ietf-822
[Top] [All Lists]

A different approach...

1991-11-06 19:54:25
Well, I wish I had had a better handle on the idea 6 months ago.  However,
If the only argument against the idea is that it's 6 months late ... then
...

THe solution to the x-RMS-to-stream problem is something called FS encoding
which we designed and implemented about 4 years ago.  FS encoding encodes
all the attributes of a file system object (recursively too so you can
send directories of directories... too) into an SMTP-mailable body part.

Binary data components are compressed and checksummed rather rigorously.

Works like a charm.  Has just about wiped out most of the FTP traffic in
some parts of the company.  In fact the biggest problem is keeping people
from sending enormous (multi-MB) messages and clogging up the long-distance
part of out corporate net.

The reason we have never published the FS spec is that we were not happy
with the details of the binary compress/checksum - it generates lines 
longer than 78 characters and really could do a more sophisticated job
of compression.

Howveer, I might just see if I can find the spec and throw it up on the
list...  It was designed to handle Mac files, RMS, and PRIMOS SEG-Dirs!

David

<Prev in Thread] Current Thread [Next in Thread>
  • A different approach..., David Robinson <=