From dcn@cbnewsd.ATT.COM Fri May 4 13:48:27 1990 From: dcn@cbnewsd.ATT.COM (david.c.newkirk) Newsgroups: alt.folklore.computers Subject: How Virtual Memory Really Works (or Doesn't Work...) Keywords: I/O Deja Vu from old Creative Computing Date: 4 May 90 14:06:04 GMT Organization: AT&T Bell Laboratories The recent posting about the Thing King reminded me of a similar parody in an old Creative Computing magazine - enjoy! I/O Deja Vu - A Farce in One Act -------------------------------- IBOX: Hello, CACHE? I'd like the data for location 54321. CACHE: I'm sorry sir. I don't have that data. Just a moment and I'll try to get it for you. CACHE: Hello, MAP? I need the address for location 54321. MAP: I don't have that address, sir. That address is stored at location 76543. Just a moment and I'll try to get it for you. Hello, I/O control? I need the data from location 76543. I/O: Just a moment. I'll try to get it. Hello, CACHE? I need the data from location 76543. CACHE: I don't seem to have that data, and I'm not about to get it. I/O: Hello, MAP? What is the address for location 76543? MAP: That address is 222543. I/O: Hello, CORE? What is the data at 222543? CORE: That data is INDIRECT PAGE TABLE 21, PAGE 44. I/O: Hello, MAP? That data you wanted is INDIRECT PAGE TABLE 21, PAGE 44. MAP: Fine. Now I need INDIRECT PAGE TABLE ENTRY 21. That would be stored at location 556. Would you get me the data at 556? I/O: I'll try. Hello, CACHE? I need the data from location 556. CACHE: That data is 6767. I/O: Hello, MAP? That data you wanted is 6767. MAP: Good. Now, let's see. 6767 plus 44 is 7033. Now, get me the data at 7033. I/O: Hold on. Hello, CACHE? I need the data from location 7033. CACHE: I'm sorry. I don't have that data. You'll have to get it from CORE. I/O: Hello, MAP? I need the address for location 7033. MAP: I'm sorry. I don't seem to have that address. I'll try to get it, but I'll probably forget what I was doing before, so you might as well too. MAP: Hello, I/O? I'm trying to get the address for 7033. That data should be stored at location 112233. Would you try to get me that data? I/O: (sigh) I'll try. Hello, CACHE? I need the data from location 112233. CACHE: That data is 4242. I/O: Hello, MAP? That data you wanted is 4242. MAP: I'll store that away. HEY IBOX, I just stored something. IBOX: Hello, CACHE? I'd like the data for location 54321. CACHE: I'm sorry sir. I don't have that data. Just a moment and I'll try... -- Dave Newkirk, att!ihlpm!dcn
I/O Deja Vu
August 5, 2008The Calendar Gap
August 5, 2008From dattier@gagme.chi.il.us Sun Feb 2 16:23:27 1992 From: dattier@gagme.chi.il.us (David W. Tamkin) Newsgroups: comp.unix.questions Subject: Re: cal: what's wrong with year 1752 Date: 31 Jan 92 22:28:51 GMT Organization: Gagme Public Access Hermitage, Chicago, Illinois 60656-1252 poulson@cs.widener.edu (Joshua R. Poulson) wrote in : | In article | xiaoy@bullet.ecf.utoronto.ca (XIAO Yan) writes: | >Can anybody enlighten me as to what's special about the September 1752, | >as far as `cal' command is concerned? | > S M Tu W Th F S | > 1 2 14 15 16 | >17 18 19 20 21 22 23 | >24 25 26 27 28 29 30 | | That's when some high record-keeper in the Catholic Church updated to | calendar to correspond for various astrological and religious events. | It corrected the calendar to correspond to holidays falling on the | correct days. Close; in *1582* the *pope* declared that October 4 should be followed by October 15 to skip ten days and get the calendar back in line with *astronomical* events. Further, century years were no longer to be leap years unless they were divisible by 400. The Julian calendar, by having too many leap years, was just too long and already ten days over. This adjustment got *solstices* and *equinoxes* on the correct dates and thus date-based holidays back to their proper relationships to the solstices and equinoxes: for example, Christmas was occurring too long after the solstice and in fact after perihelion. So why September, 1752? Protestant countries didn't go along at first, so the UK didn't convert until 1752. Whoever designed the cal program decided to use the calendar in effect in Britain and its possessions (which, at the time, included what would become the USA and Canada). In the meantime, 1700 had been a Julian leap year but not a Gregorian leap year, so the difference had grown to eleven days. (That's why September 3-13 are missing.) When the fledgling USSR converted in the early twentieth century, they were thirteen days off and found that their October Revolution had taken place on Gregorian November 7. | I could look this up, but I'm unsure of more than the fact that it was a | guy named Gregor and that's why it's the Gregorian Calendar. It was a guy named Ugo Buoncompagni, better known as Pope Gregory XIII, who gave his papal seal to the results. | Isn't history neat? In those days it was rather sloppy and unsanitary in comparison to modern times. David W. Tamkin Box 59297 Lincolnwood, Illinois 60659-0297 +1 708 518 6769 dattier@gagme.chi.il.us CIS: 73720,1570 MCI Mail: 426-1818 +1 312 463 2670
Origin of POSIX
August 5, 2008From jsq@longway.tic.com Thu May 3 19:16:29 1990 From: jsq@longway.tic.com (John S. Quarterman) Newsgroups: comp.std.unix Subject: Re: Answer to "what does POSIX stand for?" Date: 2 May 90 18:17:10 GMT Reply-To: std-unix@uunet.uu.net From: John S. Quarterman In article From: Don_Lewine@dgc.ceo.dg.com: >From the Rationale and Note section of IEEE Std 1003.1-1988 section B.1: > >"Since the interface enables application writers to write portable >applications -- it was developed with that goal in mind -- it has been >dubbed POSIX, an acronym for Portable Operations System Interface. The >name POSIX, suggested by Richard Stallman, was adopted during the printing >of the Trial Use Standard." At the time, the official name of the proposed standard was IEEE Standard Portable Operating System Environment, or POSE. Thus POSIX was a fairly obvious pun to produce something that sounded and looked similar to UNIX. The official name of the standard has changed several times since then (it was at one point Standard Portable Operating System for Computer Environments, or SPOSCE, and the name on the cover of the Full Use Standard is IEEE Standard Portable Operating System Interface for Computer Environments, or SPOSICE), but the name POSIX has stuck. Could have been worse: there exist bound draft copies of the Trial Use Standard that say IEEEIX on the cover.... >"... The term POSIX is expected to be pronounced pahz-icks, as in positive, >not poh-six, or other variations." Note that this assertion appears only in the rationale and the foreword, not in the body of the standard. This is because the committee could not standardize a pronunciation, and in fact had no consensus on what it might be. Nonetheless, there is a small clique that considers it their duty to enforce what they regard as the correct pronunciation, even though they don't all pronounce it the same way themselves. Volume-Number: Volume 19, Number 98
D-Mac's Stuff 