home *** CD-ROM | disk | FTP | other *** search
- Newsgroups: sci.crypt
- Path: sparky!uunet!haven.umd.edu!darwin.sura.net!zaphod.mps.ohio-state.edu!rpi!newsserver.pixel.kodak.com!laidbak!tellab5!cuuxb!nsscmail!ulysses!ulysses!smb
- From: smb@ulysses.att.com (Steven Bellovin)
- Subject: Re: RSA marketing weakness or lack of demand?
- Message-ID: <1992Nov6.195347.24435@ulysses.att.com>
- Date: Fri, 6 Nov 1992 19:53:47 GMT
- References: <1992Nov4.195416.4015@netcom.com> <iDLTTB12w165w@mantis.co.uk>
- Organization: AT&T Bell Laboratories
- Lines: 64
-
- In article <iDLTTB12w165w@mantis.co.uk>, mathew <mathew@mantis.co.uk> writes:
- > Well, let's see. I believe RSA was developed at MIT, and paid for by the US
- > taxpayer. I understand that when it was announced, the US government
- > attempted to cover up the whole business and bury the technology, but that
- > they failed because A.K. Dewdney of Scientific American and various other
- > people mailed out photocopies of the appropriate papers to all and sundry,
- > having been 'leaked' copies of the articles before or shortly after initial
- > publication.
-
- That story appears to be false. See ``Kahn on Codes'', p. 200.
- Briefly, an ``eccentric'' NSA employee wrote to the IEEE warning them
- that a conference session on cryptology was in violation of the law.
- For a time, MIT suspended distribution of (I think) the RSA paper. The
- Senate Intelligence Committee later investigated, and determined that
- this employee was acting on his own, not in his official capacity or
- with the approval of his superiors. Kahn appears to believe that,
- incidentally.
-
- > I gather that it was then arranged for a series of patents to be applied for
- > retrospectively. (That is, I'm told, why the patents are not valid outside
- > the US; they were applied for after publication.) These patents were
- > apparently handed on a plate to PKP, a company which seems to have done very
- > little other than threaten litigation.
-
- U.S. patent law permits applications to filed up to one year after
- publication. There is nothing underhanded about that; it's simply
- different than in the rest of the world. And it applies to everyone --
- I know someone who's applying for a patent under just those circumstances,
- and it has nothing whatsoever to do with security or cryptography.
-
- As for what PKP is -- it's a subsidiary (not wholly-owned, I think,
- though I could be wrong) of RSADSI. Its purpose in life is to hold and
- license public key patents. They're not supposed to do anything else.
- RSADSI does development; you're perfectly free to claim that their code
- is lousy, or that they do a bad job of marketing stuff, but they do
- exist. PKP's power comes from the fact that it holds the licensing
- rights to both RSA and Diffie-Hellman -- an arrangement that's arguably
- monopolistic, and hence challengeable in court. I don't know of any
- government connection.
-
- > Although they're a completely independent organization, and nothing at all to
- > do with the US government, PKP seem strangely reticent to license the RSA
- > patents to ordinary citizens who want to use RSA. In fact, they refuse point
- > blank to license the patents to people who want to use PGP. Odd, eh?
-
- That's probably a business decision. Look at it from their perspective --
- how would they enforce their license? This is freeware, code that's
- all over the Internet. How would they collect their royalties? And it's
- in competition with PEM, from which they will profit, and which has a
- built-in royalty payment structure, via certificate recertification.
- (To see what I mean here, go read the (draft, I think) RFCs on PEM, and
- see how the certificate-signing units are going to be built. They'll
- be in tamper-resistant boxes that send PEM mail back to the owners
- every time a certificate is signed!)
-
- Again -- you may not like their business decisions, nor think they make
- much sense. But hey -- American law doesn't guarantee you a profit
- from your patents, nor even that you won't go bankrupt (unless, of course,
- you're an S&L, or Chrysler...). They have their own strategies. But
- I think you're being unnecessarily paranoid. Want something to be
- paranoid about? Look at the encryption scheme just blessed for export
- by the Software Publisher's Association and NSA -- and at the key
- lengths that NSA will permit to go out, including, as I recall, RSA.
- What does that tell you?
-