Article 7029 of alt.sys.pdp10: Path: news3.best.com!news2.best.com!newsfeed.berkeley.edu!newsfeed.mathworks.com!wn3feed!worldnet.att.net!207.24.196.41!nntphub.cb.lucent.com!news From: Dennis Ritchie Newsgroups: alt.sys.pdp8,alt.sys.pdp10,alt.folklore.computers Subject: Re: Shell scripts Date: Thu, 16 Mar 2000 03:44:34 +0000 Organization: Bell Labs / Lucent Technologies Lines: 43 Message-ID: <38D058A2.A3178676@bell-labs.com> References: <951553598.660875@shelley.paradise.net.nz> <951604376.786789@shelley.paradise.net.nz> <845.92T6T4842347ghira@mistral.co.uk> <38CF9FF1.B993781A@mail.ptd.net> NNTP-Posting-Host: trux.cs.bell-labs.com Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit X-Mailer: Mozilla 4.61 [en] (WinNT; U) X-Accept-Language: en Xref: news3.best.com alt.sys.pdp8:4418 alt.sys.pdp10:7029 alt.folklore.computers:152974 Mark Crispin wrote, inter alia: > Multics was also written in a high level language, and thus was > "portable". So were several other operating systems, such as on the > Univac (written in Algol). But they weren't ported, except to similar machines in the same line of descent. > > But being written in assembly language doesn't make software non-portable. > TOPS-20 was ported to processors built by three other manufacturers (four > if you stretch TOPS-20 to include Tenex). IBM OS/360 was ported to a > number of different manufacturers' processors as well. So was MS-DOS. Likewise. Amdahl and IBM? AMD and Intel? DEC and Foonly? Gimme a break. > > Nor was the original AT&T UNIX all that portable. A number of chunks were > written in PDP-11 assembly language, and it was quite an effort to get it > running on the VAX. If you want to talk about the *first* UNIX, it was > written in PDP-7 assembly language, and the first PDP-11 UNIX was in > PDP-11 assembly language. The PDP-11 UNIX was later rewritten into C when > they got a large enough PDP-11. Actually, once it started happening around 1977, ports to significantly different architectures were done fairly fast. "Different" means things like Interdata, IBM 360/370, DEC Vax, Univac 1100, Intel 8086. > > As Barb points out, the "portability" of UNIX was more of a policy > decision of its original owners (AT&T) than something technical. But the > original AT&T licenses weren't all that nice. It was BSD that started the > explosive growth of UNIX. I wasn't aware of "policy decisions" by AT&T that caused the work, not counting the one that kept our paychecks coming in. And the original AT&T licenses were precisely those under which the important BSD work in the 1980s was carried out, not to mention the growth of the workstation industry. (I don't by any means discount the BSD work, which was enormously important, but the same important technical parts of it were done with that not-that-nice license, though it did ultimately chafe). Dennis Article 7031 of alt.sys.pdp10: Path: news3.best.com!news2.best.com!newsfeed.berkeley.edu!news.u.washington.edu!Tomobiki-Cho.CAC.Washington.EDU!mrc From: Mark Crispin Newsgroups: alt.sys.pdp8,alt.sys.pdp10,alt.folklore.computers Subject: Re: Shell scripts Date: Wed, 15 Mar 2000 23:45:20 -0800 Organization: Networks & Distributed Computing Lines: 57 Message-ID: References: <951553598.660875@shelley.paradise.net.nz> <951604376.786789@shelley.paradise.net.nz> <845.92T6T4842347ghira@mistral.co.uk> <38CF9FF1.B993781A@mail.ptd.net> <38D058A2.A3178676@bell-labs.com> NNTP-Posting-Host: tomobiki-cho.cac.washington.edu Mime-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII X-Trace: nntp1.u.washington.edu 953192723 14424 (None) 140.142.17.38 X-Complaints-To: help@cac.washington.edu NNTP-Posting-User: wambold To: Dennis Ritchie In-Reply-To: <38D058A2.A3178676@bell-labs.com> Xref: news3.best.com alt.sys.pdp8:4420 alt.sys.pdp10:7031 alt.folklore.computers:152979 On Thu, 16 Mar 2000, Dennis Ritchie wrote: > > As Barb points out, the "portability" of UNIX was more of a policy > > decision of its original owners (AT&T) than something technical. > I wasn't aware of "policy decisions" by AT&T that caused the work, not > counting the one that kept our paychecks coming in. "Good policy