Re: [ALSC-Forum] Authentication Technology -- Please try it out

[Mr. Pindar Wong]

Hi Adam,

Thanks for this, pls. see my quick comments below as I've to pack for
Australia (among other things to do this evening ... including 'eat' ).

----- Original Message -----
From: "Adam PEAKE" <pja@glocom.ac.jp>

> Are you saying that registrars are willing to handle the fees people
> will pay for membership: on top of the charge for the domain name, they
> will separate a stream of money from the individual to ICANN (ALM).
> They will aggregate, do the currency conversion and send to whatever
> bank account ICANN or the ALM organization specifies?
>
> If so, how many ICANN accredited registrars have agreed to this?  How
> many ccTLDs have agreed to this?

I believe the operative word is *could*.

Regardless if one agrees/disagrees with the Individual  Domain Name holding
criteria, which IMHO depends on you view of what ICANN does, it seemed to us
that there already exists a globally distributed infrastructure which might
be, pardon the word ... exploited (and was relevant to what ICANN does and
to which ICANN has potential access to -- i.e. the gtld and cctld registries
and registrars).

Therefore why not use this infrastructure as the initial one for operations.

It's global and it's there.

>From a financial perspective, many of the At Large activities would
constitute an incremental cost to them -- and in our view,  keeping
membership dues as low as practical was appealing. Furthermore, one can
lower the costs by various forms of aggregation (to keep the per unit costs
of financial transfers as low as possible, lower the need to understanding
how the international funds transfer infrastructure works to make a payment
(e.g. TT's), not necessitate the use of a credit card, enable other kinds of
payment (in the worst case... cash).

Other useful aspects other than financial clearing are:-

Outreach -- Use of their publicity infrastructure (to make their existing
and prospective domain name holders aware of the At large, its purpose etc.)
Outreach in local language and local customs.

Obviously the gtld/cctld registries and registrars would be concerned that
messages that go out on behalf or the At Large would be carefully
controlled, they won't be overwhelmed with customer support questions about
At Large (this may not be a good problem for them, but its a 'good' problem
to have for the at large (i.e. lots of members?)

We've been working on a paper to formally ask the cctlds and gtlds what
we're precisely asking them to do, we're almost there, but response so far
as been low.

I guess it's really a question if the leading  registries and registrars see
the value in the at large (e.g. ICANN's stability etc.) and, in the end,
take the initiative to step up to plate and volunteer.


> One advantage of the domain name holder idea seems to be that the
> registrar would handle the money and so remove the problems of
> transaction costs and other out of pocket expenses, significant problems
> with the fee system (at least for those without credit cards.)

Yup... see above.

>
> But I think we agree that transferring money (credit card, wire
> transfer, etc.) will generally provide as much information to help
> identify a member as the domain name registration system (at the present
> time.)

agreed ... that's if you require all users to pay money. If you don't then
you loose this ability to authenticate. If you don't require folks to pay
money then minimally you need to deter abuse and automated fraud.

Other mechanisms  that we're thinking about are those that involve
technologies that try to ensure that there's a living human being at the end
of the line (ideally one that doesn't involve hardware. low training costs
etc.) e.g. one currently under consideration is  the  passface technology
(flame away ;)

Essentially we're minimally try to ensure that there's a person at the end
of an email address and that the costs of automated fraud would be high
enough to be a deterrent and furthermore could  be detected before 'damage
is done'.

Also those mechanisms that avoid you having to  make up any magic numbers
(i.e. arbitrary thresholds).

>Paying money is also an indication of demonstrable interest (and
> I note you say creating informed participation is as important as any
> election, and you do not require a domain name to participate, just to
> vote.)

Correct ...  payment is an indication of demonstrable interest (may not be
the only one) -- hence me trying to pass around the virtual hat, asking for
pledges etc. (USD 12,000 and counting)

Personally... i repeat... personally... I believe that creating the informed
participation is actually *more* important than the election (which occurs
only periodically).

> So unless there is near universal agreement among registrars
> (ICANN accredited and ccTLDs) to handle the fees for the individual I
> don't see how domain name holder model adds value?
> It might be one more
> option for registration and expression of interest, but does not seem
> suitable as the default entry requirement to voting rights.

There is also the touchy questions of who one considers to be an ICANN
'stakeholder' and what is the 'public interest ICANN' as I've tried to
indicate.

Again, depending on your view of what ICANN is, you will have different
answers to both questions. If you ask 10 people what ICANN is, you'll
probably get 11 answers ;)

Nevertheless, access to the TLD infrastructure is valuable and one might not
need all of the cctld and ICANN accredited registrars initially to get
access to a representative sample of  *existing* individual domain name
holders ( mostly currently in the gtld space -- the addition of .name is a
new twist). {note:  If a cctld does not encourge/offer domain names to be
held by individuals, that individual is theoretically able to register in
another part of the namespace. The same cctld, while not offering IDNs, may
contribute to At Large outreach (at large registration and outreach aren't
mutually exclusive)

Hence the real question about access to the existing composition of today's
individual domain name holders is getting access to the existing gtld space
that allow them and not all of the TLDS (cctld and gtlds) as one might
think.

Recognising that although access to the TLD infrastructure is a nice , it is
by no means, universal from day one -- led us to consider the domain name+
concept. i.e. IDNHs and other mechanisms that can be found to meet the core
three aspects of (demonstrable interest, authentication and payment).

