From: jsavard@excxn.aNOSPAMb.cdn.invalid (John Savard) Newsgroups: alt.folklore.computers Subject: The ORIGIN of the Dominance of the 8-bit Byte Date: Sat, 21 Aug 2004 14:21:08 GMT Organization: Group Telecom - a 360networks Company Lines: 165 Message-ID: <41275572.1465186@news.ecn.ab.ca> References: <41190a85.715407@news.ecn.ab.ca> <411B57E5.DC9D2518@alum.mit.edu> <86u0v83z2a.fsf@my.domain> <41262052.61C819EC@earthlink.net> <41262d6a.9404558@news.ecn.ab.ca> NNTP-Posting-Host: h64-42-245-131.gtcust.grouptelecom.net X-Trace: utornnr1pp.grouptelecom.net 1093098072 7631 64.42.245.131 (21 Aug 2004 14:21:12 GMT) X-Complaints-To: abuse@gt.ca NNTP-Posting-Date: Sat, 21 Aug 2004 14:21:12 +0000 (UTC) X-Newsreader: Forte Free Agent 1.11/32.235 Path: nightfall.franklin.ch!pfaff2.ethz.ch!news-zh.switch.ch!switch.ch!solnet.ch!solnet.ch!news.glorb.com!logbridge.uoregon.edu!snoopy.risq.qc.ca!news.uunet.ca!newsfeed.grouptelecom.net!news.east.grouptelecom.net!not-for-mail Xref: nightfall.franklin.ch alt.folklore.computers:182400 On Fri, 20 Aug 2004 17:25:55 GMT, jsavard@excxn.aNOSPAMb.cdn.invalid (John Savard) wrote, in part: >On Fri, 20 Aug 2004 16:02:55 GMT, jchausler wrote, in >part: >>"Dowe G. Keller" wrote: >>> "Kevin G. Rhoads" writes: >>> > >I remember it being the case back in my 68K days, but nowadays I tend to >>> > >think of a word as an entity made of at least 32 bits. >>> > And before the 68K words wre often 12 bits or 8 bits, not 16. >>> And 18 and 36 and 60 and ... >>> IMHO, the PDP-11 and NOVA did much to popularize 16bit machines. >>I've always wondered how much Project Whirlwind >>influenced the selection of 16 bits as it was a 16 bit >>computer back in the days when 40 to 60 bits was >>popular. >There were computers with many unusual word lengths in the old days; 37 bits, 19 >bits, 21 bits, and so on. >Since in the "old days", computers were strictly business, they usually stored >characters that were 6 bits long, so it made sense to use word lengths that were >a multiple of that, such as 36 bits or 48 bits. >Many computers, of course, were built to do decimal arithmetic. Such computers >could have been completely decimal, and thus they might have used eight bits of >storage for a character code that could represent only 100 different characters. >The Honeywell 800 was an example of a computer with a 48-bit word length that >used 6-bit characters, and provided both binary arithmetic and packed decimal >arithmetic. It predated the IBM 360. >The IBM 360, of course, was the "900 pound gorilla" that, in replacing the >preceding IBM 7090 computer with a 36-bit word, decided things irrevocably in >favor of the 8-bit byte with everything else built from powers of two. >It basically settled the question of word lengths, and such machines as the >PDP-11 only followed in its wake. >Thus, the question of how much influence Project Whirlwind had becomes the >question of how much influence it had on the System/360. >Certainly, Whirlwind influenced SAGE, being a prototype for it, and the AN/FSQ-7 >vacuum tube computer not only had a 32-bit word length, but was built by IBM. But >it was an unusual computer, which only performed arithmetic on pairs of 16-bit >numbers. >The experimental successor to the AN/FSQ-7 computer, though, the AN/FSQ-32, had a >48-bit word length. It could operate on 48-bit integers, pairs of 24-bit >integers, or 48-bit floating-point numbers, and, thus, it was a more >general-purpose computer. This was also built by IBM; the prototype ended up at >the System Development Corporation in Santa Clara, California. >And at one point, it was connected to the Lincoln Laboratories TX-2 computer at >MIT, a computer with a 36-bit word, which also had the ability of reconfiguring >the arithmetic unit for calculations on two 18-bit numbers in parallel... this, >of course, being interesting as being part of the history of Intel's MMX feature. >But that particular moment, of course, is even more important to history, because >it was when the *Internet* was born! >Back to the current discussion - the lineal descendants of Whirlwind, then, >reverted to 48 bits from 32 as time progressed. >The rationale behind the design of the IBM 360 computer is well known. It was >intended, like the Honeywell 800, to be amenable to both scientific/mathematical >use and to commercial/accounting use. >Rather than operating on decimal numbers in character form, like a 1401, which >contributed the idea of variable-length instructions, it operated on packed >decimal numbers for greater efficiency in numeric storage. >The notion that a 32-bit word, rather than a 48-bit word, would allow *individual >bits* to be addressed was considered in the design of the System/360. That >architecture, though, never did offer any kind of bit addressing; but the fact >that the possibility of such an extension would be left open was mentioned in >initial IBM articles about the computer. >This, I think, squarely points the finger of suspicion away from Whirlwind... and >towards STRETCH. STRETCH, the IBM 7030, was bit addressable, and had a 32-bit >word. >The fact that the IBM 360 was going to be implemented across a wide range of >machines, with different microprogram formats and data bus widths, also would >have had an influence. Some of the low-end System/360 machines had 16-bit or even >8-bit buses to memory. >A 6-bit character and a 4-bit digit, though, wouldn't have totally prevented >that. You could still have had a bus width as narrow as 12 bits and still retain >compatibility with both data formats. >Was the deciding factor, then, just a matter of which *price points* the >System/360 was intended to address? >The 1403 printer did offer a TN print train that let one use lower case, and the >2741 terminal, based on the Selectric typewriter, was one of the initial >peripherals offered with the System/360. Also, IBM used a 4 of 8 code for >transmission of 6-bit characters. (The 2741 used a 6-bit code with shift >characters, not a code with unique positions for each character; code positions >actually corresponded to type element locations, and, thus, there was the >PTTC/EBCD code which was similar to BCDIC, and required a special type element, >and there was the Correspondence code, where the letters were scrambled to match >the pre-existing layout of letters on a regular Selectric typewriter.) >IBM did sell its computers on a world-wide basis, and character positions for >Japanese katakana symbols were defined within EBCDIC. >Gene Amdahl and R. W. Bemer, at IBM, were involved with the 8-bit decision; R. W. >Bemer has written historical works on character code development