Path: utzoo!utgpu!news-server.csri.toronto.edu!cs.utexas.edu!usc!wuarchive!uunet!mcsun!ukc!dcl-cs!aber-cs!athene!pcg From: pcg@cs.aber.ac.uk (Piercarlo Grandi) Newsgroups: comp.object Subject: Re: Stash Collection Message-ID: Date: 3 Dec 90 15:08:46 GMT References: <2849@esquire.dpw.com> <1990Nov27.074534.17744@Think.COM> <2853@esquire.dpw.com> <2855@esquire.dpw.com> Sender: pcg@aber-cs.UUCP Organization: Coleg Prifysgol Cymru Lines: 27 Nntp-Posting-Host: odin In-reply-to: yost@DPW.COM's message of 28 Nov 90 04:18:12 GMT On 28 Nov 90 04:18:12 GMT, yost@DPW.COM (David A. Yost) said: yost> Someone pointed out that purgeable memory blocks yost> on the Mac are a kind of stash collection. Same goes for OS/2 segments... yost> Come to think of it, I've advocated a similar principle for disk yost> file systems and netnews articles. Sort of a FIFO trash can: put yost> it in the trash queue, but it's still available for a while. I yost> use this method on my email messages and my office paper trash. yost> Very nice. There are a number of systems like this already there. Many UNIX admins have modified rm(1) not to delete but to move to a trash can directory, that is periodically purged of file solder than X days. There are some ready made rm(1) clones that do this in comp.source.{misc,unix}. Knuth in ACP volume 1 remarks that SLIP, the original Weizenbaum list processing package in Fortran, used a FIFO free list, which, he says, tended to make programs that generated dangling references run correctly for a awhile after the reference had been left dangling :-). -- Piercarlo Grandi | ARPA: pcg%uk.ac.aber.cs@nsfnet-relay.ac.uk Dept of CS, UCW Aberystwyth | UUCP: ...!mcsun!ukc!aber-cs!pcg Penglais, Aberystwyth SY23 3BZ, UK | INET: pcg@cs.aber.ac.uk Brought to you by Super Global Mega Corp .com