Re: [I18ndir] I18ndir early review of draft-schanzen-gns-10

[Christian Grothoff <[e-mail]>]

On 6/30/22 21:10, Asmus Freytag wrote:
> On 3/7/2022 10:45 AM, Schanzenbach, Martin wrote:
>> This makes it explicit that GNS names cannot be distinguished from DNS names (or special-use tlds, or any other domain name for that matter).
>> It also addresses the issue of potential leakage of names in another system.
>> Note that is leakage issue is generally a problem also for special-use names (e.g. .onion, .bit etc) but it makes sense to highlight it I think.
> 
> I thought that .onion etc are actually reserved names so there's no 
> confusion with DNS.
> 
> Or did I misread your paragraph.

Not exactly, mostly I think you're missing out on history: .onion is 
reserved _because_ we proposed that it should be. However, we proposed 6 
TLDs to be reserved, and ultimately Facebook pushed ".onion" through 
while all the others were rejected by our corporate overlords (or 
whoever runs the IETF these days). For the gory details, read the RFC 
reserving ".onion" carefully, as well as Kumari's RFC that resulted from 
the 1000-email flamewar that erupted on dnsop over our draft.

The short version: some said "gnu" was trademarked (by gnu.com, not 
gnu.org!) and could not be reserved; some said this was not an IETF 
issue but an ICANN issue; some said one needed to have 1M users of the 
pTLD before IETF would register it (demonstrate use); others said one 
could not just squad on a domain name and would have to ask for the name 
before ever using it; some suggested we should use a different syntax 
(and break compatibility with all existing applications); and some seem 
to believe that DNS is the last name resolution protocol the Internet 
will ever need.

Anyway, after discussing our draft for a few years, IETF approved the 
.onion draft (which basically was identical, except removing the 
reservations for other TLDs) in a matter of months. So Tor's .exit is 
not reserved, because Facebook doesn't need that one. So maybe we should 
write that the leakage issue is also a problem for ".exit", would that 
be better?