🚀 go-pugleaf

In reply to:
>>         By the way, since ATT owns a license, anyone who holds an ATT
>> Unix license can use compress legally, right? ie if company X pays for
>> the license to incorporate patented object Y in product Z, then
>> legitimate users of Z have noting to fear, right?

In article <3315.26b9a3fb@mccall.com> mccall!tp@deimos.cis.ksu.edu askes:
>Did you get compress from AT&T? If you didn't get it as part of the product
>covered by your AT&T license, forget it. If you did, I'd guess it would
>depend on the terms of AT&T's license (relicensing/distribution rights, and
>possible registration requirements).

The following is excerpted from the UNIX man pages supplied by Hewlett-Packard
with HP-UX V6.5 (ATT SysV R3.?)

     COMPRESS(1)                   HP-UX                   COMPRESS(1)

     NAME
          compress, uncompress, zcat - compress and expand data

     SYNOPSIS
          compress [ -d ] [ -f ] [ -v ] [ -c ] [ -V ] [ -b maxbits ] [
          name... ]
          uncompress [ -f ] [ -v ] [ -c ] [ -V ] [ name ... ]
          zcat [ -V ] [ name ... ]

     DESCRIPTION
          Compress reduces the size of the named files using adaptive
          Lempel-Ziv coding.  Whenever possible, each file is replaced...


..       Compress uses the modified Lempel-Ziv algorithm popularized
          in "A Technique for High Performance Data Compression,"
          Terry A. Welch, IEEE Computer, vol. 17, no. 6 (June 1984),
          pp. 8-19.  Common substrings in the file are first replaced...

..  AUTHOR
          Compress was developed by Joseph M. Orost, Kenneth E.
          Turkowski, Spencer W. Thomas, and James A. Woods.

As you can see, there are proper references to the authors, but no mention of
ATT, Unisys or any patent license. It is unclear from this who *supplied* the
code (ATT, HP, other)

HP also supplies two other compression utilities [ pack(1) and compact(1) ]
both based on an algorithm called Huffman Encoding. I don't know any more about
it

As an aside, the very manual pages cited were delivered compressed using LZ
encoding (via compress).

- Mark