home *** CD-ROM | disk | FTP | other *** search
- Path: sparky!uunet!auspex-gw!guy
- From: guy@Auspex.COM (Guy Harris)
- Newsgroups: comp.sys.sun.apps
- Subject: Re: Help me tar kermit.tar.z !
- Message-ID: <14625@auspex-gw.auspex.com>
- Date: 14 Sep 92 21:20:15 GMT
- References: <1992Sep12.110700.1@draco.rutgers.edu> <JDELL.92Sep14105703@zenobia.mit.edu> <1992Sep14.160034.1762@uhura.neoucom.edu>
- Sender: news@auspex-gw.auspex.com
- Organization: Auspex Systems, Santa Clara
- Lines: 28
- Nntp-Posting-Host: bootme.auspex.com
-
- >The compression algorithm is similar,
-
- The only similarity is that both algorithms (algorithms, plural)
- generally make files smaller.
-
- >but there are technical implementation differes slightly.
-
- Completely, not slightly.
-
- "pack" Huffman-encodes the file, and "unpack" undoes it; that encoding
- is done on a byte-by-byte basis.
-
- "compress" Lempel-Ziv-Welch-etc.-encodes the file, and "uncompress"
- undoes it; that encoding is done on byte *sequences*, not on individual
- bytes.
-
- >I believe one is berkeley
- >flavor while the other has systemVish roots.
-
- "pack" was originally done by somebody at the Rand Corporation, as I
- remember; it eventually ended up in SV, with some changes.
-
- "compress" was originally done by Spencer Thomas at the University of
- Utah, based on an article in IEEE Computer, June 1984, and, at least
- according to the source, was also worked on by Jim McKie at CWI, Steve
- Davies at Perkin-Elmer, Ken Turkowski of somewhere, James A. Woods of
- NASA Ames, and Joe Orost at Perkin-Elmer. BSD happened to pick it up
- before SV did, but they both picked it up eventually.
-