sorry, but ...

[Andrew Bangs]

Intercepting at switches (vs literally 'on the wire') always reminds me of Zeno's arrow paradox. We somehow have to pretend that the thing to be intercepted is stationary if this is really different from doing it on the wire, or at least that it's received and then has stuff done to it before being sent on its way. Leaving aside the issue that a single packet in a single direction might not be the same as a communication (and I can't remember if that's important) a switch could decide what to do with a packet based on the first portion of the packet (and addressing information is usually conveniently placed at the front, rather than at the back, of a packet) and be already transmitting it to the next element in the network before the tail of the packet has been received. 


In other words, there isn't necessarily a point in time after a switch has received a message and before it sends it on again. Yes, a switch is a place between the wires, but it doesn't necessarily have the entire 'message' at any instant. 


(for anyone thirsty for further reading, "cut-through switching" is a phrase your favourite search engine could investigate)

Regards,
Andrew

________________________________
From: Peter Fairbrother <zenadsl6186 at zen.co.uk>
To: UK Cryptography Policy Discussion Group <ukcrypto at chiark.greenend.org.uk> 
Sent: Tuesday, 31 July 2012, 23:03
Subject: Re: sorry, but ...

Err, while I do not think you can intercept at each and every switch in the network, I have to disagree a little on point - there is little I can see in RIPA which implies or involves flowing, except between two places.

A switch is a place - intercepting the wires between switches is not the same as intercepting at a switch after the switch has received the message and before it sends it on again.