- From: Kristof Zelechovski <giecrilj@stegny.2a.pl>
- Date: Sat, 12 Sep 2009 22:55:33 +0200
- To: "'Lloyd Wood'" <L.Wood@surrey.ac.uk>, "'Infinity Linden'" <infinity@lindenlab.com>
- Cc: <hybi@ietf.org>, <uri-review@ietf.org>, "'Daniel R. Tobias'" <dan@tobias.name>, "'URI'" <uri@w3.org>, "'Lisa Dusseault'" <lisa.dusseault@gmail.com>
1. DNS can use UDP. 2. There is no need to do interplanetary name lookup each time; the local resolver can have all desired hosts preconfigured. 3. Even if this fails, you can have numeric addresses in an URI authority, so it does not preclude using URI if only numeric addresses are feasible. (BTW, domain names are for people; appliances do not need them at all unless their host has an unstable IP number.) IMHO, Chris -----Original Message----- From: uri-review-bounces@ietf.org [mailto:uri-review-bounces@ietf.org] On Behalf Of Lloyd Wood Sent: Saturday, September 12, 2009 12:01 AM To: Infinity Linden Cc: hybi@ietf.org; Lloyd Wood; uri-review@ietf.org; Daniel R. Tobias; URI; Lisa Dusseault Subject: Re: [Uri-review] [hybi] ws: and wss: schemes [snip] But URIs in space? URIs are predicated on an authoritative DNS. In deep space, DNS as it stands is likely not to be useful - lookups take too long, contacts are scheduled and planned, late binding, rather than early binding, is needed. And TCP's assumptions and timers aren't a good fit with the environment. But that doesn't rule out IP or other protocols in the suite.
Received on Saturday, 12 September 2009 20:56:34 UTC