Path: utzoo!utgpu!news-server.csri.toronto.edu!rpi!zaphod.mps.ohio-state.edu!wuarchive!uunet!mcsun!hp4nl!star.cs.vu.nl!ast
From: ast@cs.vu.nl (Andy Tanenbaum)
Newsgroups: comp.os.minix
Subject: Re: Bitnet-sites (was Re: amoeba ??)
Message-ID: <9447@star.cs.vu.nl>
Date: 25 Mar 91 09:26:53 GMT
References: <48414@nigel.ee.udel.edu> <9073@plains.NoDak.edu>
Sender: news@cs.vu.nl
Organization: Fac. Wiskunde & Informatica, VU, Amsterdam
Lines: 22

In article <9073@plains.NoDak.edu> overby@plains.NoDak.edu (Glen Overby) writes:
>The one thing I'm looking forwards to a Minix sources group for is to be
>able to stuff EVERYTHING that comes across it into a nice armor uuencoded
>shield so you BITNET people can quit telling the rest of us to run our
>networks like it's EBCDIC!

A question that has not really come up, but perhaps will have to inevitably
is whether things sent to that group should be uuencoded or not.  Regardless
of whether uuencoding actually saves bandwidth or not, it does have the
property that a noncompressed (i.e. plaintext) file of about 80K will be
rejected by the news system is being too long (> 64K), but the same
compressed uuencoded file will fit in 64K.  Thus NOT uuencoding requires
large postings to be split in more pieces, which I think is a nuisance.
If I have a single file that is 200K, I would just as soon send it as 200K,
rather than having to split it and encase the pieces and number than etc
just because the network prefers everything in bite-sized pieces.

This should not be interpreted as an argument for uuencoding everything,
but it is another factor to be considered, and one that effects everyone,
not just bitnet sites.

Andy Tanenbaum (ast@cs.vu.nl)