which drew on >his activities. The STRETCH, incidentally, used an 8-bit character code and did >include lower-case letters in that code. >So I think I would definitely say that STRETCH contributed more than Whirlwind to >the current dominance of the 8-bit byte; and the System/360 design was also >influenced by the need for decimal arithmetic, which neither STRETCH nor >Whirlwind had. I may note that R. W. Bemer, who recently passed away, also worked on STRETCH. Also, since Ken Olsen worked on Whirlwind, it may indeed be that this somewhat influenced his decision, later on, to make the PDP-11 a 16-bit machine. And, furthermore, since the PDP-11 was the flagship machine of the Digital Equipment Corporation, the company that was felt to make computers for independent thinkers, for rebels, for those who wanted to escape the grayness of IBM... sort of the Apple Computer of its day... the fact that even *they* embraced the 8-bit byte could well have helped to bring about its *total* dominance, even though the STRETCH, through the 360, ensured the 8-bit byte would always have a strong presence. The PDP-11, of course, is the machine I curse every day, for having demonstrated that a computer could be consistently little-endian. It influenced and inspired the designers of the 8080 and the 6502 in this respect, and is the reason that the current 80386 architecture, manifest on the Intel Pentium brand microprocessors we use (call them not Pentiae, at your peril) is little-endian. If it influenced today's microprocessors in that respect, it could well have influenced them in word length. To what extent was the PDP-11 a 16-bit machine because of Whirlwind, and to what extent was it a 16-bit machine because of the existing popularity of power-of-two word sizes thanks to the 360? By the time the PDP-11 came out, lower-case letters already had a defined place in ASCII. Thus, the 6-bit character would clearly have been limiting, tying the computer to the past. A 7-bit character is something awkward to cut into smaller parts, and few other computer makers used it. DEC did already have a perfectly good line of machines with an 18-bit word; and the (36-bit) PDP-10 showed that this can lead to 9-bit characters as easily as 6-bit characters. Advances in electronic circuit technology, however, led DEC at that time to want a "clean sheet of paper" to build a design based on different principles; the PDP-11 has six general registers, instead of doing all its arithmetic between the memory and one accumulator. The IBM 360 demonstrated the advantages of general registers to a great many people, and this concept's lineage goes back to the Pegasus, a British drum and magnetostrictive delay line vacuum-tube machine. John Savard http://home.ecn.ab.ca/~jsavard/index.html ###### Message-ID: <4127767C.13556CE4@yahoo.com> From: CBFalconer Reply-To: cbfalconer@worldnet.att.net Organization: Ched Research X-Mailer: Mozilla 4.75 [en] (Win98; U) X-Accept-Language: en MIME-Version: 1.0 Newsgroups: alt.folklore.computers Subject: Re: The ORIGIN of the Dominance of the 8-bit Byte References: <41190a85.715407@news.ecn.ab.ca> <411B57E5.DC9D2518@alum.mit.edu> <86u0v83z2a.fsf@my.domain> <41262052.61C819EC@earthlink.net> <41262d6a.9404558@news.ecn.ab.ca> <41275572.1465186@news.ecn.ab.ca> Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit Lines: 22 Date: Sat, 21 Aug 2004 16:25:30 GMT NNTP-Posting-Host: 12.76.138.140 X-Complaints-To: abuse@worldnet.att.net X-Trace: bgtnsc04-news.ops.worldnet.att.net 1093105530 12.76.138.140 (Sat, 21 Aug 2004 16:25:30 GMT) NNTP-Posting-Date: Sat, 21 Aug 2004 16:25:30 GMT Path: nightfall.franklin.ch!pfaff2.ethz.ch!news-zh.switch.ch!switch.ch!solnet.ch!solnet.ch!news.glorb.com!wn13feed!worldnet.att.net!bgtnsc04-news.ops.worldnet.att.net.POSTED!53ab2750!not-for-mail Xref: nightfall.franklin.ch alt.folklore.computers:182410 John Savard wrote: > ... snip ... > > To what extent was the PDP-11 a 16-bit machine because of > Whirlwind, and to what extent was it a 16-bit machine because > of the existing popularity of power-of-two word sizes thanks > to the 360? I suspect neither. It is simply a matter of getting the most bang for the buck. Registers and bit fields that control various things are completely used, without wastage. Please configure your newsreader to limit output line length to something shorter - 65 chars is nice. -- fix (vb.): 1. to paper over, obscure, hide from public view; 2. to work around, in a way that produces unintended consequences that are worse than the original problem. Usage: "Windows ME fixes many of the shortcomings of Windows 98 SE". - Hutchison ###### From: jsavard@excxn.aNOSPAMb.cdn.invalid (John Savard) Newsgroups: alt.folklore.computers Subject: Re: The ORIGIN of the Dominance of the 8-bit Byte Date: Sat, 21 Aug 2004 18:46:22 GMT Organization: Group Telecom - a 360networks Company Lines: 11 Message-ID: <41279862.6436797@news.ecn.ab.ca> References: <41190a85.715407@news.ecn.ab.ca> <411B57E5.DC9D2518@alum.mit.edu> <86u0v83z2a.fsf@my.domain> <41262052.61C819EC@earthlink.net> <41262d6a.9404558@news.ecn.ab.ca> <41275572.1465186@news.ecn.ab.ca> <4127767C.13556CE4@yahoo.com> NNTP-Posting-Host: h64-42-245-134.gtcust.grouptelecom.net X-Trace: utornnr1pp.grouptelecom.net 1093113985 7012 64.42.245.134 (21 Aug 2004 18:46:25 GMT) X-Complaints-To: abuse@gt.ca NNTP-Posting-Date: Sat, 21 Aug 2004 18:46:25 +0000 (UTC) X-Newsreader: Forte Free Agent 1.11/32.235 Path: nightfall.franklin.ch!pfaff2.ethz.ch!news-zh.switch.ch!switch.ch!solnet.ch!solnet.ch!news.glorb.com!news.alt.net!newsfeed.grouptelecom.net!news.east.grouptelecom.net!not-for-mail Xref: nightfall.franklin.ch alt.folklore.computers:182416 On Sat, 21 Aug 2004 16:25:30 GMT, CBFalconer wrote, in part: >Please configure your newsreader to limit output line length to >something shorter - 65 chars is nice. Apologies; I had to lengthen it to include a table in one post, and forgot to change it back. John Savard http://home.ecn.ab.ca/~jsavard/index.html ###### Newsgroups: alt.folklore.computers Subject: Re: The ORIGIN of the Dominance of the 8-bit Byte References: <41190a85.715407@news.ecn.ab.ca> <41262052.61C819EC@earthlink.net> <41262d6a.9404558@news.ecn.ab.ca> <41275572.1465186@news.ecn.ab.ca> Reply-To: jhaynes@alumni.uark.edu Organization: University of Arkansas Alumni X-Newsreader: trn 4.0-test72 (19 April 1999) From: haynes@alumni.uark.edu (Jim Haynes) Lines: 52 Message-ID: <8fNVc.6970$2L3.1221@newsread3.news.atl.earthlink.net> Date: Sat, 21 Aug 2004 19:19:32 GMT NNTP-Posting-Host: 4.226.60.30 X-Complaints-To: abuse@earthlink.net X-Trace: newsread3.news.atl.earthlink.net 1093115972 