decision" could also mean "lack of bad policy decision." Had AT&T behaved the way that DEC or Honeywell did, there would have been no release of UNIX outside of AT&T until the 3B series was produced. The PDP-11, Interdata, etc. versions never would have seen the outside of Bell Labs. The UNIX license would have prohibited running UNIX on anything other than a 3B system. It isn't clear to me whether it was conscious corporate decision, accident of history, or guerilla action on the part of the programmers. You can probably comment on this point... ;-) The point is, AT&T behaved differently. Either AT&T made a policy decision not to restrict UNIX to AT&T-supplied hardware; or AT&T failed to make a policy decision to restrict UNIX to AT&T-supplied hardware. Either way, it was possible by policy to port UNIX to other platforms. And in the end run, that's what counted. However, a policy (or lack thereof), no matter how laudable, isn't a technical aspect. Technically, there were other systems that were as portable as UNIX; but policy blocked their portability. I contend that policy is more important that technology in deciding what succeeds. VHS vs. Beta; Intel vs. Motorola;... history is littered with superior technology that died because of bad policy of its owners. UNIX had undeniably, overwhelmingly superior policy. Technology is an entirely different matter. The claim that UNIX is unconditionally superior technology to [Multics | ITS | TOPS-20 | LispM] leaves a bitter taste for the erstwhile partisans of the latter. It's the same reason that a claim that VHS is superior technology to Beta annoys former Beta fans. It's the sign of a poor winner. I know that you (DMR) have never been this way. As I recall, you once made a very gracious statement about Ray Tomlinson and Dan Murphy that was much appreciated in the Tenex/TOPS-20 world. I wish that I could say the same about certain other individuals. I mean, sheesh, not that long ago someone was insisting to me that UNIX invented device-independent I/O and file/memory mapping! -- Mark -- * RCW 19.190 notice: This email address is located in Washington State. * * Unsolicited commercial email may be billed $500 per message. * Science does not emerge from voting, party politics, or public debate. Article 7042 of alt.sys.pdp10: Path: news3.best.com!news1.best.com!nntp.primenet.com!nntp.gctr.net!howland.erols.net!newsfeed.skycache.com!portc01.blue.aol.com!newsstand.cit.cornell.edu!news.graphics.cornell.edu!news From: westin*nospam@graphics.cornell.edu (Stephen H. Westin) Newsgroups: alt.sys.pdp8,alt.sys.pdp10,alt.folklore.computers Subject: Unix history/policy (was: Shell scripts) Date: 16 Mar 2000 12:54:18 -0500 Organization: Cornell University Program of Computer Graphics Lines: 40 Sender: westin@DIESEL Message-ID: References: <38D058A2.A3178676@bell-labs.com> <8ar58l$btj$1@agate.berkeley.edu> NNTP-Posting-Host: diesel.graphics.cornell.edu X-Newsreader: Gnus v5.5/Emacs 20.3 Xref: news3.best.com alt.sys.pdp8:4428 alt.sys.pdp10:7042 alt.folklore.computers:153018 korpela@islay.ssl.berkeley.edu (Eric J. Korpela) writes: > In article , > Mark Crispin wrote: > >On Thu, 16 Mar 2000, Dennis Ritchie wrote: > >> > As Barb points out, the "portability" of UNIX was more of a policy > >> > decision of its original owners (AT&T) than something technical. > >> I wasn't aware of "policy decisions" by AT&T that caused the work, not > >> counting the one that kept our paychecks coming in. > > > >"Good policy decision" could also mean "lack of bad policy decision." > > > >Had AT&T behaved the way that DEC or Honeywell did, there would have been > >no release of UNIX outside of AT&T until the 3B series was produced. The > >PDP-11, Interdata, etc. versions never would have seen the outside of Bell > >Labs. The UNIX license would have prohibited running UNIX on anything > >other than a 3B system. > > IIRC, AT&T was prohibited at the time from selling computer hardware. That would be from the consent decree of 1956, which I believe was from a suit between IBM and AT&T. It basically said that IBM wouldn't do phones, and AT&T wouldn't do computers. > I think > it was