Path: utzoo!attcan!uunet!lll-winken!lll-lcc!mordor!joyce!ames!oliveb!sun!breakpoint!jpayne From: jpayne%breakpoint@Sun.COM (Jonathan Payne) Newsgroups: comp.emacs Subject: Re: Buffer data structures Keywords: data structures, buffers Message-ID: <58551@sun.uucp> Date: 30 Jun 88 16:38:57 GMT References: <18612@cornell.UUCP> <394@sce.UUCP> Sender: news@sun.uucp Reply-To: jpayne@sun.UUCP (Jonathan Payne) Organization: Sun Microsystems, Mountain View Lines: 23 In article <394@sce.UUCP> graham@sce.UUCP (Doug Graham) writes: >On a related note, I would like to [ask] Jonathan Payne about the structure >he used for Jove. It uses a linked list of line structures kept >in memory. The actual text of the line is kept in "virtual memory" >with a virtual address in the line structure pointing to it. >I would like to know why he chose to keep the line >structures in memory rather than putting them in "virtual memory" >as well. This decision means that Jove cannot be used to edit >files containing a large number of lines on machines with a limited >amount of memory. Are there advantages to doing things this way? > >Doug. Ignorance. You kidding? I hadn't even HEARD of virtual memory when I started JOVE. A neat idea, I guess. I always hated the way JOVE's main limitation was the number of lines it could hold. Of course, I haven't noticed ever since I switched to a VAX, and now to Sun's. I love pdp11's, but Boy! who wants to think about memory in such silly ways as that anymore? Oh, and you'd be amazed how fast a busy computer can copy a megabyte when the gap moves. (Or maybe you wouldn't.) Mind-boggling. Computers are so fast, even when they're slow.