4.226.60.30 (Sat, 21 Aug 2004 12:19:32 PDT) NNTP-Posting-Date: Sat, 21 Aug 2004 12:19:32 PDT Path: nightfall.franklin.ch!pfaff2.ethz.ch!news-zh.switch.ch!switch.ch!solnet.ch!solnet.ch!news.glorb.com!border1.nntp.dca.giganews.com!border2.nntp.dca.giganews.com!nntp.giganews.com!elnk-atl-nf1!newsfeed.earthlink.net!stamper.news.atl.earthlink.net!newsread3.news.atl.earthlink.net.POSTED!c4a26796!not-for-mail Xref: nightfall.franklin.ch alt.folklore.computers:182421 I've always believed the popularity of the 36 bit word was due simply to the fact that 36 has more factors than anything nearby. In the early days when storage was in extremely short supply you had to pack things into a word. 36 bits lets you pack items of 1, 2, 3, 4, 6, 9, 12, or 18 bits into a word with no waste. Six bits was especially important at the time because of the widespread use of six-bit BCD character codes. If you wanted to make a minicomputer at the time it must have been natural to choose 18 bits as half of 36 and thus almost as nice with respect to factors and especially six-bit characters. As you mentioned, STRETCH used an 8-bit character code because everybody realized six bits was no longer enough - and 8 bits fits nicely with the power-of-2 situation in that STRETCH was addressable to the single bit level. So in STRETCH you had the feature that by lopping off low-order address bits you could address items of 1, 2, 4, 8, etc. bits in a consistent manner. Apparently in the 360 line the ability to address down to the bit was not considered to be worth the extra 3 bits of address that it required. Both STRETCH and 360 made a break with the past in which fixed-word-length machines addressed to the word and then used special tricks to get to fields within a word. IBM seems to have hedged its bets on character code design - they participated in the ASCII effort and then developed their own EBCDIC code when the ASCII community would not yield to IBMs desires. What IBM desired, and got in EBCDIC was a code that perpetuated the breaks in the alphabetical sequence that exist in punched-card code. The STRETCH designers had eliminated them and apparently found that was too bitter a pill for the corporation to swallow. IBM even brought back six-bit codes with their 96-column cards. And then there was the largely non-function "ASCII bit" in System/360 that was apparently necessary to get the government to buy a non-ASCII computer system. Presumably Control Data went to 60 bits because the customer wanted more precision than their previous 48 bit computer afforded, and they wanted the word size to be a multiple of 12. CDC liked to read and write cards in column binary, hence 12 bits per column. The makers of 36 bit word machines took a variety of measures to cope with 8-bit character codes. DEC chose to allow arbitrary-sized bytes packed into 36 bit words, with leftover bits at the end of the word left unused. G.E. chose to have means to deal with 6-bit or 9-bit bytes in words. This introduced a nagging problem with what to do with the 9th bit in handling 8-bit bytes. If you want to read and write tapes you have to have a mode that preserves the whole 36 bit words and another mode that produces strings of 8-bit characters. I don't know what Univac did about the problem - someone else can tell us. -- jhhaynes at earthlink dot net ###### From: jsavard@excxn.aNOSPAMb.cdn.invalid (John Savard) Newsgroups: alt.folklore.computers Subject: Re: The ORIGIN of the Dominance of the 8-bit Byte Date: Sat, 21 Aug 2004 22:04:48 GMT Organization: Group Telecom - a 360networks Company Lines: 23 Message-ID: <4127c650.18196556@news.ecn.ab.ca> References: <41190a85.715407@news.ecn.ab.ca> <41262052.61C819EC@earthlink.net> <41262d6a.9404558@news.ecn.ab.ca> <41275572.1465186@news.ecn.ab.ca> <8fNVc.6970$2L3.1221@newsread3.news.atl.earthlink.net> NNTP-Posting-Host: h64-42-245-90.gtcust.grouptelecom.net X-Trace: utornnr1pp.grouptelecom.net 1093125892 272 64.42.245.90 (21 Aug 2004 22:04:52 GMT) X-Complaints-To: abuse@gt.ca NNTP-Posting-Date: Sat, 21 Aug 2004 22:04:52 +0000 (UTC) X-Newsreader: Forte Free Agent 1.11/32.235 Path: nightfall.franklin.ch!pfaff2.ethz.ch!news-zh.switch.ch!switch.ch!solnet.ch!solnet.ch!news.glorb.com!cyclone.bc.net!news.uunet.ca!newsfeed.grouptelecom.net!news.east.grouptelecom.net!not-for-mail Xref: nightfall.franklin.ch alt.folklore.computers:182427 On Sat, 21 Aug 2004 19:19:32 GMT, haynes@alumni.uark.edu (Jim Haynes) wrote, in part: >I don't know what Univac did >about the problem - someone else can tell us. I'm no expert on this, but here in Edmonton, the provincial government uses mostly IBM computers - they got in the news as buying the first specimen of a model of z/Architecture machine recently - and the municipal government uses mostly Univac computers. I remember speaking to someone who basically said that lower-case letters were basically not available with a Univac machine. While Unisys now sells machines with 8-bit characters, it may be that they, therefore, addressed the problem by resolutely ignoring it, and refusing to support 8-bit characters in any significant way on their 36-bit machines. John Savard http://home.ecn.ab.ca/~jsavard/index.html ###### Newsgroups: alt.folklore.computers From: hoh@invalid.invalid (Goran Larsson) Subject: Re: The ORIGIN of the Dominance of the 8-bit Byte Message-ID: Organization: [1] + 5934 done /bin/rm -rf ~/ & X-Newsreader: trn 4.0-test73 (May 24, 2000) References: <41190a85.715407@news.ecn.ab.ca> <41262d6a.9404558@news.ecn.ab.ca> <41275572.1465186@news.ecn.ab.ca> <8fNVc.6970$2L3.1221@newsread3.news.atl.earthlink.net> Lines: 13 NNTP-Posting-Host: 213.101.45.254 X-Complaints-To: news-abuse@swip.net X-Trace: nntpserver.swip.net 1093127342 213.101.45.254 (Sun, 22 Aug 2004 00:29:02 MET DST) NNTP-Posting-Date: Sun, 22 Aug 2004 00:29:02 MET DST X-Sender: s-1124858@d213-101-45-254.swipnet.se Date: Sat, 21 Aug 2004 22:26:26 GMT Path: nightfall.franklin.ch!pfaff2.ethz.ch!news-zh.switch.ch!switch.ch!solnet.ch!solnet.ch!newsfeed.tiscali.ch!feed.news.tiscali.de!uio.no!news.tele.dk!news.tele.dk!small.news.tele.dk!newsfeed1.swip.net!swipnet!nntpserver.swip.net!not-for-mail Xref: nightfall.franklin.ch alt.folklore.computers:182430 In article <8fNVc.6970$2L3.1221@newsread3.news.atl.earthlink.net>, Jim Haynes wrote: > What IBM desired, and > got in EBCDIC was a code that perpetuated the breaks in the alphabetical > sequence that exist in punched-card code. Why did IBM desire breaks in the alphabetical sequence? Just compatibility with old habits or did the breaks give IBM some important advantage that the rest of the world didn't