only after the break up of the phone company in the 80s that AT&T was > allowed to enter the computer sales buisness. Well, I believe that 1976 was the end of the consent decree (though I don't know why I believe that); one result was the release of a new generation of Teletype terminals (the 40 series) to catch up with the rest of the world. It was too little, too late, of course; at that point, I think DEC dominated with DECwriters and VT100s. -- -Stephen H. Westin Any information or opinions in this message are mine: they do not represent the position of Cornell University or any of its sponsors. Article 7046 of alt.sys.pdp10: Path: news3.best.com!news1.best.com!su-news-hub1.bbnplanet.com!nycmny1-snh1.gtei.net!news.gtei.net!newsfeed.mathworks.com!panix!news.panix.com!not-for-mail From: jeffj@panix.com (Jeff Jonas) Newsgroups: alt.sys.pdp8,alt.sys.pdp10,alt.folklore.computers Subject: Re: Unix history/policy (was: Shell scripts) Date: 16 Mar 2000 17:39:23 -0500 Organization: Jeff's House of Electronic Parts Lines: 32 Message-ID: <8arnqr$amj$1@panix.com> References: <8ar58l$btj$1@agate.berkeley.edu> <8arb9r$1eci$1@news.enteract.com> NNTP-Posting-Host: panix.com X-Trace: news.panix.com 953246364 3111 166.84.0.226 (16 Mar 2000 22:39:24 GMT) X-Complaints-To: abuse@panix.com NNTP-Posting-Date: 16 Mar 2000 22:39:24 GMT X-Newsposter: trn 4.0-test55 (26 Feb 97) Xref: news3.best.com alt.sys.pdp8:4433 alt.sys.pdp10:7046 alt.folklore.computers:153027 >>> IIRC, AT&T was prohibited at the time from selling computer hardware. >> That would be from the consent decree of 1956, which I believe was >> from a suit between IBM and AT&T. It basically said that IBM wouldn't >> do phones, and AT&T wouldn't do computers. >IBM and AT&T both signed consent decrees in 1956, but unrelated to >one another. Both were the result of antitrust proceedings by the >US government. IBM agreed to sell equipment instead of only leasing >it; AT&T agreed to restrict itself to the telecommunications business. *grunt* nobody's giving authoritative references, but hey, it's only USENET :-) I think many of thought "that makes since" in light of the more recent tit-for-tat posturing of the early 90s where AT&T was pondering video on demand and challenging the cable industry, and the cable industry wanted to offer phone service. Here we are 10 years later and cable service still sux and is overpriced. The closest they've come to telephony is offering cable internet, which is STILL slow to deploy in most areas. I can't follow the telecom industry as closely as I did but I believe that AT&T did succeed in making video on demand technology which servers niche markets. Still no picturephone, though and the video/picture phones were a flop. Even videoconferencing seems to be a passing fad. "Gee, Brain, what are we going to do with all this left over bandwidth" "set up a porn web server, Pinky. It'll help me ... RULE THE WORLD" -- Jeffrey Jonas jeffj@panix(dot)com The original Dr. JCL and Mr .hide Article 7047 of alt.sys.pdp10: Path: news3.best.com!news1.best.com!su-news-hub1.bbnplanet.com!news.gtei.net!newsfeed.berkeley.edu!newsfeed.enteract.com!betanews.enteract.com!not-for-mail From: Eric Fischer Newsgroups: alt.sys.pdp8,alt.sys.pdp10,alt.folklore.computers Subject: Re: Unix history/policy (was: Shell scripts) Date: 16 Mar 2000 23:18:47 GMT Organization: EnterAct Corp Turbo-Elite News Server Lines: 34 Message-ID: <8arq4n$1re5$1@news.enteract.com> References: <8ar58l$btj$1@agate.berkeley.edu> <8arb9r$1eci$1@news.enteract.com> <8arnqr$amj$1@panix.com> NNTP-Posting-Host: shell-3.enteract.com X-Trace: news.enteract.com 953248727 60869 207.229.143.42 (16 Mar 2000 23:18:47 GMT) X-Complaints-To: abuse@enteract.com NNTP-Posting-Date: 16 Mar 2000 23:18:47 GMT X-Newsreader: trn 4.0-test72 (19 April 1999) Originator: enf@enteract.com (Eric Fischer) Xref: news3.best.com alt.sys.pdp8:4434 alt.sys.pdp10:7047 alt.folklore.computers:153031 Jeff Jonas wrote: > >IBM and AT&T both signed consent decrees in 1956, but unrelated to > >one another. Both were the result of antitrust proceedings by the > >US government. IBM agreed to sell equipment