From mboxrd@z Thu Jan 1 00:00:00 1970 X-Spam-Checker-Version: SpamAssassin 3.4.4 (2020-01-24) on polar.synack.me X-Spam-Level: X-Spam-Status: No, score=0.1 required=5.0 tests=BAYES_05,INVALID_MSGID autolearn=no autolearn_force=no version=3.4.4 X-Google-Language: ENGLISH,ASCII-7-bit X-Google-Thread: 109fba,baaf5f793d03d420 X-Google-Attributes: gid109fba,public X-Google-Thread: fc89c,97188312486d4578 X-Google-Attributes: gidfc89c,public X-Google-Thread: 10db24,4cf070091283b555 X-Google-Attributes: gid10db24,public X-Google-Thread: 103376,97188312486d4578 X-Google-Attributes: gid103376,public X-Google-Thread: 1014db,6154de2e240de72a X-Google-Attributes: gid1014db,public From: dewar@cs.nyu.edu (Robert Dewar) Subject: Re: What's the best language to learn? [was Re: Should I learn C or Pascal?] Date: 1996/09/07 Message-ID: #1/1 X-Deja-AN: 179085362 references: <4vqt9t$3jk@ns.broadvision.com> organization: Courant Institute of Mathematical Sciences newsgroups: comp.edu,comp.lang.ada,comp.lang.c,comp.lang.c++,comp.unix.programmer Date: 1996-09-07T00:00:00+00:00 List-Id: George II was one of the ICL operating systems, I used George 4 extensively on an ICL 1906A at the University of Leeds in the early 70's, and was amused by the front page of the manual, which said that "George 4 uses the well established techniques of virtual memory and demand paging to manage memory", or somesuch (after all Atlas did precede it by many years). That's an interesting contrast to an IBM spokesperson describing the first virtual memory implementations of OS/3x0 much later, who, when asked about the Atlas, said he had never heard of it, and that IBM was too busy developing advanced technology to waste time looking at ancient history (I don't know if that is a verbatim quote, I was not there, but that is the way a friend described the Florida Share meeting). George 4 had two other nice features, which still seem advanced today. First, it had a completely fluid file system that spilled from disk to tape in a transparent manner. I once overheard one of the system programmers saying to another "we are getting a little pressed for disk space, we have 300 megabytes of disks, and nearly 900 megs of online files" :-) Leeds kept four tape drives dedicated to constantly writing updated information (a kind of constant backup) and reading old tapes. When I asked someone there how to delete an old version of a file (George had version numbers like VMS), he replied "why would you ever delete a file, you never know if you might need it", and indeed, if an unused file wanders to a dusty tape, that makes sense. When I returned each summer to the university of Leeds, I would log in, and it would take 5 minutes to list my top level directory, because it was on some old tape that the operator had to find and mount. The other thing that was interesting was that george 4 really used a virtual machine setup. You got presented a virtual machine whose hardware I/O devices you could configure, and which you could attach to the real devices or to virtual devices on disk, which in the simplest cases where simply files. The ICL range at the time covered a huge range from tiny minicomputers to big virtual memory mainframes. You really got upwards and downwards compatibility across the range, because at the low end, the machine became a real machine with virtually no operating system, and you ran your program directly abvove the hardware, and that same program run under George 4 using a virtual machine configured to be identical to the hardware of the low end machine assumed by the program, so every program was in fact terribly device specific and hardware configuration specific, but it didn't matter, since you simply configured your virtual machine to match the requirements of the program. Sorry for the off-topic reminiscence, but the keyword "George" brings back some fond memories. By the way it was during these summers at Leeds that (a) I learned Algol-68 and got involved in that scene, which indirectly lead to my involvement in Ada and (b) wrote Macro-SPITBOL.