single representative to EuroISPA. That single
> representative has one vote on the EuroISPA council.

And if ISPA UK decided to participate on PAB, it would be ISPA UK
having that one vote on PAB. Of course, it would be ISPA UK signing
the gTLD-MoU, so it would only be ONE more signatory to the MoU.

> EuroISPA discussed alternative weighted voting schemes and may
> yet adjust the articles to permit them. In the end we decided that
> each member should have one vote because the disparity between
> the number of members represented was simply not that great.

So it seems that (currently) not even you accept weighting schemes.
They are generally complicated to implement, and in the case of PAB,
could easily be cheated on.

> ISPA UK is a member of EuroISPA. It is not a member of the European
> Union. The proposal is that ISPA UK's members should sign the
> gTLD MOU and join the PAB. ISPA UK has no vote in European Union
> deliberations. ISPA UK's members would certainly expect to have
> votes on the PAB.

If those ISP's want to have votes on PAB, then it's easy, they sign the
gTLD-MoU, they say they DO want to participate in the PAB deliberations,
and they put someone in to participate. This generally means listening
in, but would help if sometime they contribute. Votes are not really
taken that often, so they wouldn't really be troubled by the "hassle".
Of course, if you personally are trying to get that representation you
could feel a little frustrated at them having freedom to vote for
themselves... (please forward my email to your ISP members).

> > Then there is the matter of a form of cheating that all of a sudden
> > becomes possible -- a British ISP, for example, could be
> > represented two times -- it could sign on its own, and it could be
> > represented by UK ISPA. Of course, it could be represented many
> > times over, depending on how many organizations it belonged to.
>
> If this is cheating, then ISOC is cheating. The last time that I
> looked, I counted a dozen "ISOC" signatures on the gTLD MOU. There
> was ISOC, then ISOC here, ISOC there, ...

You're mixing things. You're mixing signatures on the gTLD-MoU with
voting capacity on PAB. Remember, signatures are by organisations,
voices
on PAB are individuals. I've personally got mixed views about the
different
chapters of ISOC signing the MoU, however they do not have one person on
PAB who collectively counts as 10 votes. To be honest, I don't see too
much of ISOC in PAB (except for Sascha maybe) participating, so you
can't
even argue that ISOC has hijacked PAB.

> In fact, we aren't talking about cheating in any form. We are talking
> about N signatories to the gTLD MOU choosing to appoint a single
> representative who will vote on behalf of each. Your approach casually
> deprives signatories of their votes.

No, signing the gTLD-MoU doesn't give you automatically a vote on PAB,
it
opens the doors for you to participate or not in PAB if you want, and
participation means precisely that. Not that you name someone else on
your
behalf. I can go around my neighbourhood tomorrow and get all my local
stores to sign up to the gTLD-MoU and have them name me their rep. That
would certainly raise some eyebrows.

> > > But there is a simple quid pro quo: if ISPA UK's 83+ members appoint
> > > someone to represent their interests, we want the fact that that person
> > > is speaking on behalf of 83+ organisations and their million+
> > > customers to be recognised.
> >
> > Following your reasoning, if you signed for EuroISPA, then you believe
> > that you should be given 500+ votes, because, according to your
> > letter, EuroISPA represents between 500 and 1000 ISPs. Or
> > alternatively, the various members of EuroISPA could sign, and between
> > them command 500 to 1000 votes.
>
> You may find it amusing to argue like this, but what I actually proposed
> was that our members sign, each appointing a representative. In cases
> where one person represented many signatories, that person should have
> a proportionate number of votes.

You can have that person actively participate and say "I'm talking on
behalf
of these X participants", but when it comes to voting, it will have to
be
those participants voting, and not the "rep". For them to participate it
would
mean that they are on PAB. Of course, even this method leaves doors open
to
hijacking, and you could relatively easily find a way. I hope you will
try to follow more the spirit than the wording.

> What I am really proposing is that the Internet community take over the
> PAB. I think that industry in particular should join en masse. Yes,
> EuroISPA's members should join. The CIX should urge its members to join
> on the same basis. So should every other trade association in the Internet.
>
> It is in our interest that the circus end.

Exactly, and it is in our interest to have broad participation, not just
figurative participation. If you want to weigh votes, how do you do it?
Based on company assets? On users servered? On bandwidth traffic? On how
many IP addresse the organisation has? On number of organisations? All
of
these can be faked. Today, so can email addresses, but as we're not
generally voting, and discussions reach rough consensus, then the fact
that
"bogus" addresses could exist is alleviated somewhat.

> > What would work to achieve good, though, would be for EuroISPA to sign
> > the MoU, and to get as many other groups as possible to sign and
> > participate, as well. It is simply true that the MoU and EuroISPA
> > have many common goals, and share a common belief that the Internet is
> > international, and not the property of the USG. It is also true that
> > many of the ideas you have presented have already been bruited in PAB;
> > with your participation they could very well pass.
>
> I doubt very much that the EuroISPA Council will have been persuaded by
> your arguments. We have not signed the gTLD MOU up until now because
> the consensus has been that those involved would gleefully include our
> members in their count of supporters while minimizing our actual influence
> through procedural devices - just as you have outlined above.

Well, maybe EuroISPA won't sign, but if its individual members haven't
signed
so far, then that's up to them. EuroISPA signs, then the person who goes
on
PAB gets a vote. A single ISP signs, same thing. If the person is one
and the
same, then he should know who he's voting for, but he gets one vote, not
two.

Yours, John Broomfield.