understand? -- Göran Larsson http://www.mitt-eget.com/ ###### From: David R Brooks Newsgroups: alt.folklore.computers Subject: Re: The ORIGIN of the Dominance of the 8-bit Byte Date: Sun, 22 Aug 2004 07:12:29 +0800 Message-ID: <69lfi0l02h0i9i3nra30qhfa391au188t0@4ax.com> References: <41190a85.715407@news.ecn.ab.ca> <411B57E5.DC9D2518@alum.mit.edu> <86u0v83z2a.fsf@my.domain> <41262052.61C819EC@earthlink.net> <41262d6a.9404558@news.ecn.ab.ca> <41275572.1465186@news.ecn.ab.ca> X-Newsreader: Forte Agent 1.92/32.572 X-No-Archive: yes MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit Lines: 31 NNTP-Posting-Host: 203.59.155.35 X-Trace: 1093129969 per-qv1-newsreader-01.iinet.net.au 22832 203.59.155.35 Path: nightfall.franklin.ch!pfaff2.ethz.ch!news-zh.switch.ch!irazu.switch.ch!switch.ch!HSNX.atgi.net!newsfeed.iinet.net.au!newsfeed.iinet.net.au!per-qv1-newsreader-01.iinet.net.au!not-for-mail Xref: nightfall.franklin.ch alt.folklore.computers:182433 jsavard@excxn.aNOSPAMb.cdn.invalid (John Savard) wrote: :On Fri, 20 Aug 2004 17:25:55 GMT, jsavard@excxn.aNOSPAMb.cdn.invalid (John :Savard) wrote, in part: :>The IBM 360, of course, was the "900 pound gorilla" that, in replacing the :>preceding IBM 7090 computer with a 36-bit word, decided things irrevocably in :>favor of the 8-bit byte with everything else built from powers of two. : :>It basically settled the question of word lengths, and such machines as the :>PDP-11 only followed in its wake. : Another driver (towards 4N bits) was the development at the end of the '60s, of TTL/MSI logic packages. They were mostly 4 bits to a package (adder, register etc.) : :The PDP-11, of course, is the machine I curse every day, for having demonstrated :that a computer could be consistently little-endian. It influenced and inspired :the designers of the 8080 and the 6502 in this respect, and is the reason that :the current 80386 architecture, manifest on the Intel Pentium brand :microprocessors we use (call them not Pentiae, at your peril) is little-endian. Another reason for little-endian, if you are trying to save hardware (as we were back then) is that you basically count UP to address consecutive bytes. Now you already have an UP counter in your machine: it's called the instruction counter. With a little cunning, that can access data bytes as well: everything reads in the same order. Of course, the softweare people (having only hexadecimal dumps to work from) now curse you, because they have to mentally reverse everything they read, but that's life :-) ###### From: jsavard@excxn.aNOSPAMb.cdn.invalid (John Savard) Newsgroups: alt.folklore.computers Subject: Re: The ORIGIN of the Dominance of the 8-bit Byte Date: Sun, 22 Aug 2004 00:56:56 GMT Organization: Group Telecom - a 360networks Company Lines: 52 Message-ID: <4127ed5f.28197423@news.ecn.ab.ca> References: <41190a85.715407@news.ecn.ab.ca> <41262d6a.9404558@news.ecn.ab.ca> <41275572.1465186@news.ecn.ab.ca> <8fNVc.6970$2L3.1221@newsread3.news.atl.earthlink.net> NNTP-Posting-Host: h64-42-245-158.gtcust.grouptelecom.net X-Trace: utornnr1pp.grouptelecom.net 1093136220 22353 64.42.245.158 (22 Aug 2004 00:57:00 GMT) X-Complaints-To: abuse@gt.ca NNTP-Posting-Date: Sun, 22 Aug 2004 00:57:00 +0000 (UTC) X-Newsreader: Forte Free Agent 1.11/32.235 Path: nightfall.franklin.ch!pfaff2.ethz.ch!news-zh.switch.ch!switch.ch!news.imp.ch!news.imp.ch!snoopy.risq.qc.ca!news.uunet.ca!newsfeed.grouptelecom.net!news.east.grouptelecom.net!not-for-mail Xref: nightfall.franklin.ch alt.folklore.computers:182439 On Sat, 21 Aug 2004 22:26:26 GMT, hoh@invalid.invalid (Goran Larsson) wrote, in part: >Why did IBM desire breaks in the alphabetical sequence? Just compatibility >with old habits or did the breaks give IBM some important advantage that >the rest of the world didn't understand? The breaks in themselves were not desired, as such. Before there were such things as computers, IBM made equipment to total information recorded on punched cards, and equipment to sort punched cards. So, punched cards were a *pre-existing* technology used as a computer input medium. A punched card had twelve rows of holes. They typically stood for +, -, 0, 1, 2, 3, 4, 5, 6, 7, 8, and 9. Letters were indicated by punching two holes at once; +1 through +9 stood for the letters A through J; -1 through -9 stood for K through R, and 02 through 09 stood for S through Z. What IBM wanted was a code that could be generated from the existing punched-card code with as few gates as possible. Other companies with completely different codes did the same thing; for example, while IBM always used natural binary-coded decimal, other companies that used the excess-3 representation for decimal digits also used codes that corresponded to punched card code. Other computers used Flexowriter codes that were like a 6-bit version of 5-level TTY code, in which the alphabetic sequence was not merely broken, it was scrambled. (The letter E was 10000 in 5-level code; the letter T was 00001, and the space was 00100: the idea was to minimize wear and tear.) Then there was Fieldata, which was specified by the military, and which made more sense. A 6-bit version of this code was used on many later Univac computers. ASCII didn't come along till later - although it was around by the time the 360 came along, the BCDIC 6-bit code had long been in existence, and naturally, IBM wanted to preserve as much compatibility as possible between old and new. John Savard http://home.ecn.ab.ca/~jsavard/index.html ###### From: jsavard@excxn.aNOSPAMb.cdn.invalid (John Savard) Newsgroups: alt.folklore.computers Subject: Re: The ORIGIN of the Dominance of the 8-bit Byte Date: Sun, 22 Aug 2004 01:01:19 GMT Organization: Group Telecom - a 360networks Company Lines: 35 Message-ID: <4127ef7f.28741176@news.ecn.ab.ca> References: <41190a85.715407@news.ecn.ab.ca> <411B57E5.DC9D2518@alum.mit.edu> <86u0v83z2a.fsf@my.domain> <41262052.61C819EC@earthlink.net> <41262d6a.9404558@news.ecn.ab.ca> <41275572.1465186@news.ecn.ab.ca> <69lfi0l02h0i9i3nra30qhfa391au188t0@4ax.com> NNTP-Posting-Host: h64-42-245-158.gtcust.grouptelecom.net X-Trace: utornnr1pp.grouptelecom.net 1093136481 22353 64.42.245.158 (22 Aug 2004 01:01:21 GMT) X-Complaints-To: abuse@gt.ca NNTP-Posting-Date: Sun, 22 Aug 2004 01:01:21 +0000 (UTC) X-Newsreader: Forte Free Agent 1.11/32.235 Path: nightfall.franklin.ch!pfaff2.ethz.ch!news-zh.switch.ch!switch.ch!newscore.univie.ac.at!news-spur1.maxwell.syr.edu!news.maxwell.syr.edu!cyclone.bc.net!news.uunet.ca!newsfeed.grouptelecom.net!news.east.grouptelecom.net!not-for-mail Xref: nightfall.franklin.ch alt.folklore.computers:182440 On Sun, 22 Aug 2004 07:12:29 +0800, David R Brooks wrote, in part: > Another reason for little-endian, if you are trying to save hardware >(as we were back then) is that you basically count UP to address >consecutive bytes. Now you already have an UP counter in your machine: >it's called the instruction counter. With a little cunning, that can >access data bytes as well: everything reads in the same order. > Of course, the softweare people (having only hexadecimal dumps to >work from) now curse you, because they have to mentally reverse >everything they read, but that's life :-) Well, nowadays, gates aren't so rare. You ARE right that this was the basic reason behind little-endian. But before the PDP-11 came along, if a computer had a 16-bit word, the two bytes in that word would be placed there in big-endian order. If, to save counting backwards, 32-bit integers happened to be least significant word first, well, that's an exotic type outside of the normal hardware types. It was with the PDP-11 that someone had the brilliant idea of accessing the bytes *within* a word (fetched in a single bus access) in reverse order, so that now everything could be consistent. Before the PDP-11, being consistently big-endian, like a 360, was always the desired goal, but there were occasional exceptions to that tolerated in order to use the up counter. John Savard http://home.ecn.ab.ca/~jsavard/index.html ###### NNTP-Posting-Date: Sat, 21 Aug 2004 23:07:28 -0500 From: "Bill Leary" Newsgroups: alt.folklore.computers References: <41190a85.715407@news.ecn.ab.ca> <41262d6a.9404558@news.ecn.ab.ca> <41275572.1465186@news.ecn.ab.ca> <8fNVc.6970$2L3.1221@newsread3.news.atl.earthlink.net> <4127ed5f.28197423@news.ecn.ab.ca> Subject: Re: The ORIGIN of the Dominance of the 8-bit Byte Date: Sun, 22 Aug 2004 00:16:40 -0400 X-Priority: 3 X-MSMail-Priority: Normal X-Newsreader: Microsoft Outlook Express 6.00.2800.1437 X-MimeOLE: Produced By Microsoft MimeOLE V6.00.2800.1441 Message-ID: Lines: 29 X-Trace: sv3-m8FKKaUzDQJ8SRa0ZRiKkS3qPbzzjzIPyQzueO5Go0nW7mmEFDTmdrS3ooTWWPOYn+fYUDq0+lxCqu9!G5b/3DwwRbSEGbgOAn+FaVJrutFG2HCKIg4v9Eum+I0W5sqGskINZAxZ1vA= X-Complaints-To: abuse@giganews.com X-DMCA-Notifications: http://www.giganews.com/info/dmca.html X-Abuse-and-DMCA-Info: Please be sure to forward a copy of ALL headers X-Abuse-and-DMCA-Info: Otherwise we will be unable to process your complaint properly X-Postfilter: 1.3.13 Path: nightfall.franklin.ch!pfaff2.ethz.ch!news-zh.switch.ch!switch.ch!solnet.ch!solnet.ch!news.glorb.com!border1.nntp.dca.giganews.com!nntp.giganews.com!local1.nntp.dca.giganews.com!news.giganews.com.POSTED!not-for-mail Xref: nightfall.franklin.ch alt.folklore.computers:182451 "John Savard" wrote in message news:4127ed5f.28197423@news.ecn.ab.ca... > On Sat, 21 Aug 2004 22:26:26 GMT, hoh@invalid.invalid (Goran > Larsson) wrote, in part: > > >Why did IBM desire breaks in the alphabetical sequence? Just compatibility > >with old habits or did the breaks give IBM some important advantage that > >the rest of the world didn't understand? > > The breaks in themselves were not desired, as such. > > ((..omitted..)) > > ASCII didn't come along till later - although it was around > by the time the 360 came along, the BCDIC 6-bit code had > long been in existence, and naturally, IBM wanted to > preserve as much compatibility as possible between old and > new. Not to cast doubts on any of Mr. Savard's comments (which, from my own memory seem quite accurate), there's an interesting examination of the history of character codes at http://tronweb.super-nova.co.jp/characcodehist.html which may add something to this, if you're sufficiently interested. - Bill ###### From: Rich Alderson Newsgroups: alt.folklore.computers Subject: Re: The ORIGIN of the Dominance of the 8-bit Byte Date: 22 Aug 2004 00:29:14 -0400 Organization: PANIX Public Access Internet and UNIX, NYC Lines: 32 Sender: alderson+news@panix5.panix.com Message-ID: References: <41190a85.715407@news.ecn.ab.ca> <411B57E5.DC9D2518@alum.mit.edu> <86u0v83z2a.fsf@my.domain> <41262052.61C819EC@earthlink.net> <41262d6a.9404558@news.ecn.ab.ca> <41275572.1465186@news.ecn.ab.ca> NNTP-Posting-Host: panix5.panix.com X-Trace: reader1.panix.com 1093148954 26218 166.84.1.5 (22 Aug 2004 04:29:14 GMT) X-Complaints-To: abuse@panix.com NNTP-Posting-Date: Sun, 22 Aug 2004 04:29:14 +0000 (UTC) X-Newsreader: Gnus v5.7/Emacs 21.3 Path: nightfall.franklin.ch!pfaff2.ethz.ch!news-zh.switch.ch!switch.ch!newscore.univie.ac.at!news-spur1.maxwell.syr.edu!news.maxwell.syr.edu!panix!not-for-mail Xref: nightfall.franklin.ch alt.folklore.computers:182452 jsavard@excxn.aNOSPAMb.cdn.invalid (John Savard) writes: First off, was it *really* necessary to quote your own 122-line message to add some commentary? > By the time the PDP-11 came out, lower-case letters already had a defined > place in ASCII. Thus, the 6-bit character would clearly have been limiting, > tying the computer to the past. A 7-bit character is something awkward to cut > into smaller parts, and few other computer makers used it. Yes--and had for 7 or 8 years. > DEC did already have a perfectly good line of machines with an 18-bit word; > and the (36-bit) PDP-10 showed that this can lead to 9-bit characters as > easily as 6-bit characters. Advances in electronic circuit technology, > however, led DEC at that time to want a "clean sheet of paper" to build a > design based on different principles; the PDP-11 has six general registers, > instead of doing all its arithmetic between the memory and one accumulator. You clearly know nothing about the PDP-6 and PDP-10: This architecture has 16 general-purpose registers (called accumulators in that universe of discourse)-- more general-purpose than those in the System/360 which was introduced at the same time as the PDP-6, since floating-point as well as integer arithmetic is done in them. Not all PDPs are the same. -- Rich Alderson | /"\ ASCII ribbon | news@alderson.users.panix.com | \ / campaign against | "You get what anybody gets. You get a lifetime." | x HTML mail and | --Death, of the Endless | / \ postings | ###### Sender: CStacy@BOHR Newsgroups: alt.folklore.computers Subject: Re: The ORIGIN of the Dominance of the 8-bit Byte References: <41190a85.715407@news.ecn.ab.ca> <411B57E5.DC9D2518@alum.mit.edu> <86u0v83z2a.fsf@my.domain> <41262052.61C819EC@earthlink.net> <41262d6a.9404558@news.ecn.ab.ca> <41275572.1465186@news.ecn.ab.ca> From: cstacy@news.dtpq.com (Christopher C. Stacy) Message-ID: Lines: 17 User-Agent: Gnus/5.09 (Gnus v5.9.0) Emacs/21.2 MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii Date: Sun, 22 Aug 2004 04:52:19 GMT NNTP-Posting-Host: 