Re: The case of dotless domains

On 07/14/2013 10:05 PM, S Moonesamy wrote:
> At 06:53 14-07-2013, Yoav Nir wrote:
> > http://tools.ietf.org/html/draft-moonesamy-dotless-domains-00
> That memo discusses about the case of the dotless domains in terms of
> the technical standards.  Comments are welcome.
Interesting, but pardon my being blunt, it does not seem to cover any 
new ground. You did say one useful thing:
IETF specifications usually provide guidance to ensure interoperability. 
 Whether dotless domains are harmful or not is a policy matter.
> At 13:11 13-07-2013, Ofer Inbar wrote:
> > What this brings to mind is that we used to have implicit DNS domain
> > search in the early days of DNS.  When edu.com accidentally hijacked
> > a huge chunk of the Internet, most of the net very quickly got rid of
> > implicit search, and we got the explicit DNS search feature that many
> > people are discussing now.
> Yes.
Can you (or Ofer) define how you're using the terms "explicit" and 
"implicit" in terms of DNS search, and what their relevance is to the 
topic of dotless domains? And no, I'm not being snarky, I think part of 
the problem here is a fundamental misunderstanding of how the vast 
majority of hosts are configured currently.
> > If some new TLD gets used in a dotless fashion in a way that truly
> > does cause major trouble, I expect we'll see sites all over the net
> > quickly deploying DNS resolvers that discard A, AAAA, or MX records
> > at the top level, to protect their users.
> There is already deployed code to do that.
True, for better or worse. :)  BIND's implementation also starts off 
with a reasonable default list of delegation-only domains, and allows 
you to customize the list.
... which is an excellent example of market forces influencing software 
development.
> At 20:14 14-07-2013, Doug Barton wrote:
> > It is unarguably true that as things currently stand there will be
> > "problems" with dotless domains. How widespread, and how serious those
> > problems become is yet to be seen.
> I haven't seen anyone in the IETF arguing that sufficient market demand
> overrides comments about problems that will appear.
I haven't seen anyone in the IETF arguing that the moon is made of green 
cheese either. What does your comment have to do with anything? Sorry 
(again) to be blunt, but the IETF pounding the "this thing is bad!" drum 
is not a particularly effective tool. [1]
We have to be realistic about how and where our influence is relevant, 
and more importantly how and where it is not. Detailing the technical 
problems with the proposal is relevant IETF work, but that job has 
already been done more thoroughly by my esteemed former colleagues on 
the SSAC.
> > So either this is a good idea that will gain traction, and therefore
> > appropriate software support; or it is a bad idea that will go away on
> > its own. Either way, making a fuss
> It is up to those who decide to decide whether dotless domains is a good
> idea that will gain traction or a bad idea that will go away on it own.
You vastly overestimate your own importance in the grand scheme of 
things. Don't worry, it's a common human problem. :)  Rather than 
repeating my point for the third time, see what I wrote just above.
> At 20:25 14-07-2013, Dave Crocker wrote:
> > In contrast, assertions about "market demand" ensuring that "software
> > folks... will make them work" rests on a fuzzy concept of market
> > forces -- for example, the market of users isn't likely to be issuing
> > a formal or informal 'demand' about any of this, and a model of
> > altering installed-base behavior that has, I believe, has no
> > historical precedent.
> Yes.
I already explained why Dave's perspective is wrong in my previous 
message, which you apparently chose to ignore. Your example of 
delegation-only domains for DNS software further proves my point.
Doug

[1] cf. NAT