Here one can imagine elements to a number of scenarios

1) email account holder who subscribes to an ICANN-related mailing list
(e.g. one in local langauge run by a cctld/ISOC chapter)
2) someone who walks in, shows-up to a certain place (e.g. ISOC chapter,
university, local ISP etc.)
3) email account holder who completes an online process to register, then
*pays* to send ICANN/ALSO a sealed *handwritten* envelope (distributes the
financial burden of postage from ICANN to the At Large member, minimal
payment is for paper + postage, gets around the problem of a badly formed
postal address (i.e. return to sender posts), unfortunately requires the
person to be able to know how to write.
4) other ways that one hasn't had the brainpower to think of.

Obviously all of this will evolve over time.

Certainly all the TLDS (both cctlds and gtlds) can assist with At large
outreach. Minimally a link on their homepage and/or providing  some local
language mailing lists for discussing ICANN or At Large issues.


>
> And of course the domain name holder model still has problems with
> equity (developing nations), with definition of individual (most names
> are hold by businesses), with the different registration practices among
> ccTLDs (some have encouraged individual registrations, some have
> restricted/prevented), with timing/registration cycles. Focusing a
> membership around domain names would likely shift the organization even
> further away from the issues of the ASO/PSO. During the White Paper
> process many of us wondered if it was a good idea to lump the stable and
> well function IP registry and protocol actives of IANA with the
> controversial, politicized problems of domain names (I don't mean to
> teach you to suck eggs...)  It bothers me that domain name holding as
> entry point to membership will make things worse in this regard.

Indeed... and their are other issues as well.

It's certainly not a perfect system (far from it), but its a starting
point -- doesn't necessarily have to be the only one (again, depending on
your view of who is a 'stakeholder' and what constitutes the 'public
interest in ICANN')

Given all the confusion, FUD about ICANN, what it is really about ( a today
worry), what it *could* evolve into (a tomorrow worry)  etc., I would argue
that it would be prudent to adopt as wide a criteria as possible that can
revolves around the 3 criteria (demonstrable interest, authentication,
payment ) -- since there's no agreement on (stakeholder/public interest).

Cheers Adam ...

and by the way... although you've certainly contributed alot of sweat equity
(your time) ... what can I put you down for as a pledge (and beer doesn't
count ;)

p.

PS: I really beginning to hate instant noodles.

>
> Kind regards,
>
> Adam
>
>
>
>
> > * that collection of membership dues will enable a stable ALSO that will
> > have resources to encourage informed participation (ideally from a much
> > wider audience)
> > * that such collection of membership dues can create greater confidence
in
> > membership uniqueness
> > * that financial clearing can be done in a currency that is most
convenient
> > * that effective participation mechanisms  within a local internet
community
> > are encouraged
> > * that there is the possibility of having local resources that have some
> > clue as to what the DNS, IP addresses, protocol identifiers are all
about
> > * that membership dues might be able to be contained and reasonable
> > * that familiar distributed database and registration technologies used
in
> > the DNS can also be used to help create the electoral roll  (effectively
a
> > voting registry) (e.g. using whois, EPP etc.)
> > * that stands some chance of actually being implementable
> >
> > This is, of course, assuming that the TLD registries and registrars
actually
> > buy into this whole process and what we are talking about is
implementating
> > a global At Large Membership of a non-trivial size.
> >
> > Does this mean that using Individual Domain Name Holders should be the
> > *only*  entry point into the At Large Membership?
> >
> > This again  really depends on what your view of what ICANN is actually
all
> > about ( I guess the 'root' of the problem ;) -- and perhaps, as others
have
> > argued,  ICANN needs to further clearly limit what activities it engages
in
> > (using 'picket fences' or whatever ...)
> >
> > Nevertheless, as Carl mentioned at the ICANN AGM, we continue to look
for
> > other mechanisms that can constitute 'domain name plus'.
> >
> > i.e. while it may be debateable what constitutes a 'stakeholder' and
what is
> > 'the public interest in ICANN' depending on your view of the ICANN
world,
> > other mechanisms should address issues of
> >
> > 1) Demonstrable Interest
> > 2) Authentication
> > 3) Payment
> >
> > Using Individuals Domain Name Holders is a starting point ... it may not
> > necessarily have to be the only one... given the confusion of what ICANN
is
> > all about.
> >
> > Cheers,
> >
> > p.
> >
> > ----- Original Message -----
> > From: "Eric Dierker" <eric@hi-tek.com>
> > To: "Bruce Young" <Bruce@barelyadequate.info>
> > Cc: "Denise Michel ALSC" <dmichel@atlargestudy.org>;
> > <forum@atlargestudy.org>
> > Sent: Sunday, February 17, 2002 11:48 AM
> > Subject: Re: [ALSC-Forum] Authentication Technology -- Please try it out
> >
> > >
> > > Let us make damn sure that the ALSC understands this question.
> > >
> > > What is your primary reason for suggesting that domain name
registration
> > has
> > > any thing to do with voting in the At-Large?
> > >
> > > And what is your reasoning behind that decision?
> > >
> > > Sincerely,
> > > Eric
> > >
> > > Bruce Young wrote:
> > >
> > > > Denise wrote:
> > > >
> > > > >The financial transaction accompanying a domain name
> > > > >acquisition can contribute to authentication, but it was not the
> > primary
> > > > >reason the ALSC recommended this approach.
> > > >
> > > > Could you illuminate us then as to the primary reason, since the one
you
> > > > mention above was the primary argument made in your final report?
> > > >
> > > > Bruce Young
> > > > Portland, Oregon
> > > > Bruce@barelyadequate.info
> > > > http://www.barelyadequate.info
> > >
> > >