Path: utzoo!utgpu!news-server.csri.toronto.edu!cs.utexas.edu!wuarchive!usc!jarthur!nntp-server.caltech.edu!piglet!madler
From: madler@piglet.caltech.edu (Mark Adler)
Newsgroups: comp.sys.handhelds
Subject: Re: W A R N I N G  compress & ftp can corrupt hp48 bin files
Keywords: Ouch!
Message-ID: <1990Oct15.220403.15473@nntp-server.caltech.edu>
Date: 15 Oct 90 22:04:03 GMT
References: <9006@jarthur.Claremont.EDU> <40717@eerie.acsu.Buffalo.EDU> <1990Oct15.170509.9366@nntp-server.caltech.edu> <40831@eerie.acsu.Buffalo.EDU> <1990Oct15.172722.25374@jarvis.csri.toronto.edu>
Sender: news@nntp-server.caltech.edu
Organization: California Institute of Technology, Pasadena
Lines: 21
Nntp-Posting-Host: piglet.caltech.edu


I apologize if I offended James Cloos with my comments about assumptions
where the fault may lay.  In this case, I'M the one who made a bad
assumption about how he came to his conclusions.  I have assisted quite
a few people over email in doing these kind of transfers, and, so far,
it has been resolved each time by having the user do the right thing.

Also, I made that posting the same time I sent mail to James suggesting
things to try to get it to work, so his reply came long after that first
posting.  (Newsgroup posting tends to be rather slower than email.)

It certainly is tricky to get it to work, especially since some machines
don't like to have arbitrary length files.  Binary transfers require care
and attention to spot problems along the way to aid in debugging the
entire link.

Again, I apologize to James and to the net for this additional, but
appropriate bandwidth noise.

Mark Adler
madler@piglet.caltech.edu