On Fri, 22 Aug 1997, Tim May wrote:
We haven't discussed this point in a while, but the belief most cryptologists have is roughly this:
The world-wide community of mathematics and cryptology researchers, linked through open publication of new research results, is GREATER than the cloistered NSA and GCHQ communities of researchers.
Thus, as bright as Brian Snow or Don Coppersmith or John Conway may be, the "edge" the NSA may have once had is largely gone. Which is not to say that they are not still a formidable technical organization, with substantial computer resources.
This also applies to the substantial computer resources and technology as well. Something on TLC got me thinking about this even before I read this message. The NSA has a problem in that, except maybe for quantum cryptography, they no longer have an advantage of kind, merely of degree, and the market is narrowing that gap daily. Before, only they had the resources to do something like the machine NCR built to crack enigma messages, and could build many one-of-a-kind machines to do individual cracking. They still can, but it isn't efficient to do so today. They can build 1000 custom ASICs, but they will be more expensive than 10,000 off-the-shelf CPU chips - they can't do engineering any better or cheaper than Intel or DEC, and they don't have millions of customers to spread the fixed costs over. When a camcorder is more complex than most weapons systems, but is available at the local mall, and when I can buy SMP servers from an 800 number, I have the same thing the NSA has, only smaller. But the NSA is still a finite size (given the earlier posts about terawatts), and I can link my computer with thousands or millions of others. DEScrack was only done on a relatively few computers. If a really big prize was offered (In this lotto, you just run this screen saver which uses less bandwidth than pointcast... The chinese radio lottery via the internet), you could get almost every computer into the act. As technology goes forward, any advantage of largess will be overcome by greater numbers of small systems - at some point the large college campuses will have more cpu cycles than the NSA because students bring the latest technology with them. Unless the NSA is radically different, the concept of putting a SMP system on everyone's desk (upgrading it every few years) and linking them isn't going to go over as well as getting 100 new cray supercomputers. --- reply to tzeruch - at - ceddec - dot - com ---