On Thu, Apr 7, 2016 at 5:34 PM Adrien de Croy <adr...@qbik.com> wrote:
> > ------ Original Message ------ > From: "David Conrad" <d...@virtualized.org> > To: "Philip Homburg" <pch-dn...@u-1.phicoh.com> > Cc: "dnsop@ietf.org" <dnsop@ietf.org> > Sent: 8/04/2016 12:38:26 a.m. > Subject: Re: [DNSOP] Alternative Special-Use TLD problem statement draft > > > Philip, > > On Apr 7, 2016, at 7:57 AM, Philip Homburg <pch-dn...@u-1.phicoh.com> > wrote: > > However, having lived through a period where many different naming > systems where > use in parallel and seeing what benefits it brought to have to consider > just DNS, > I think a model where IETF and ICANN actively control the consistency of > the > internet name space is best. > > > I do not believe IETF or ICANN have that level of control. The Internet is > known for "permissionless innovation" for a reason. > > > I have created naming systems myself. And there are many reasons to > dislike > DNS and do something different. > > > Right. > > > But ultimately, for the stability of the internet, it is best to not do > that. > Write some experimental code, write a few papers and be done with it. > Then, > if you still care about the problem, work within the IETF to improve DNS. > > > I believe the point of the special use registry is that these are > protocols that are not DNS and have no interest in being in the DNS, but > which make use of domain name conventions. The alternative to the special > use registry is not that such names won't exist, rather it is that the > names will collide with names in the DNS. > > > > > > > the question is whether we should fundamentally change the domain name > system to accommodate this or not. My position is we should not. > > Just because TOR asks for .onion doesn't mean it should be given it. > > I understand the IETF is supposed to obtain consensus, but I didn't see > anything in http WG on this until after the fact. Special use names has > wide-ranging repercussions. > > If an author of a product chooses to devise a new non-dns resolution > mechanism, and use names that are confusingly similar to domain names, and > use names that may be issued by ICANN in the DNS, then I'm sorry but they > need to look for an alternative technical solution. > ... or choose a different string(s) so that they are *not* confusingly similar to the DNS. If someone designs a product which uses the suffix string .c0m the response could be "That's a security (phishing) issue, how abut you instead use <something else>?". .onion A: was not a name which had been applied for nor existed in the DNS and B: had a large installed base. > Not make everyone else in the world change to suit them. > Yup. However (even though I have a badge which claims otherwise) we are not the Internet police. Using a string to delineate "your" portion of a namespace is not something that the IETF can prevent ( .onion launched a sucessful resolution system without "us") - if we provide a process which has some reasonable review, and is not ridiculously painful to follow, we can maintain a registry of these. If the process is difficult / arcane / etc we will simply have "squatters". W > > Regards > > Adrien > > > > > > Regards, > -drc > (speaking only for myself) > > > _______________________________________________ > DNSOP mailing list > DNSOP@ietf.org > https://www.ietf.org/mailman/listinfo/dnsop >
_______________________________________________ DNSOP mailing list DNSOP@ietf.org https://www.ietf.org/mailman/listinfo/dnsop
