Path: utzoo!attcan!uunet!wuarchive!mailrus!jarvis.csri.toronto.edu!torsqnt!hybrid!chance!john
From: john@chance.UUCP (John R. MacMillan)
Newsgroups: comp.sources.d
Subject: Re: Why "shar: Shell Archive  (v1.22)" is bad
Summary: Knapsacking not worth it
Keywords: shar
Message-ID: <1989Sep24.190434.9060@chance.UUCP>
Date: 24 Sep 89 19:04:34 GMT
References: <14502@bloom-beacon.MIT.EDU> <4155@cbnewsh.ATT.COM>
Reply-To: john@chance.UUCP (John R. MacMillan)
Organization: Haphazard
Lines: 15

In article <4155@cbnewsh.ATT.COM> wcs@cbnewsh.ATT.COM (Bill Stewart 201-949-0705 ho95c.att.com!wcs) writes:
|	Has anyone written someting like this?  The general case is
|	a knapsack / bin-packing problem that takes a list of items
|	and outputs a bunch of lists each containing less than N KB.
|	Implementation issues include output formats, and whether sizes
|	belong in the input or should be determined by the bin-packer.

I hacked around with this a while ago, and decided it's not really
worth the effort.  Usually I saw either no reduction in the number of
parts, or one fewer, so the savings in net bandwith isn't very big
(one set of news headers and one shar heading).
-- 
John R. MacMillan           "Don't you miss it...don't you miss it...
john@chance.UUCP             Some of you people just about missed it."
...!utcsri!hcr!chance!john        -- Talking Heads