Eric Murray <[EMAIL PROTECTED]> queried the Listocracy:
>Does anyone know the legal status of RC4 in the US?
>
>I know that a cipher purporting to be RC4 was published on
>Cypherpunks by Anonymous, and that various crypto packages
>have RC4 or "EC4". My question is, has RSA taken anyone to
>court in the US for using RC4 without buying a license from RSA?
To my knowledge, neither RSADSI, nor the new firm formed by the
combination of RSADSI and Security Dynamics last year -- RSA Security, Inc.
-- has ever dragged any individual or enterprise into court for using an
unlicensed implementation of RC4.
I may have missed something over the years, but there are today an
abundance of commercial products on the market which use ARC4, "Apparently
RC4," or some similar independent implementation of Rivest's RC4 -- with no
tithe to RSA, and no kickback that I can see. (There was a fascinating
discussion of all this a year or two ago, on either Cypherpunks or here on
C2's Cryptography. Search for an evocative subject-line: "None Dare Speak
its Name," or "The Cipher None Dare Name.")
RSA is reportedly far more conscientious in defending its trademark
on the label "RC4." RC2, RC4, RC5, RC6 -- and probably MD2, MD4, and MD5,
Rivest's freeware message digests or hashs -- are registered trademarks.
("RC," Rivest once told me, simply stands for "Ron's Code" -- a workbench
label for crypto in development that somehow escaped into the hands of the
RSA marketing guys.)
I suspect that RSA did send out more than a few nastygrams to OEMs
or other mass marketeers about "illicit use" of RC4, but -- at least in
recent years -- its complaints probably went to commercial enterprises which
both (a) sought to resell the algorithm in the US, and (b) blatently used
the RC4 label in a way that is likely to confuse many people as to the
source of the RC4 implementation code. In the real world, RC4 is today
almost exclusively associated with the implementation of RC4 in the RSA
BSAFE toolkits, which have been licensed to some 700 OEMs, designed into
thousands of products, and installed in a half *billion* user machines.
[Gothic legend to the contrary, I have also never heard of RSA
rousting _any_ US firm or individual who used their own unlicensed
implemention of RSApkc (and RC2 and RC4) in various homebrew SSL grafts, or
in the famous freeware SSL kits: SSLeay or OpenSSL. The elegant design of
RC4 has been widely studied and discussed in academia.]
[Vendors of commercial products which include various ARC4
implementations putz along untouched. The IETF also has several RFCs which
document and refer to various independent and equivalent ACR4
implementations. Indeed, in order to market Eric Young's BSAFE SSL-C
toolkit out of RSA-Australia, Young -- the "eay" of SSLeay in a previous
life, now the CTO of RSA-Australia -- had to prove to both the Australian
and US export control mavens that Young's implementation of RC4 for SSL-C
was based on wholly non-RSADSI and non-American sources.]
Of course, back in the early-90s when the reverse-engineered RC4
code was first anonymously posted to the Net, it immediately became clear
that the old combination of software license, trade secret status,
copyright, and trademark IP with which RSADSI had tried to protect and
control access to Rivest's RC4 algorithm was an utter failure.
As I think the insightful Greg Boiles, Esq., has pointed out
elsewhere, the possibility of unattributed global publication guts many of
the traditional IP defenses for commercial know-how or technical savvy in
commercial products.
Anyone who can deconstruct or reverse-engineer a proprietary and
secret design or formula -- be it the Coca Cola formula or Ron Rivest's RC4
-- can use the Net to duck the retribution that was at the heart of most
traditional IP defenses. (Ya gotta have a 16 year-old's sense of
invulnerability to sign your name to the deed, be you a DVD devil, an angel,
a curious teen scientist, or just a guy who doesn't like to pay royalties to
artists or inventors;-)
The fate of RC4 -- the anonymous post to the Net has led to the
widespread use of unlicensed "ARC4" in hundreds of commercial software
products today -- led RSADSI (and many other corporations) to conclude than
only patents could effectively protect software IP.
In 199o, there were 1,300 US patents issued for software
innovations. In 1999, there were 22,500 software patents issued. <sigh> I
have always believed the rogue publication of RC4 was a major accelerator in
this trend.
With the threat of anonymous publication of trade secrets on the
Net, it seems inevitable, at least in hindsight, that patents were to become
more important; more broaded applied; and more broadly construed.
Ain't nothin' else, folks, which can define and defend an
intellectual property claim for a novel and non-obvious invention -- while
assuming that everyone can access and study the innovation itself. The Net
makes it impossible to identify who spills a secret. A patent make secrecy
unnecessary.
RSA, of course, got two patents on RC5, the elegant symmetric cipher
Ron Rivest developed next. I presume there are also patents pending on RC6,
Rivest's AES candidate (although RSA, like other AES candidates, promised
free access for all in the unlikely event that RC6 becomes the AES.)
Suerte,
_Vin
PS. As I presume is obvious, only RSA Security speaks for RSA on this
touchy subject. IANAL. I offer no legal advice. I have also been a
consultant to RSA for many years, which has doubtless warped my judgement.
