Re: [DNSOP] Adoption and Working Group Last Call for draft-appelbaum-dnsop-onion-tld

On Thu, Jun 4, 2015 at 3:53 PM, John Levine <johnl@taugh.com> wrote:
>>> This is a really good point.   I think there does need to be a .ALT registry in order for .ALT to be able to
>>address anything other than experimental uses.
>>And I think this would actually be a good thing.
>>
>>If we created a registry for alt, how would alt not be just another
>>TLD with exactly the same status as any other domain name registry?
>>You can already register a name in the DNS registries and not turn it
>>on in the DNS.
>
> I think the key difference would be that it would accept any number of
> entries for the same string, and would have a pointer to the place
> where you can download the code that implements it.  If you use
> foo.alt and I use foo.alt, well, that's our problem.  There is also no
> reason to have only one such registry, or why any organization with a
> name starting with "I" would run any of them.

I think that such a list / resource would be a fine idea, but I think that:
A: it would be good to avoid calling it a "registry" (that term has
specific meaning within the DNS world), and
B: it would also be good if someone (or someones) other than the IETF
ran them. This could be a person, like John for exmaple[0], or just
something like a wikipedia page.... Some of my reason for writing the
.alt draft was because I get more than enough ICANN politics at ICANN
meetings -- I *so* don't want special use names to become an
attractive niusence and have legal / trademark fights when someone
launches an alternate name resolution system for finding drugs and
calls it 'coke.alt'.

Having a place where I could go figure out what piece of software I
need to install to resolve http://0xdeadbeef.kitten.alt would be
really useful (even if the resource said that this could be any of 3
different alternate resolution methods - if it looks like a bunch of
hex is is probably KittenNet (install KittenRes0.23.tgz), if the
string is mainly badly spelt "words" it's likely LoLCat, try install
ICanHazNames from http://example.net :-)).

>
> As I mentioned before, given that the whole point of .alt is that
> people are implementing things that look like DNS names but are
> resolved in some other way, the winner of any such conflict is the one
> with widely used running code.

Yah. If I'm launching a new namespace that resolves based upon
<something>, I have an incentive to choose a string that isn't already
being used by some other large, well known project, in the same way
that it would be silly for me to write a new UNIX program that does
something like cowsay (but with kittens) and call it 'cat'.

W

>
> R's,
> John
>
> _______________________________________________
> DNSOP mailing list
> DNSOP@ietf.org
> https://www.ietf.org/mailman/listinfo/dnsop

-- 
I don't think the execution is relevant when it was obviously a bad
idea in the first place.
This is like putting rabid weasels in your pants, and later expressing
regret at having chosen those particular rabid weasels and that pair
of pants.
   ---maf