Date: Sat, 11 Nov 1995 14:41:45 -0700 (MST) From: Terry Lambert <terry@lambert.org> To: hsu@cs.hut.fi (Heikki Suonsivu) Cc: peter@freefall.freebsd.org, freebsd-current@freefall.freebsd.org, brad@fcr.com, paulus@cs.anu.edu.au Subject: Re: cvs commit: src/usr.sbin/pppd RELNOTES Message-ID: <199511112141.OAA08131@phaeton.artisoft.com> In-Reply-To: <199511110306.FAA06778@shadows.cs.hut.fi> from "Heikki Suonsivu" at Nov 11, 95 05:06:02 am
next in thread | previous in thread | raw e-mail | index | archive | help
> And there is the nasty part: > > > Patents. > > ******** > > > > The BSD-Compress algorithm used for packet compression is the same as > > that used in the Unix "compress" command. It is apparently covered by > > U.S. patents 4,814,746 (owned by IBM) and 4,558,302 (owned by Unisys), > > and corresponding patents in various other countries (but not > > Australia). If this is of concern, you can build the package without > > including BSD-Compress. To do this, edit net/ppp-comp.h to change the > > definition of DO_BSD_COMPRESS to 0. The bsd-comp.c files are then no > > longer needed, so the references to bsd-comp.o may optionally be > > removed from the Makefiles. > > Has anyone looked at it, if it could be modified to do zip compression > instead. In addition to being patent-free it would improve performance? > There is a libary for gzip routines which was used for ssh quite > successfully, but I do not know how well it would sit into a kernel. > > The streams code does not apply FreeBSD, but as it is contaminated with a > pretty restrictive copyright (no-no for an ISP: "...nor to provide a > commercial service"). Is there something else than streams code which is > covered by this copyright? The decompression is identical for both the patent-covered (LZW) and non-covered (LZ) techniques. It's the compression process itself that uses Unisys's Terry Welch's techniques that is covered. Plain old Lempel-Ziv, which isn't covered, simply generates the data not as compressed as the Welch refinement causes it to be. The LZ decompressor can decompress both LZ and LZW compressed data, and the LZW decompressor is the same. So the restriction by the patent is on the compression process, and only on a particular method of compression. So it's possible to offer compatability without violating the patent. Terry Lambert terry@lambert.org --- Any opinions in this posting are my own and not those of my present or previous employers.
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?199511112141.OAA08131>