From: Anand Kumria [akumria@acm.org]
Sent: Monday, 30 September 2002 5:35 AM
To: Jo Lim
Subject: Re: new 2LD proposal


Hi Jo,

From the minutes and associated web pages I note that only Mark Tearle has a proposal 
for .conf.au listed. While my own is substanially similiar I'd appreciate it if you 
could bring it to the attention of the New Names committee.

> In the course of discussions to date, it has become clear to the Panel 
> that the creation of new 2LDs raises a number of important policy 
> issues. The Panel has decided to release two discussion papers to 
> invite public feedback on these issues:
> - New 2LDs Discussion Paper at 
> http://www.auda.org.au/policy/panel-newname-2002/new2LDs-discuss.html

The discussion paper says "Recent changes to domain name allocation policy in the 
existing 2LDs have made it easier for registrants to get the domain name they want. 
Therefore, the Panel believes that the onus is squarely on the proponents of change 
to demonstrate that the change will bring commensurate benefits". I think at this point 
it is still to early to evaluate the impact (benficial or otherwise) of having multiple 
registrants.

For example will .net.au continue to be de-valued? It was initially for network-related 
entities but now has the same charter as .com.au

Yet some of the recent registrants in both .com.au and .net.au, for
example: livingthing.net.au and dreamandachieve.com.au clearly don't fit the charter 
(from http://www.auda.org.au/register/: "For commercial entities, such as companies 
(with ACN as registered through ASIC), and businesses (registered with state 
governments").

Worse entities now clearly feel that nothing within .au suits them (e.g. 
cityrail.info).

Something the committee may wish to consider is why livingthing.net.au was not 
registered under livingthing.nsw.gov.au. 

Why dreamandachieve.com.au was not registered under dreamandachieve.dolly.com.au or 
dreamandachieve.ninemsn.com.au (in fact it all goes to dolly.ninemsn.com.au, why 
wasn't that promoted?).

Cityrail.info merely maps to cityrail.nsw.gov.au, why did cityrail feel the need 
to do this?

All of these registrantions de-value .com.au and .net.au (and .au)

They make it less useful for the community at large because you can no longer make 
the assumption that an entity within either of those domains is actually an organisation. 
That was a valid assumption previously.

It seems clear that the "new regime" allows organisations with a dislike for four-level 
domains to go a level up.

Onto the various points raised within the discussion paper:

1. Preservation of the overall integrity and usability of the .au domain space

One of the most compelling aspects of registration within .au has been the fairly 
focused nature of registrants. Entities in one area would generally have to demonstrate 
a clear need to be in another.

The lack of clear registration space ( .comp.au; .campaign.au; et al) has meant people 
register where ever they can. 

I suspect taht once a domain has been approved, normal promotion would make any selection 
memorable. Aafter all, a substanial amount of money will be spent educating people about 
it.

As I've pointed out above organisations appear to willing to promote third-level domains 
with some vigour but fourth-level domains haven't yet caught on.

2. DNS hierarchy issues 

RE: "defensive registrations". Most companies that I am aware of already do this as a 
mater of policy within .net.au and .com.au (e.g www.optus.net.au and www.optus.com.au) 
- I believe the committee is mistaken if they believe that most companies do not do this.

Having fairly clear registration criteria (e.g. principally educational, principally 
medicial) reduces the instance of this and contributes to making the DNS more useful.

3. Purpose of the DNS

While the initial intent of the DNS was to map names into numbers (and vice versa). Most 
current users would actually perceive it as a service directory. The fairly common usage 
of www.something.com.au and ftp.something.com.au are examples.

For those within large entites, descriptive / directory locations are actually mapped into 
DNS i.e. The Printer in Sales is normally mapped as sales_printer._lpd._tcp.something.com.au

There is some feeling within the IETF (the ZeroConf WG) that due to the wide-spread 
deployment of DNS resolvers, layering a directory service on top will be of benefit to most 
users. 

Both Microsoft and Apple have incorporated this into their respective offerings (Rendezvous 
and Active Directory respectively) so it can only become more common over time.

Assisting users by transitioning the national domain space to facilite ease of discovery, 
by stratification of the domain space, would be benefical.

4. Precedent-setting 

The committee appears to be worried which of the two models, industry (an economic 
grouping) or profession (a personal grouping) it should select.

The simplest way is to gauge demand. I would suggest that the only effective way to do 
this would be to require registrants within an existing .au space give it up and migrate 
to the new one offer.

Evaluating the take up after a period of time (say 1 year) would give guidance on which 
model the public and registrants prefer. That would require some kind of comittment to 
an experiment from auDA - they may end up with a "unique" area like csiro.au again. 

5. Commercial advantage considerations 

The committee has a very valid point; with reference to my specific closed 2LD proposals 
(.med.au and .emb.au). I can see some benefit to having .med.au open - that way no 
organisation within that space would be behold to any registrar.

For other areas perhaps a better option would be to allow any accredited registrar to 
process the domain after it had been 'approved'. This also take away the need for any 
proposed management body to demonstrate their funding credentials - you can relying on 
the existing competitive regime. 

The approval body may levy a fee for each domain processed if funding is required for that 
function. However having .emb.au open may be more confusing - should anyone be allowed to 
register us.emb.au or greece.emb.au?

6. User demand

I believe attempting to evaluate non-registrant demand is ultimately futile. Registrants 
are the ones who will determine whether they will invest money into promoting their identity 
online.

Despite a lack of visibility and a challenging registration process conf.au has been used 
successfully by number of organisations (Mark Tearle's submissions idenifies quite a number). 
There is clearly demand for it. [1]

I believe that the best way to gauge demand from would-be registrants would be to require 
them to give up any existing registration. For some domains this may not be appropriate 
(specifically conf.au) but in general I believe it is.

Would minters.com.au be willing to give it up and become minters.law.au? Or would they 
value the conformity that minters.com / minters.com.au provides them?

Would individual lawyers listed in at lawsociety.com.au be more likely to take up a direct 
law.au regsitration? Similiarly for other proposed domains which would stratify .au

I think the main thing to keep in mind is whether any proposed 2LD would be a more 
appropriate option for a would-be registrant than existing domains.

Good luck.

Thanks,
Anand

[1]: I've previously been, and continue to be, involved with linux.conf.au