home *** CD-ROM | disk | FTP | other *** search
- Path: sparky!uunet!stanford.edu!rutgers!cmcl2!panix!mdonovan
- From: mdonovan@panix.com (Michael Donovan)
- Newsgroups: comp.compression
- Subject: Compressor Engine (was: Re: Compressing decimal expansion of pi)
- Summary: Writing a portable [de]compressor in a virtual machine
- Message-ID: <BzxKJ2.Cxq@panix.com>
- Date: 27 Dec 92 18:25:50 GMT
- References: <1992Dec22.221324.14115@pollux.lu.se> <1992Dec23.201005.18142@crd.ge.com> <1992Dec27.133611.3281@pollux.lu.se>
- Organization: Panix Public Access Internet & Unix, NYC
- Lines: 24
-
- >In article <1992Dec23.201005.18142@crd.ge.com> davidsen@crd.ge.com (bill davidsen) writes:
- >>Archivers starting with ARC used to
- >>try several compression algorithms and use the one best suited to the
- >>data. And they got better compression than any of the existing programs.
- >>[...]
- >> You could take the analogy to the point of called a compressed file a
- >>program which executes on a pseudo-machine called an expandor, who's
- >>output is the unique original file.
-
- Has this been done? I don't mean for custom, special-purpose things like
- generating Pi, but in general? If a compressor/expandor set were written
- using some sort of pseudo-code within a general-purpose archive program,
- then collections of files could be compressed with the latest compression,
- and distributed with the needed pseudo-code, and still be usable on any
- platform! The pseudo-code could be stored in the same archive, using a
- guaranteed-known compression method (like zip 1.9's), and the archive
- program could automatically add the code to it's library, possibly after
- compiling it to native code!
-
- Theoretically, this would be the last archive program you'd ever need!
-
- --
- Mike Donovan mdonovan@panix.com
- "These opinions are mine! Mine mine mine!"
-