domo@tsa.co.uk (Dominic Dunlop) (11/21/90)
In article <5174@tardis.cs.ed.ac.uk> gtoal@tardis.cs.ed.ac.uk (Graham Toal) writes: > I am very much against that Cobol hacker Nicholson's > campaign against computing, with all the half-truths and veiled hints > she drops about 'information from the security services'. I specifically > object to her trying to get powers for tapping data lines. I saw an > advert in a computing mag once for a device which scans telephone lines > looking for calls with modem data on them. Yes. Isn't Emma wonderful? A credit to the British Conservative party, and all that. So much for politics (and note the Follow-up-To). What strikes me is that high-speed modems which rely on echo-cancellation to work at all should be fairly difficult to tap -- unless one can get access to the signal while it's travelling four wires rather than two. On two wires, the send and receive signals are mixed up, and, as I understand it, almost impossible to disentangle without precise knowledge of one end of the conversation. (Which, or course, the modems at either end of the circuit have, and feed into their echo cancellers.) Governments inevitably have the funding to do the almost impossible, and can get access to four-wire circuits beyond the local loop. But I'd expect the task of cracking V.32 or TrailBlazer exchanges to be beyond the scope of an off-the-shelf box attached to a telephone line with a couple of crocodile clips. Comments, anybody? -- Dominic Dunlop