instead of only leasing > >it; AT&T agreed to restrict itself to the telecommunications business. > > *grunt* nobody's giving authoritative references, but hey, it's only > USENET :-) You want references? OK, you get references. John Brooks, _Telephone: the First Hundred Years_, Harper & Row, 1976: The agreement, or consent decree, settling the case was wrapped up in Washington right after New Year's Day, 1956.... First, Western Electric would not be separated from AT&T; second, Western Electric, except for its government defense work, agreed to confine itself to manufacturing equipment of types bought by the Bell System, rather than seeking other markets; third, the Bell System agreed not to engage in any business other than common-carrier telecommunications and "incidental operations"; and fourth, the Bell System agreed to grant nonexclusive licenses and related technical information to any applicant on fair terms. (pp. 254-255) The only history of IBM I have at hand is _THINK_ by William Rodgers, Stein and Day, 1969, and it doesn't give any solid details about IBM's consent decree. But the text of the decree seems to be available at http://www.essential.org/antitrust/ibm/ibm1956cd.html I can dig for more details if that's not good enough. eric Article 7045 of alt.sys.pdp10: Path: news3.best.com!news1.best.com!su-news-hub1.bbnplanet.com!nycmny1-snh1.gtei.net!news.gtei.net!newsfeed.mathworks.com!panix!news.panix.com!not-for-mail From: jeffj@panix.com (Jeff Jonas) Newsgroups: alt.sys.pdp8,alt.sys.pdp10,alt.folklore.computers Subject: Re: Unix history/policy (was: Shell scripts) Date: 16 Mar 2000 17:32:18 -0500 Organization: Jeff's House of Electronic Parts Lines: 43 Message-ID: <8arndi$9k6$1@panix.com> References: <8ar58l$btj$1@agate.berkeley.edu> NNTP-Posting-Host: panix.com X-Trace: news.panix.com 953245939 2982 166.84.0.226 (16 Mar 2000 22:32:19 GMT) X-Complaints-To: abuse@panix.com NNTP-Posting-Date: 16 Mar 2000 22:32:19 GMT X-Newsposter: trn 4.0-test55 (26 Feb 97) Xref: news3.best.com alt.sys.pdp8:4432 alt.sys.pdp10:7045 alt.folklore.computers:153026 >> IIRC, AT&T was prohibited at the time from selling computer hardware. >That would be from the consent decree of 1956, which I believe was >from a suit between IBM and AT&T. It basically said that IBM wouldn't >do phones, and AT&T wouldn't do computers. Geez, and both failed miserably when they tried Rolm was IBM's phone system and IBM withdrew/gave up. AT&T tried selling their own PCs and workstations usually based on the WeCo 32xxx CPU and named 3b2/5/...: AT&T tried selling Olivetti and Convergent PCs too. All of AT&T Information System was totally disolved in the 90s. >Well, I believe that 1976 was the end of the consent decree (though I >don't know why I believe that); one result was the release of a new >generation of Teletype terminals (the 40 series) to catch up with the >rest of the world. It was too little, too late, of course; at that >point, I think DEC dominated with DECwriters and VT100s. I am still sorrowed that the Teletype brand-name was not maintained since it probaly had more name-brand recognition than any other name! I think you're right that in the 70s, Teletype machines were big and clunky compared to the competitors, but they did some ground-breaking work in the 80s with the BLIT, 5620 and other high resolution graphic and windowing "smart" terminals. And I used to have a stack of Teletype Dataspeed 40 chain printers that were so rugged they refused to die! And cheap to run too: used DecWriter ribbons! There was another factor to Teletype's downfall: they tried desperately to break into the IBM mainframe market. They had IBM compatible terminal cluster controllers and terminals, yet few places outside of AT&T risked the ire of IBM and actually deployed them. Many shops remained "true blue" using only genuine IBM terminals and all despite the higher cost and being older technology. The end of the consent decree allowed for me to work at AT&T: I mostly consulted to AT&T's Information System: developing the Unix OS and applications. -- Jeffrey Jonas jeffj@panix(dot)com The original Dr. JCL and Mr .hide