68.163.132.93 X-Complaints-To: abuse@verizon.net X-Trace: trndny07 1093150339 68.163.132.93 (Sun, 22 Aug 2004 00:52:19 EDT) NNTP-Posting-Date: Sun, 22 Aug 2004 00:52:19 EDT Path: nightfall.franklin.ch!pfaff2.ethz.ch!news-zh.switch.ch!switch.ch!solnet.ch!solnet.ch!news.glorb.com!cyclone1.gnilink.net!spamkiller2.gnilink.net!gnilink.net!trndny07.POSTED!8b1f9295!not-for-mail Xref: nightfall.franklin.ch alt.folklore.computers:182454 >>>>> On 22 Aug 2004 00:29:14 -0400, Rich Alderson ("Rich") writes: Rich> jsavard@excxn.aNOSPAMb.cdn.invalid (John Savard) writes: >> DEC did already have a perfectly good line of machines with an 18-bit word; >> and the (36-bit) PDP-10 showed that this can lead to 9-bit characters as >> easily as 6-bit characters. Advances in electronic circuit technology, >> however, led DEC at that time to want a "clean sheet of paper" to build a >> design based on different principles; the PDP-11 has six general registers, >> instead of doing all its arithmetic between the memory and one accumulator. Rich> You clearly know nothing about the PDP-6 and PDP-10: This architecture has 16 Rich> general-purpose registers (called accumulators in that universe of discourse)-- Rich> more general-purpose than those in the System/360 which was introduced at the Rich> same time as the PDP-6, since floating-point as well as integer arithmetic is Rich> done in them. Not to mention that each register could also be a stack pointer. ###### From: jsavard@excxn.aNOSPAMb.cdn.invalid (John Savard) Newsgroups: alt.folklore.computers Subject: Re: The ORIGIN of the Dominance of the 8-bit Byte Date: Sun, 22 Aug 2004 10:55:54 GMT Organization: Group Telecom - a 360networks Company Lines: 16 Message-ID: <41287b58.3594127@news.ecn.ab.ca> References: <41190a85.715407@news.ecn.ab.ca> <411B57E5.DC9D2518@alum.mit.edu> <86u0v83z2a.fsf@my.domain> <41262052.61C819EC@earthlink.net> <41262d6a.9404558@news.ecn.ab.ca> <41275572.1465186@news.ecn.ab.ca> NNTP-Posting-Host: h64-42-245-157.gtcust.grouptelecom.net X-Trace: utornnr1pp.grouptelecom.net 1093172157 26699 64.42.245.157 (22 Aug 2004 10:55:57 GMT) X-Complaints-To: abuse@gt.ca NNTP-Posting-Date: Sun, 22 Aug 2004 10:55:57 +0000 (UTC) X-Newsreader: Forte Free Agent 1.11/32.235 Path: nightfall.franklin.ch!pfaff2.ethz.ch!news-zh.switch.ch!irazu.switch.ch!switch.ch!news.maxwell.syr.edu!cyclone.bc.net!news.uunet.ca!newsfeed.grouptelecom.net!news.east.grouptelecom.net!not-for-mail Xref: nightfall.franklin.ch alt.folklore.computers:182462 On 22 Aug 2004 00:29:14 -0400, Rich Alderson wrote, in part: >You clearly know nothing about the PDP-6 and PDP-10: The PDP-4 and PDP-9 were DEC's 18-bit computers, and *they* did not have general registers. The PDP-11 could have been a cheaper PDP-15. It could *not* have been a PDP-10. They hardly could have made a 36-bit computer at the price of a 16-bit computer. So they did need a new architecture. John Savard http://home.ecn.ab.ca/~jsavard/index.html ###### From: jsavard@excxn.aNOSPAMb.cdn.invalid (John Savard) Newsgroups: alt.folklore.computers Subject: Re: The ORIGIN of the Dominance of the 8-bit Byte Date: Sun, 22 Aug 2004 11:08:54 GMT Organization: Group Telecom - a 360networks Company Lines: 27 Message-ID: <41287c62.3860018@news.ecn.ab.ca> References: <41190a85.715407@news.ecn.ab.ca> <41262d6a.9404558@news.ecn.ab.ca> <41275572.1465186@news.ecn.ab.ca> <8fNVc.6970$2L3.1221@newsread3.news.atl.earthlink.net> <4127ed5f.28197423@news.ecn.ab.ca> NNTP-Posting-Host: h64-42-245-157.gtcust.grouptelecom.net X-Trace: utornnr1pp.grouptelecom.net 1093172937 320 64.42.245.157 (22 Aug 2004 11:08:57 GMT) X-Complaints-To: abuse@gt.ca NNTP-Posting-Date: Sun, 22 Aug 2004 11:08:57 +0000 (UTC) X-Newsreader: Forte Free Agent 1.11/32.235 Path: nightfall.franklin.ch!pfaff2.ethz.ch!news-zh.switch.ch!switch.ch!news.ip-plus.net!newsfeed.ip-plus.net!news.tesion.net!news.belwue.de!newsfeed01.sul.t-online.de!newsfeed00.sul.t-online.de!t-online.de!border2.nntp.dca.giganews.com!nntp.giganews.com!nf3.bellglobal.com!snoopy.risq.qc.ca!news.uunet.ca!newsfeed.grouptelecom.net!news.east.grouptelecom.net!not-for-mail Xref: nightfall.franklin.ch alt.folklore.computers:182466 On Sun, 22 Aug 2004 00:16:40 -0400, "Bill Leary" wrote, in part: >Not to cast doubts on any of Mr. Savard's comments (which, from my own >memory seem quite accurate), there's an interesting examination of the >history of character codes at >http://tronweb.super-nova.co.jp/characcodehist.html which may add something >to this, if you're sufficiently interested. It _is_ interesting, although it fails to mention that ITA #1 is based on Baudot's code, while the completely different ITA #2, although often called Baudot code, because it was Emile Baudot who invented the basic principle, is based on the code originated by Donald Murray for his later printing telegraph device. May I recommend http://home.ecn.ab.ca/~jsavard/crypto/tele03.htm ... although this site of mine is not devoted to history, so it doesn't display the original Baudot code, so I shall also recommend http://www.vauxelectronics.com/gil/tty/docs/smith--teletype-codes.htm http://www.smithsrisca.demon.co.uk/binary-specimen.html John Savard http://home.ecn.ab.ca/~jsavard/index.html ###### From: stanb45@dial.pipex.com (Stan Barr) Newsgroups: alt.folklore.computers Subject: Re: The ORIGIN of the Dominance of the 8-bit Byte References: <41190a85.715407@news.ecn.ab.ca> <41262d6a.9404558@news.ecn.ab.ca> <41275572.1465186@news.ecn.ab.ca> <8fNVc.6970$2L3.1221@newsread3.news.atl.earthlink.net> <4127ed5f.28197423@news.ecn.ab.ca> <41287c62.3860018@news.ecn.ab.ca> Organization: Metropolis Grafix Reply-To: stanb45@dial.pipex.com Message-ID: X-Newsreader: slrn (0.9.5.2 UNIX) Date: 22 Aug 2004 12:45:21 GMT Lines: 17 NNTP-Posting-Date: 22 Aug 2004 12:45:21 GMT NNTP-Posting-Host: 62-241-188-39.dsl.pipex.com X-Trace: 1093178721 news-text.dial.pipex.com 20244 62.241.188.39:4123 X-Complaints-To: abuse@uk.uu.net Path: nightfall.franklin.ch!pfaff2.ethz.ch!news-zh.switch.ch!switch.ch!newscore.univie.ac.at!uninett.no!news.uit.no!newsfeed1.e.nsc.no!bnewspeer00.bru.ops.eu.uu.net.MISMATCH!bnewsoutpeer01.bru.ops.eu.uu.net!bnewsinpeer00.bru.ops.eu.uu.net!bnewspost00.bru.ops.eu.uu.net!emea.uu.net!not-for-mail Xref: nightfall.franklin.ch alt.folklore.computers:182476 On Sun, 22 Aug 2004 11:08:54 GMT, John Savard wrote: > >It _is_ interesting, although it fails to mention that ITA #1 is based >on Baudot's code, while the completely different ITA #2, although often >called Baudot code, because it was Emile Baudot who invented the basic >principle, Re-invented is perhaps a better description, Francis Bacon invented a 5-bit