[from somelist]

Subject: Re: [s-t] The return of Das Blinkenlight Date: Mon, 31 Jan 2005 19:00:49 -0500

...

In the early 90's I was a product manager for a (now-defunct) company that made LAN hubs-- this was when a 10Base-T port would cost you a couple

This reminded me of a story from a few years ago.

Apparently a lot of modem manufacturers tied the activity light on the modem directly to the circuit which modulated the sound.

Then someone realized that with a telescope, and and optical transister, one could read that datastream as if hooked to the modem directly.

And astonishing numbers of businesses had their modem pools facing windows, because the blinkenlights looked impressive.

<http://applied-math.org/optical_tempest.pdf> Not just modems. Some Cisco routers, even at megabit rates. 2002 publication, although the research was over the previous couple of years. And (for instance) the Paradyne Infolock 2811-11 DES encryptor, which has an LED on the plaintext data. How we laughed. The paper also covers using LEDs (such as keyboard LEDs) as covert data channels. And yes, it cites Cryptonomicon. I'm not sure whether this was more or less cool than Marcus Kuhn's work on reconstructing CRT displays from reflected light, by reverse convolution with the impulse-response curves of the various phosphors. Both papers are fantastic reads, very accessible, very stimulating. <http://www.cl.cam.ac.uk/~mgk25/ieee02-optical.pdf> Nick B ----- End forwarded message ----- -- Eugen* Leitl <a href="http://leitl.org">leitl</a> ______________________________________________________________ ICBM: 48.07078, 11.61144 http://www.leitl.org 8B29F6BE: 099D 78BA 2FD3 B014 B08A 7779 75B0 2443 8B29 F6BE http://moleculardevices.org http://nanomachines.net [demime 1.01d removed an attachment of type application/pgp-signature]