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