telegraph code in 1623, but he didn't have the hardware to make use of it :-) -- Cheers, Stan Barr stanb .at. dial .dot. pipex .dot. com (Remove any digits from the addresses when mailing me.) The future was never like this! ###### From: jsavard@excxn.aNOSPAMb.cdn.invalid (John Savard) Newsgroups: alt.folklore.computers Subject: Re: The ORIGIN of the Dominance of the 8-bit Byte Date: Sun, 22 Aug 2004 17:42:40 GMT Organization: Group Telecom - a 360networks Company Lines: 23 Message-ID: <4128da48.811961@news.ecn.ab.ca> References: <41190a85.715407@news.ecn.ab.ca> <41262d6a.9404558@news.ecn.ab.ca> < 41275572.1465186@news.ecn.ab.ca> <8fNVc.6970$2L3.1221@newsread3.news.atl.earthli nk.net> <4127ed5f.28197423@news.ecn.ab.ca> <41287c62.3860018@news.ecn.ab.ca> NNTP-Posting-Host: h64-42-245-132.gtcust.grouptelecom.net X-Trace: utornnr1pp.grouptelecom.net 1093196564 4867 64.42.245.132 (22 Aug 2004 17:42:44 GMT) X-Complaints-To: abuse@gt.ca NNTP-Posting-Date: Sun, 22 Aug 2004 17:42:44 +0000 (UTC) X-Newsreader: Forte Free Agent 1.11/32.235 Path: nightfall.franklin.ch!pfaff2.ethz.ch!news-zh.switch.ch!switch.ch!solnet.ch !solnet.ch!news.glorb.com!news.alt.net!newsfeed.grouptelecom.net!news.east.group telecom.net!not-for-mail Xref: nightfall.franklin.ch alt.folklore.computers:182496 On 22 Aug 2004 12:45:21 GMT, stanb45@dial.pipex.com (Stan Barr) wrote, in part: >Re-invented is perhaps a better description, Francis Bacon invented a 5-bit >telegraph code in 1623, but he didn't have the hardware to make use of it :-) True, true! The principle of using binary encoding for the letters of the alphabet was invented by him. But Baudot invented the principle of sending letters by wire in a fixed-length binary code, combining the inventions of Bacon and Morse. Donald Murray's printing telegraph worked like Baudot's, with five wires between machines and with a "letters space" and a "figures space". It wasn't until the Morkrum that teletypewriters took the form we know today, with a space, letters shift and figures shift, and a code modified from Murray's slightly in respect of punctuation marks. Thus, while the code in use today was invented in New Zealand, and the basic principle in France, the finishing touches for the modern teletypewriter were applied in America! John Savard http://home.ecn.ab.ca/~jsavard/index.html ###### From: David R Brooks Newsgroups: alt.folklore.computers Subject: Re: The ORIGIN of the Dominance of the 8-bit Byte Date: Mon, 23 Aug 2004 06:37:57 +0800 Message-ID: References: <41190a85.715407@news.ecn.ab.ca> <41262d6a.9404558@news.ecn.ab.ca> < 41275572.1465186@news.ecn.ab.ca> <8fNVc.6970$2L3.1221@newsread3.news.atl.earthli nk.net> <4127ed5f.28197423@news.ecn.ab.ca> <41287c62.3860018@news.ecn.ab.ca> <4128da48.811961@news.ecn.ab.ca> X-Newsreader: Forte Agent 1.92/32.572 X-No-Archive: yes MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit Lines: 18 NNTP-Posting-Host: 203.59.155.35 X-Trace: 1093214297 per-qv1-newsreader-01.iinet.net.au 22822 203.59.155.35 Path: nightfall.franklin.ch!pfaff2.ethz.ch!news-zh.switch.ch!switch.ch!solnet.ch !solnet.ch!newspeer1.nwr.nac.net!nntp.cifnet.net!HSNX.atgi.net!newsfeed.iinet.ne t.au!newsfeed.iinet.net.au!per-qv1-newsreader-01.iinet.net.au!not-for-mail Xref: nightfall.franklin.ch alt.folklore.computers:182524 The listing I saw of Bacon's code shows the letters A-Z assigned in a naturla binary-counting order. If correct, this seems to imply that Bacon also discovered binary addition (you can't count without an addition operator). Of course, Bacon proposed his system for steganography, in which 2 different typefaces would stand for '1' & '0' in a concealing text text. jsavard@excxn.aNOSPAMb.cdn.invalid (John Savard) wrote: :On 22 Aug 2004 12:45:21 GMT, stanb45@dial.pipex.com (Stan Barr) wrote, :in part: : :>Re-invented is perhaps a better description, Francis Bacon invented a 5-bit :>telegraph code in 1623, but he didn't have the hardware to make use of it :-) : :True, true! The principle of using binary encoding for the letters of :the alphabet was invented by him. ###### Message-ID: <412938C2.B08E71D0@yahoo.com> From: CBFalconer Reply-To: cbfalconer@worldnet.att.net Organization: Ched Research X-Mailer: Mozilla 4.75 [en] (Win98; U) X-Accept-Language: en MIME-Version: 1.0 Newsgroups: alt.folklore.computers Subject: Re: The ORIGIN of the Dominance of the 8-bit Byte References: <41190a85.715407@news.ecn.ab.ca> <41262d6a.9404558@news.ecn.ab.ca> < 41275572.1465186@news.ecn.ab.ca> <8fNVc.6970$2L3.1221@newsread3.news.atl.earthli nk.net> <4127ed5f.28197423@news.ecn.ab.ca> <41287c62.3860018@news.ecn.ab.ca> <4128da48.811961@news.ecn.ab.ca> Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit Lines: 30 Date: Mon, 23 Aug 2004 00:30:45 GMT NNTP-Posting-Host: 12.76.134.252 X-Complaints-To: abuse@worldnet.att.net X-Trace: bgtnsc04-news.ops.worldnet.att.net 1093221045 12.76.134.252 (Mon, 23 Au g 2004 00:30:45 GMT) NNTP-Posting-Date: Mon, 23 Aug 2004 00:30:45 GMT Path: nightfall.franklin.ch!pfaff2.ethz.ch!news-zh.switch.ch!irazu.switch.ch!swi tch.ch!news.maxwell.syr.edu!wn13feed!worldnet.att.net!bgtnsc04-news.ops.worldnet .att.net.POSTED!53ab2750!not-for-mail Xref: nightfall.franklin.ch alt.folklore.computers:182532 John Savard wrote: > ... snip ... > > True, true! The principle of using binary encoding for the letters of > the alphabet was invented by him. But Baudot invented the principle of > sending letters by wire in a fixed-length binary code, combining the > inventions of Bacon and Morse. > > Donald Murray's printing telegraph worked like Baudot's, with five wires > between machines and with a "letters space" and a "figures space". It > wasn't until the Morkrum that teletypewriters took the form we know > today, with a space, letters shift and figures shift, and a code > modified from Murray's slightly in respect of punctuation marks. > > Thus, while the code in use today was invented in New Zealand, and the > basic principle in France, the finishing touches for the modern > teletypewriter were applied in America! What about the semaphore code used at sea, and which you may have learned in Boy Scouts for a merit badge? I was thinking about it recently and can't remember how those arm positions could encode 26 letters. It seems impossible. -- Chuck F (cbfalconer@yahoo.com) (cbfalconer@worldnet.att.net) Available for consulting/temporary embedded and systems. USE worldnet address! ###### From: D.J. Newsgroups: alt.folklore.computers Subject: Re: The ORIGIN of the Dominance of the 8-bit Byte Date: Sun, 22 Aug 2004 21:19:58 -0500 Organization: TychoTown Tycho Crater Ice Cream Parlour Message-ID: <4skii0517off39p261q7lmddmaknb5f7rr@4ax.com> Distribution: world Reply-To: D.J. References: <41190a85.715407@news.ecn.ab.ca> <41262d6a.9404558@news.ecn.ab.ca> < 41275572.1465186@news.ecn.ab.ca> <8fNVc.6970$2L3.1221@newsread3.news.atl.earthli nk.net> <4127ed5f.28197423@news.ecn.ab.ca> <41287c62.3860018@news.ecn.ab.ca> <4128da48.811961@news.ecn.ab.ca> <412938C2.B08E71D 0@yahoo.com> X-Newsreader: Forte Agent 2.0/32.652 MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit X-Complaints-To: abuse@supernews.com Lines: 16 Path: nightfall.franklin.ch!pfaff2.ethz.ch!news-zh.switch.ch!switch.ch!news.imp. ch!news.imp.ch!solnet.ch!solnet.ch!newsfeed.wirehub.nl!62.58.50.20.MISMATCH!nntp feed.zonnet.nl!in.100proofnews.com!tdsnet-transit!newspeer.tds.net!newspeer.radi x.net!news-spur1.maxwell.syr.edu!news.maxwell.syr.edu!sn-xit-03!sn-xit-04!sn-xit -01!sn-post-02!sn-post-01!supernews.com!corp.supernews.com!not-for-mail Xref: nightfall.franklin.ch alt.folklore.computers:182549 CBFalconer wrote: ] What about the semaphore code used at sea, and which you may have ] learned in Boy Scouts for a merit badge? I was thinking about it ] recently and can't remember how those arm positions could encode ] 26 letters. It seems impossible. Here ya go: http://www.anbg.gov.au/flags/semaphore.html JimP. -- http://www.linuxgazette.net/ Linux Gazette http://evergame.drivein-jim.net/ August 6, 2004 Everquest http://www.drivein-jim.net/ July 14, 2004: http://crestar.drivein-jim.net/new.html August 15, 2004 AD&D ###### From: Marco S Hyman Newsgroups: alt.folklore.computers Subject: Re: The ORIGIN of the Dominance of the 8-bit Byte Date: 22 Aug 2004 19:48:06 -0700 Organization: S.N.A.F.U. -- http://www.snafu.org/ Lines: 12 Sender: marc@neko.snafu.org Message-ID: References: <41190a85.715407@news.ecn.ab.ca> <41262d6a.9404558@news.ecn.ab.ca> < 41275572.1465186@news.ecn.ab.ca> <8fNVc.6970$2L3.1221@newsread3.news.atl.earthli nk.net> <4127ed5f.28197423@news.ecn.ab.ca> <41287c62.3860018@news.ecn.ab.ca> <4128da48.811961@news.ecn.ab.ca> <412938C2.B08E71D0 @yahoo.com> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii X-Trace: news.uni-berlin.de WMW1Sr0v3SC6R7npiv6nXAed4fqTion0ACBvqza5oY9cuD8yE= X-Orig-Path: dumbcat.snafu.org!_news User-Agent: Gnus/5.09 (Gnus v5.9.0) Emacs/21.3 Path: nightfall.franklin.ch!pfaff2.ethz.ch!news-zh.switch.ch!irazu.switch.ch!swi tch.ch!news.belwue.de!fu-berlin.de!uni-berlin.de!not-for-mail Xref: nightfall.franklin.ch alt.folklore.computers:182555 CBFalconer writes: > What about the semaphore code used at sea, and which you may have > learned in Boy Scouts for a merit badge? I was thinking about it > recently and can't remember how those arm positions could encode > 26 letters. It seems impossible. Google is your friend :-) http://www.anbg.gov.au/flags/semaphore.html // marc ###### X-Trace-PostClient-IP: 68.147.131.211 From: Brian Inglis Newsgroups: alt.folklore.computers Subject: Re: The ORIGIN of the Dominance of the 8-bit Byte Organization: Systematic Software Reply-To: Brian.Inglis@SystematicSw.ab.ca Message-ID: References: <41190a85.715407@news.ecn.ab.ca> <411B57E5.DC9D2518@alum.mit.edu> <86u0v83z2a.fsf@my.domain> <41262052.6 1C819EC@earthlink.net> <41262d6a.9404558@news.ecn.ab.ca> <41275572.1465186@news. ecn.ab.ca> <41287b58.3594127@news.ecn.ab.ca> X-Newsreader: Forte Agent 1.93/32.576 English (American) MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit Lines: 31 Date: Mon, 23 Aug 2004 08:44:47 GMT NNTP-Posting-Host: 24.71.223.147 X-Complaints-To: abuse@shaw.ca X-Trace: pd7tw1no 1093250687 24.71.223.147 (Mon, 23 Aug 2004 02:44:47 MDT) NNTP-Posting-Date: Mon, 23 Aug 2004 02:44:47 MDT Path: nightfall.franklin.ch!pfaff2.ethz.ch!news-zh.switch.ch!switch.ch!solnet.ch !solnet.ch!news.glorb.com!border1.nntp.dca.giganews.com!nntp.giganews.com!pd7cy2 so!shaw.ca!pd7tw1no.POSTED!53ab2750!not-for-mail Xref: nightfall.franklin.ch alt.folklore.computers:182565 On Sun, 22 Aug 2004 10:55:54 GMT in alt.folklore.computers, jsavard@excxn.aNOSPAMb.cdn.invalid (John Savard) wrote: >On 22 Aug 2004 00:29:14 -0400, Rich Alderson > wrote, in part: > >>You clearly know nothing about the PDP-6 and PDP-10: > >The PDP-4 and PDP-9 were DEC's 18-bit computers, and *they* did not have >general registers. Weren't the -1 and -7 too? I know the -15 was also 18 bit. >The PDP-11 could have been a cheaper PDP-15. The PDP-15 was a flip-chip single AC design with a fast FPP and GPU and limited physical memory, like a process control control machine with other stuff bolted on to make it a scientific graphics machine. >It could *not* have been a PDP-10. They hardly could have made a 36-bit >computer at the price of a 16-bit computer. So they did need a new >architecture. Possible with transistors on PCBs, but might have been considered too risky with an expensive mainframe system. -- Thanks. Take care, Brian Inglis Calgary, Alberta, Canada Brian.Inglis@CSi.com (Brian dot Inglis at SystematicSw dot ab dot ca) fake address use address above to reply ###### From: jsavard@excxn.aNOSPAMb.cdn.invalid (John Savard) Newsgroups: alt.folklore.computers Subject: Re: The ORIGIN of the Dominance of the 8-bit Byte Date: Mon, 23 Aug 2004 11:05:02 GMT Organization: Group Telecom - a 360networks Company Lines: 14 Message-ID: <4129cf01.52022581@news.ecn.ab.ca> References: <41190a85.715407@news.ecn.ab.ca> <41262d6a.9404558@news.ecn.ab.ca> < 41275572.1465186@news.ecn.ab.ca> <8fNVc.6970$2L3.1221@newsread3.news.atl.earthli nk.net> <4127ed5f.28197423@news.ecn.ab.ca> <41287c62.3860018@news.ecn.ab.ca> <4128da48.811961@news.ecn.ab.ca> <412938C2.B08E71D 0@yahoo.com> NNTP-Posting-Host: h64-42-245-113.gtcust.grouptelecom.net X-Trace: utornnr1pp.grouptelecom.net 1093259107 8347 64.42.245.113 (23 Aug 2004 11:05:07 GMT) X-Complaints-To: abuse@gt.ca NNTP-Posting-Date: Mon, 23 Aug 2004 11:05:07 +0000 (UTC) X-Newsreader: Forte Free Agent 1.11/32.235 Path: nightfall.franklin.ch!pfaff2.ethz.ch!news-zh.switch.ch!irazu.switch.ch!swi tch.ch!news.maxwell.syr.edu!cyclone.bc.net!sjc1.usenetserver.com!news.usenetserv er.com!wesley.videotron.net!newsfeed.grouptelecom.net!news.east.grouptelecom.net !not-for-mail Xref: nightfall.franklin.ch alt.folklore.computers:182578 On Mon, 23 Aug 2004 00:30:45 GMT, CBFalconer wrote, in part: >What about the semaphore code used at sea, and which you may have >learned in Boy Scouts for a merit badge? I was thinking about it >recently and can't remember how those arm positions could encode >26 letters. It seems impossible. http://home.ecn.ab.ca/~jsavard/crypto/intro.htm has a picture; there are 28 possibilities, 8*7/2, as might be expected. John Savard http://home.ecn.ab.ca/~jsavard/index.html