Re: [wg-c] Votebot sample ballot.

["Jim Glanz" <bidquik@bigsky.net>]

This seems to be a reasonable approach toward a solution.  I say go with it.

Jim Glanz

-----Original Message-----
From: Roeland M.J. Meyer <rmeyer@mhsc.com>
To: Robert F. Connelly <rconnell@psi-japan.com>; wg-c@dnso.org
<wg-c@dnso.org>
Cc: Dan Busarow <dan@dpcsys.com>; Jonathan Weinberg <weinberg@mail.msen.com>
Date: Thursday, September 09, 1999 8:23 AM
Subject: RE: [wg-c] Votebot sample ballot.


>> Robert F. Connelly: Thursday, September 09, 1999 1:07 AM
>>
>> At 23:45 08-09-1999 -0700, Roeland M.J. Meyer wrote:
>
>> >This is
>> >because we can ONLY answer the "how many" issue, with
>> >consensus,  when
>> >we all understand [and have agreed to] the "How to add TLDs" process
>> >question. Without that answer the debate recurses back to the "if any
>> >question".
>> >
>> >I have suggested a process that I think will work. More
>> >importantly, it
>> >is a sequence of steps that will get us there.
>> >
>> >-------------------------------------------------------------
>> >Strategy:
>> >Handle one step at a time, in order:
>> >Evidence of each step's completion is;
>> >         1) a consensus position, based on a vote.
>> >         2) Recommendation for modification of the next step
>> (if any).
>> >
>> >Steps:
>> >         1) Do we add any new Registry/gTLDs?
>>
>> Dear Roeland:
>>
>> I see what you mean.  First vote up or down, zero gTLDs or
>> more that zero
>> gTLDs.
>>
>> I'm not sure we need to have this many additional ballots.
>> It takes one
>> week to run a VoteBot.
>
>We shouldn't feel such time-pressure becasue this is a complex and
>highly-charged issue. Further, the answers to  2-5 are not trivial.
>
>> The advantage of Dan's system (which Joop's may share) is
>> that we can see  just who votes for what.
>
>I see your point for an open vote and I agree with it. Joop's system can
>handle that without the turn-around time of Dan's VoteBot. However, I
>don't see that making much of a difference.
>
>> I like it -- except for the long lead time.  But, let's
>> dispose of this  first issue quickly, zero or more than zero.
>
>Yes, I agree. However, I expect 3-5 to be rather complex. Let me list
>what I expect the evidence of completion (EOC), for those steps, to be.
>(SLA= Service Level Agreement)
>
>> >         2) Where do we register the TLD registry (root registry)?
> EOC:
> 1) Agreement that ICANN/DNSO will run a root registry.
> 2) SLA for the ICANN/IANA/DNSO root-registry.
> 3) Operational requirements.
> 4) Basic process for adding new TLDs.
>
>> >         3) Minimum services expected from a TLD Registry.
> EOC:
> 1) Minimum SLA for a TLD registry.
>
>> >         4) Minimum services expected from a gTLD.
> EOC:
> 1) Minimum SLA for a gTLD registry.
>
>> >         5) Minimum  business requirements for a Registry.
> EOC:
> 1) Minimum stability requirements for the registry.
> 2) Fall-back/reassignment plans for event of registry business
>failure.
>
>> >         6) How many registries?
>> >         7) How many gTLDs per registry?
>
>> >Also, there are NO  SHORTCUTS. That
>> >is the very reason we are in this jam to begin with. Let us all
>> >excersize some self-discipline and stick to the process, whatever
>> >process we agree to.
>
>As you can see, I expect most of the output, of 2-5, to be documents.
>These are non-trivial documents. Once these documents are instantiated,
>and ONLY after these documents are instantiated, do I expect reasonable
>debate on items 6 and 7, without fear of deadlock.
>
>Any attempt at short-cutting this process would be a waste of time,
>IMHO.
>
>--------------------
>Roeland M.J. Meyer, CEO
>Morgan Hill Software Company, Inc.
>http://www.mhsc.com/
>mailto://rmeyer@mhsc.com
>--------------------
>