[68031] in North American Network Operators' Group
Re: How reliable does the Internet need to be? (Was: Re: Converged Network Threat)
daemon@ATHENA.MIT.EDU (Stephen Sprunk)
Sat Feb 28 05:24:05 2004
From: "Stephen Sprunk" <stephen@sprunk.org>
To: "Steve Gibbard" <scg@gibbard.org>
Cc: "North American Noise and Off-topic Gripes" <nanog@merit.edu>
Date: Sat, 28 Feb 2004 04:17:11 -0600
Errors-To: owner-nanog-outgoing@merit.edu
Thus spake "Steve Gibbard" <scg@gibbard.org>
> When sending somebody e-mail you assume they'll probably check their
> e-mail and receive the message eventually, but you have no idea if they'll
> get it right away, or if they'll notice it along with all the other e-mail
they get.
> When phoning somebody, you know right away whether they answer,
> and you know right away how they respond to whatever you have to say.
In theory, that's a job for DSN.
> If you really need to get in touch with somebody right now, do you call
> their presumably more reliable land line, or their presumably less
> reliable cell phone?
I think this was intended as rhetorical, but I'll bite... If I want to
reach someone _right now_, I call their cell phone, because the odds are
significantly higher they'll be near it.
E911 is a mess even for non-mobile users in a single office building. Full
compliance, at least with my state's regulations, requires phone numbers be
localized to within a 100ft2 space. This means a VoIP system must be able
to automatically trace a user's IP address, MAC address, physical port,
cable run, and at least portion of a floor -- not to mention figuring out
what the correct telco circuit to route the 911 call out, which obviously
varies depending on where the user happens to be sitting.
S
Stephen Sprunk "Stupid people surround themselves with smart
CCIE #3723 people. Smart people surround themselves with
K5SSS smart people who disagree with them." --Aaron Sorkin
