James Huber <[EMAIL PROTECTED]> asked:
>> I show the RSA patent 4,405,829 (expires Sept. 29th 2000) and
>>the "idea" patent 5,214,703 (good for another 10 years) but don't
>>find a patent on rc5 ????
>>
>>Who owns rc5 and how is it restricted?
Bodo Moeller suggested:
> Try 5,835,600 or 5,724,428 (I haven't looked closely which one is actually
RC5).
Both. Ron Rivest actually has two RC5 patents, and both are
commonly entitled, "Block encryption algorithm with data-dependent
rotations." It was RC5's particular technique for data-dependent
rotations that drew most of the attention from the crypto community, but the
key-schedule table -- integral to the design of RC5 -- was broken out and
patented separately, at the request of the US Patent and Trademark Examiner.
Anyone can find them with a patent search under "Rivest, Ronald" on
IBM's Intellectual Property site: <http://patent.womplex.ibm.com/>
Mr. Huber then got excited:
>Got it,
> YES ! RSA has again shot itself in the foot. Understand that
>I'm not a lawyer either, however as a Manager of Software development
>I have always been told by the legal boys, you MUST enforce a
>copyright and/or patent in ALL cases or you can't enforce it in any.
>US law requires equal treatment to all.
Un huh.
> To that end, RSA placed RC2 & RC5 in the "public domain" when
>they contributed it to the US Government backed PKCS efforts with
>the following statments taken right from their web-site:
Rivest invented RC2 in 1987. It was controlled and licensed as a
trade secret by RSADSI for many years, but in 1992, Ron published an
Internet RFC describing it, effectively putting it in the public domain.
This was part of RSA's drive to promote its S/MIME protocol for
secure encrypted mail as an IETF standard, at the request of Netscape. RC2
was already widely deployed in S/MIME implementations from a dozen-odd
international vendors, proprietary software like Lotus Notes and MS Windows
-- as well, of course, as the SSL suites in browsers and web servers.
>4. The "Intellectual Property Issues" section shall also include
> text stating the Contributor's intent for licensing its patents and
> trademarks (existing and future) that may apply to the contribution.
> The text may take one of the following forms (following the practice
> of IEEE Standards):
>
> a) A general disclaimer to the effect that the Contributor will
> not enforce any of its present or future patent(s) or trademarks
> whose use would be required to implement the Contribution or
> portions thereof if incorporated in PKCS documents.
>
> b) A statement that a license will be made available to all
> applicants without compensation.
>
> c) A statement that a license will be made available to all
> applicants under reasonable rates, with reasonable terms
> and conditions that are demonstrably free of any unfair discrimination.
>
>RSA when proposing it's replacemanet to the NSA for DES initially
>elected for option "A", then tried to play a game and claim option
>"B".....Now officially is trying to claim option "C".
Ummm. Don't think so. I'm not sure what you are referring to with
your reference to the "government backed PKCS" documents. (IEEE? ANSI X9?)
You might want to review the RSA Labs stuff on RSA's PKCS:
<http://www.rsasecurity.com/rsalabs/>
The typical reference point for PKCS in modern crypto, particularly
off the RSA Security website, would be to the famous series of Public Key
Cryptographic Standards (PKCS) that RSA developed (most in the early 1990s)
as proprietary but freely available implementation standards.
RSA had to do it, just because the NSA and its Allied intelligence
agencies had squashed all efforts to have the various US and/or
international standards-making organizations define open standards for any
public key cryptosystems. (Mind you, RSApkc is not patented outside the US.)
There has seldom been a more influential series of best-practice
models. I think almost anyone would agree that PKCS #1 thru PKCS #14 laid
the basis for all subsequent efforts to standardize PKC, PKI, and
Certificate definition, management, and practice -- but to say the PKCS
development was "government backed" is to turn history upside down. And
miss the point besides.
(RSA's relationship with those who want access to its algorithms
free or cheaply is a family food fight. RSA relationship with the NSA,
until maybe two years ago, was war. The NSA tried for most of a decade to
crush RSADSI, displace it in the market, or get direct control of the
RSApkc patent. And what would have happened if they had succeeded, you
might ask;-)
> Oops, that doesn't work in a court of law and would explain
>why RSA has failed to ever enforce this (MD2/5) or the more
>common RSA patents....They can't.
Ummmm. Ron Rivest has freely published his message digest (MD)
series of hashes for public use. Turns out that good hash algorithms are
pretty rare.
Ron published his MD4 Message Digest Algorithm (RFC 1320) and MD5
(RFC 1321) both in April, 1992. Those were the only two that came into
widespread use. RSA has also recommended, for several years now, that
implementors graduate to (MD5-based) SHA-1 for most applications.
AFAIK, neither Prof. Rivest nor his RSADSI ever made any attempt to
enforce an intellectual property claim on the MD hashes. See Rivest's
website at MIT: <http://theory.lcs.mit.edu/~rivest/publications.html>.
I'd be surprised if you have found anyplace -- other than the NIST's
AES competition, where Rivest's RC6 is a long-shot candidate to replace the
DES -- where RSA has offered RC5 (or any patented RC6 internal to RC6) free
of its intellectual property entanglements. RSA has been defining IP in
crypto for a few years now, and their lawyers claim they've gotten better at
it. Not perfect, but better;-)
(If RC6 is selected as the next AES, said RSAS, the company will
concede all patent rights associated with RC6. It has, I believe, withstood
considerable pressure from the US government and others which attempted to
force RSAS, now, to renounce any and all patent claims it might have on any
other AES candidate algorithms. Some of the other candidates may rely upon
RC5's patented method of data-dependent rotations.)
> To further shoot their own foot off, they included MD5 in the
>public releases of RSAREF a full 4 years BEFORE they were granted
>a patent and included no claims as to a patent being pending.
Un huh. All that foot-shooting.
There never were any patents -- never any attempts to patent --
Rivest's hashes, AFAIK. That's MD5, the free hash. RC5 is the patented
symmetric algorithm. RC6 -- unless the NSA does something really unexpected
and lets NIST make RC6 the AES -- will probably be patented too.
>And the RSAREF license forces them option "B" for commercial use.
>They are refusing to even speak to anyone regarding it....Cool.
>That means to me.....it's mine to use as long as I make a formal
>in writing statement to the registered offices of their company
>they must either produce the original license and terms or forfit
>their right to do so.
Un huh. Bound to be interesting.
I do hope you don't forget to point to the List about the experience.
> We will be making it official in writing just as soon as the
>legal types draw up the papers. This covers only our company but
>anyone else in the US should be free to do the same.
Un huh.
Suerte,
_Vin
______________________________________________________________________
OpenSSL Project http://www.openssl.org
User Support Mailing List [EMAIL PROTECTED]
Automated List Manager [EMAIL PROTECTED]
