[9934] in Commercialization & Privatization of the Internet
re: Internet "PayPhones"
daemon@ATHENA.MIT.EDU (Per Gregers Bilse)
Thu Jan 27 15:41:27 1994
From: Per Gregers Bilse <bilse@eu.net>
Date: Thu, 27 Jan 1994 21:40:32 +0200
In-Reply-To: <9401271548.AA05591@iscmed.med.ge.com>
To: com-priv@psi.com
On Jan 27, 9:48, Larry Walker wrote:
> >need which many travellers have. What we'd need for something
> >else than a cmd-line is genuine mobile IP, which isn't here yet.
>
> Interesting point. But why not use dynamic IP address assignment on the
> dialin routers that your mobile users call in to?
It was considered, but would limit (the initial version of) the
service to people with 'IP-capable' machines. Currently, that leaves
a lot of people out in the dark.
Personally, I also don't think cmd-line email is such a terrible
thing.-/
> >adventurous users can run whatever protocol they want on top of
> >the telnet session.
>
> This is an interesting concept: I've heard references to running PPP over a
> telnet session running on an async dialup connection. Do you have any
> experience with attempting this?
I haven't tried it personally, but the experience of my colleague is
that it works, but leaves something to be desired in the performance
stakes. Low-speed IP/PPP is clunky enough already; adding a telnet
layer (and another IP layer) gives a bit too much wrapping around
your data for good response. In particular the telnet layer would be
likely to break traffic into a lot of small packets. But for
non-interactive traffic this doesn't make much difference, though; it
just slows things down.
--
bilse <bilse@EU.net> +31 20 592 5109 (dir: 5110); fax +31 20 592 5163
