compress/doc/NOTES

   1
   2   $FreeBSD: src/usr.bin/compress/doc/NOTES,v 1.2 2002/10/16 12:42:15 charnier Exp $
   3
   4 From: James A. Woods <jaw@eos.arc.nasa.gov>
   5
   6 >From vn Fri Dec  2 18:05:27 1988
   7 Subject: Re: Looking for C source for RSA
   8 Newsgroups: sci.crypt
   9
  10 # Illegitimi noncarborundum
  11
  12 Patents are a tar pit.
  13
  14 A good case can be made that most are just a license to sue, and nothing
  15 is illegal until a patent is upheld in court.
  16
  17 For example, if you receive netnews by means other than 'nntp',
  18 these very words are being modulated by 'compress',
  19 a variation on the patented Lempel-Ziv-Welch algorithm.
  20
  21 Original Ziv-Lempel is patent number 4,464,650, and the more powerful
  22 LZW method is #4,558,302.  Yet despite any similarities between 'compress'
  23 and LZW (the public-domain 'compress' code was designed and given to the
  24 world before the ink on the Welch patent was dry), no attorneys from Sperry
  25 (the assignee) have asked you to unplug your Usenet connection.
  26
  27 Why?  I can't speak for them, but it is possible the claims are too broad,
  28 or, just as bad, not broad enough.  ('compress' does things not mentioned
  29 in the Welch patent.)  Maybe they realize that they can commercialize
  30 LZW better by selling hardware implementations rather than by licensing
  31 software.  Again, the LZW software delineated in the patent is *not*
  32 the same as that of 'compress'.
  33
  34 At any rate, court-tested software patents are a different animal;
  35 corporate patents in a portfolio are usually traded like baseball cards
  36 to shut out small fry rather than actually be defended before
  37 non-technical juries.  Perhaps RSA will undergo this test successfully,
  38 although the grant to "exclude others from making, using, or selling"
  39 the invention would then only apply to the U.S. (witness the
  40 Genentech patent of the TPA molecule in the U.S. but struck down
  41 in Great Britain as too broad.)
  42
  43 The concept is still exotic for those who learned in school the rule of thumb
  44 that one may patent "apparatus" but not an "idea".
  45 Apparently this all changed in Diamond v. Diehr (1981) when the U. S. Supreme
  46 Court reversed itself.
  47
  48 Scholars should consult the excellent article in the Washington and Lee
  49 Law Review (fall 1984, vol. 41, no. 4) by Anthony and Colwell for a
  50 comprehensive survey of an area which will remain murky for some time.
  51
  52 Until the dust clears, how you approach ideas which are patented depends
  53 on how paranoid you are of a legal onslaught.  Arbitrary?  Yes.  But
  54 the patent bar the the CCPA (Court of Customs and Patent Appeals)
  55 thanks you for any uncertainty as they, at least, stand to gain
  56 from any trouble.
  57
  58 =-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=
  59 From: James A. Woods <jaw@eos.arc.nasa.gov>
  60 Subject: Re: Looking for C source for RSA (actually 'compress' patents)
  61
  62         In article <2042@eos.UUCP> you write:
  63         >The concept is still exotic for those who learned in school the rule of thumb
  64         >that one may patent "apparatus" but not an "idea".
  65
  66 A rule of thumb that has never been completely valid, as any chemical
  67 engineer can tell you.  (Chemical processes were among the earliest patents,
  68 as I recall.)
  69
  70         ah yes -- i date myself when relaying out-of-date advice from elderly
  71         attorneys who don't even specialize in patents.  one other interesting
  72         class of patents include the output of optical lens design programs,
  73         which yield formulae which can then fairly directly can be molded
  74         into glass.  although there are restrictions on patenting equations,
  75         the "embedded systems" seem to fly past the legal gauntlets.
  76
  77         anyway, I'm still learning about intellectual property law after
  78         several conversations from a Unisys (nee sperry) lawyer re 'compress'.
  79
  80         it's more complicated than this, but they're letting (oral
  81         communication only) software versions of 'compress' slide
  82         as far as licensing fees go.  this includes 'arc', 'stuffit',
  83         and other commercial wrappers for 'compress'.  yet they are
  84         signing up licensees for hardware chips.  Hewlett-Packard
  85         supposedly has an active vlsi project, and Unisys has
  86         board-level LZW-based tape controllers.  (to build LZW into
  87         a disk controller would be strange, as you'd have to build
  88         in a filesystem too!)
  89
  90         it's byzantine
  91         that Unisys is in a tiff with HP regarding the patents,
  92         after discovering some sort of "compress" button on some
  93         HP terminal product.  why?  well, professor Abraham Lempel jumped
  94         from being department chairman of computer science at technion in
  95         Israel to sperry (where he got the first patent), but then to work
  96         at Hewlett-Packard on sabbatical.  the second Welch patent
  97         is only weakly derivative of the first, so they want chip
  98         licenses and HP relented.  however, everyone agrees something
  99         like the current Unix implementation is the way to go with
 100         software, so HP (and UCB) long ago asked spencer Thomas and i to sign
 101         off on copyright permission (although they didn't need to, it being pd).
 102         Lempel, HP, and Unisys grumbles they can't make money off the
 103         software since a good free implementation (not the best --
 104         i have more ideas!) escaped via Usenet.  (Lempel's own pascal
 105         code was apparently horribly slow.)
 106         i don't follow the IBM 'arc' legal bickering; my impression
 107         is that the pc folks are making money off the archiver/wrapper
 108         look/feel of the thing [if ms-dos can be said to have a look and feel].
 109
 110         now where is telebit with the compress firmware?  in a limbo
 111         netherworld, probably, with sperry still welcoming outfits
 112         to sign patent licenses, a common tactic to bring other small fry
 113         into the fold.  the guy who crammed 12-bit compress into the modem
 114         there left.  also what is transpiring with 'compress' and sys 5 rel 4?
 115         beats me, but if sperry got a hold of them on these issues,
 116         at&t would likely re-implement another algorithm if they
 117         thought 'compress' infringes.  needful to say, i don't think
 118         it does after the above mentioned legal conversation.
 119         my own beliefs on whether algorithms should be patentable at all
 120         change with the weather.  if the courts finally nail down
 121         patent protection for algorithms, academic publication in
 122         textbooks will be somewhat at odds with the engineering world,
 123         where the textbook codes will simply be a big tease to get
 124         money into the patent holder coffers...
 125
 126         oh, if you implement LZW from the patent, you won't get
 127         good rates because it doesn't mention adaptive table reset,
 128         lack thereof being *the* serious deficiency of Thomas' first version.
 129
 130         now i know that patent law generally protects against independent
 131         re-invention (like the 'xor' hash function pleasantly mentioned
 132         in the patent [but not the paper]).
 133         but the upshot is that if anyone ever wanted to sue us,
 134         we're partially covered with
 135         independently-developed twists, plus the fact that some of us work
 136         in a bureaucratic morass (as contractor to a public agency in my case).
 137
 138         quite a mess, huh?  I've wanted to tell someone this stuff
 139         for a long time, for posterity if nothing else.
 140
 141 james
 142