Variola: You say a lotta good shit here, but you're really out of your area in this case. You seem to miss the basic points, and then fill in your blindspot with pure theoretical conjecture. Let me point out some of the lil' flaws in your thinking....
With all due respect, you think Ft. Meade uses the same COTS crap as you are forced to deal with? Bwah hah hah.
For some things, sure. Actually I know from first hand experience. (I've actually been in an NSA, DISA, and a few other experimental network nodes.) Lots of the equipment I saw was from the big vendors, most notably Lucent and Nortel. Somewhere deeper than I had access to, however, they almost certainly use special silicon.
Gilmore et al used a bunch of old Sun Chassis for his & Kocher's DEScracker. You think this is somehow more than 100 watts, in a diplo suitcase, nowadays?
OK, so you're saying that this suitcase takes in say 10 OC-192s, demuxes all of them down to the DS1 level (we're at 50,000 DS1s), demaps and unpacks the ATM cells, and then reassembles all of the packets therein? Questions: 1) How does this majic box store all that data? 2) I've been in dozens of COs myself, and have worked extensively with people who have spent (collectively speaking) centuries in them. They never saw such a magic box a you describe, and indeed would certainly know about someone trying to install one. Or perhaps the NSA has developed a cloaking device making the box invisible? 2) What silicon does it use? Are you saying that the government can do a LOT better than 0.13 microns these days? Somehow I doubt it. Look at the off-the-shelf SONET chip architectures. Sure, there's lots of stuff onboard that you wouldn't need for what you're talking about, but getting rid of that stuff would still put the most advanced chip lightyears behyind what you're talking about. 3) If the majic box doesn't store the data, how does it get it back to HQ? Telepathy? Or, does it use a bank of lasers that somehow are several orders of magnitude more efficient that off-the-shelf lasers? (And let us remember that there's a fundamental constraint with bulk optics..an optical multiplexer or ciculator can't be an order of magnitude smaller than the wavelength it will support.) JA's comments about fiber exhaust are dead-on, and were not known to most of the Telecom Bubble participants. (Indicates the dude knows what he's talking about with respect to telecom.) But dark fibers aren't a real concern. It would be easy to develop a DWDM system that operated over the L or M bands, "under" the C-band wavelengths used by a carrier. So the problem isn't the fiber, it's lighting it. As for my comments about cable landings, I explicity stated that the splices back to VA were seen and known. And yes, I was in a position to know. (There's not a lot you can hide in a CO...it's not like they staff them with NSA agents or something.) As for trolling, well when I do it I do it with friggin' style m'friend. But sometimes, the truth is so mundane it looks fairly boring. Sorry to dissappoint you. I'm going to have to confiscate your copy of "Deepness in the Sky"... -TD _________________________________________________________________ MSN Toolbar provides one-click access to Hotmail from any Web page FREE download! http://toolbar.msn.click-url.com/go/onm00200413ave/direct/01/