Re: comments on Jim's and Jeff's addenda to the gTLD MOU

[Jim Dixon <jdd@matthew.uk1.vbc.net>]

On Thu, 18 Sep 1997, andi payn wrote:

> I think that someone could easily get a test system set up and running well
> within a month, and let CORE take it over and begin running it the way they
> want to once they're incorporated and ready to go. Once they're happy with
> it, which should be another month at most, they can start accepting
> registrarions.

Agreed.  This is how things have been done traditionally on the Internet,
and this is why the Internet has grown so quickly.
 
> I agree that CORE should be professionally managed, but I don't think that
> this belongs in the gTLD-MoU. If it belongs anywhere, it's in either the
> CORE-MoU or the CORE articles of incorporation, both of which were written
> by iPOC and can be modified by CORE only with POC's approval, I believe.

What we are trying to do here is to collect a list of things that people
agree upon.  Which MOU they belong in is not very important.  

Try to keep in mind that CORE does not at this moment exist.  So
everything, just everything, is up in the air.

> >>>   6       if at all possible, existing shared registry software should
> >>>           be used in CORE's operations
> >>
> >> "Existing shared registry"?? Now you sound funny. Are you against
> progress? Has
> >> ... 
> The main point of starting with existing software is to get things rolling
> as soon as possible. 

Precisely.  If there was no pressure of any sort, it might be better to
design an ideal system.  And if there was no experience to draw on, we
would have to start from scratch.  But there are many national registries
that have been running for quite some time.  It is foolish to ignore 
their experience.  At least one of these is a shared registry, just like
CORE, that has been running in one form or another for several years. 
That's Nominet in the UK.

>                     Even if nothing out there is a perfect fit, it may be
> easier to modify existing software than to start from scratch. Hopefully,
> the software used by Nominet or another existing registry will be either
> usable or close enough to use as a starting point. The second reason to
> start with existing software is, of course, that it's presumably been proven
> through use--that is, you already know how reliable it is.

Also, people always use systems in odd ways that the system designer
doesn't anticipate.  Real experience is invaluable.

> >>   8       CORE should be strongly encouraged to use public domain
> software
> >>
> >> Public domain db that large, powerful and reliable? Well, if they
> exist.... But
> >> again fitting this is a
> >> second order choice, after service quality.

All over the world there are usenet news machines taking hundreds of
thousands of hits a day.  These are large, powerful, reliable machines
using public domain software.  I don't know any news machine using
commercial software that can outperform machines using public domain
software.

We have learned to be very careful explaining this fact of life to
our customers.  Their expensive commercial software is run into the
ground by our news machines running free, public domain software.

> >  I agree again here.  But it is possible to use public domain software
> >in many instances.  This is no exception.
> 
> Much of the software that drives the Internet is public domain, or at least
> GNU- or similarly-licensed free software. 

Yep.  We have the choice of at least two different futures.  In one, the
world's registries use expensive and incompatible proprietary software
solutions.  In the other, they use open standards and public domain 
software and systems that can communicate using standard protocols.

>  >>   10      a gTLD name server should put put on line as quickly as
> >>           possible, preferably by 15 October
> >> ...
> 
> I was going to say that if this had been done differently from the
> beginning, there would be plenty of people volunteering time and equipment
> to make this happen. Fortunately, there still are some.

Indeed.

> There's no reason you can't have a server on line with test data before you
> finish the selection process. As I pointed out before, this would be a good
> thing. CORE, once they got organized, could spend a month or so testing and
> tinkering with the system, and barring any serious problems or objections,
> they could begin registering names well before the end of the year.

Exactly.
 
> Of course there is no one set of standards used by every jurisdiction in the
> world, but there are some basic principles that are used in most
> jurisdictions (enough for WIPO to consider them international), and they are
> not the principles in use in the MoU (or, for that matter, in Jim's
> counterproposal).

Can you elaborate?
 
> If the use of a trademarked name by someone other than the trademark holder
> causes confusion, or is meant to mislead, or dilutes a "strong" trademark,
> it's a violation. There are exceptions for prior use by the non-holder, fair
> use and/or satirical intent by the non-holder, and misuse of the trademark
> by the trademark holder. If anyone finds these standards objectionable in
> some way, please explain to me how.

Just for the record, could you define "strong trademark" and justify 
its special treatment?

> OK, Amadeu, did you read my proposals (the third set)? I think that what
> we've come up so far with would be enough to alleviate the fears of hundreds
> of people who were previously hesitant or unwilling to sign, and remember
> that at this stage we're only a few hours and three proposals into the first
> phase...

Yes, indeed :-)
 
--
Jim Dixon                                                 Managing Director
VBCnet GB Ltd                http://www.vbc.net        tel +44 117 929 1316
-----------------------------------hobbies---------------------------------
Member of Council                                                 President
Internet Services Providers Association                       EuroISPA EEIG
http://www.ispa.org.uk                              http://www.euroispa.org
tel +44 171 976 0679                                     tel +32 2 503 2265