Sender: lynn@LYNNPC Newsgroups: alt.folklore.computers Subject: 3270 protocol Reply-To: Anne & Lynn Wheeler From: Anne & Lynn Wheeler Message-ID: Organization: Wheeler&Wheeler Lines: 13 User-Agent: Gnus/5.0808 (Gnus v5.8.8) Emacs/20.7 MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii Date: Wed, 26 Sep 2001 11:31:32 GMT NNTP-Posting-Host: 199.174.229.213 X-Complaints-To: abuse@earthlink.net X-Trace: newsread2.prod.itd.earthlink.net 1001503892 199.174.229.213 (Wed, 26 Sep 2001 04:31:32 PDT) NNTP-Posting-Date: Wed, 26 Sep 2001 04:31:32 PDT X-Received-Date: Wed, 26 Sep 2001 04:28:05 PDT (newsmaster1.prod.itd.earthlink.net) Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!news-ge.switch.ch!newsfeed00.sul.t-online.de!t-online.de!news-spur1.maxwell.syr.edu!news.maxwell.syr.edu!newsfeed.direct.ca!look.ca!newsfeed1.earthlink.net!newsfeed.earthlink.net!newsmaster1.prod.itd.earthlink.net!newsread2.prod.itd.earthlink.net.POSTED!not-for-mail Xref: chonsp.franklin.ch alt.folklore.computers:91142 a couple days ago somebody sent me a question on 3270 protocol which I didn't know the answer to ... but it jog'ed some memory cells that i've been trying to remember what were the terms used for the 3272/3277 and 3274/3278/9 protocol. I have some vague recollection that one of the terms was CUT and may refer to the 3272/3277 protocol ... but I can't remember the other term (and/or even sure CUT is one of the terms). does anybody remember the two terms? -- Anne & Lynn Wheeler | lynn@garlic.com - http://www.garlic.com/~lynn/ ###### From: jmaynard@thebrain.conmicro.cx (Jay Maynard) Newsgroups: alt.folklore.computers Subject: Re: 3270 protocol References: Reply-To: jmaynard@conmicro.cx Message-ID: User-Agent: slrn/0.9.6.4 (Linux) Lines: 19 Date: Wed, 26 Sep 2001 13:04:47 GMT NNTP-Posting-Host: 162.33.222.2 X-Complaints-To: abuse@coretel.net X-Trace: news.abs.net 1001509487 162.33.222.2 (Wed, 26 Sep 2001 09:04:47 EDT) NNTP-Posting-Date: Wed, 26 Sep 2001 09:04:47 EDT Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!news-ge.switch.ch!news-fra1.dfn.de!news0.de.colt.net!colt.net!nycmny1-snf1.gtei.net!nycmny1-snh1.gtei.net!cpk-news-hub1.bbnplanet.com!news.gtei.net!nntp.abs.net!news.abs.net!not-for-mail Xref: chonsp.franklin.ch alt.folklore.computers:91115 On Wed, 26 Sep 2001 11:31:32 GMT, Anne & Lynn Wheeler wrote: >a couple days ago somebody sent me a question on 3270 protocol which I >didn't know the answer to ... but it jog'ed some memory cells that >i've been trying to remember what were the terms used for the >3272/3277 and 3274/3278/9 protocol. I have some vague recollection >that one of the terms was CUT and may refer to the 3272/3277 protocol >... but I can't remember the other term (and/or even sure CUT is one >of the terms). Close. CUT is the protocol used by dumb 327x devices (control-unit terminal). I don't recall that the 3277 version was called anything besides "type A" (referring to the customization parameter you had to set; when customizing a 3274, you set up a number of type A ports for 3277s, and a number of type B ports for 3278s). (Or maybe it's the other way around.) The other protocol you're thinking of is DFT (distributed function terminal), which offloaded a bunch of intelligence to the downstream device. The device's microcode was downloaded from storage in the controller. The first such device was the 3290. ###### From: Don Quixote Newsgroups: alt.folklore.computers Subject: Re: 3270 protocol Reply-To: donq+afc@eclipse.net Message-ID: <5874rtc9cnovjjjet9t6jt7pfn95n0dasf@news.fishead.meowkitty.com> References: X-Newsreader: Forte Agent 1.8/32.548 MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit Lines: 14 Date: Wed, 26 Sep 2001 14:24:02 -0400 NNTP-Posting-Host: 207.207.196.181 X-Complaints-To: news@netcarrier.com X-Trace: news.netcarrier.net 1001538327 207.207.196.181 (Wed, 26 Sep 2001 17:05:27 EDT) NNTP-Posting-Date: Wed, 26 Sep 2001 17:05:27 EDT Organization: NetCarrier Internet Services Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!news-ge.switch.ch!newsfeed00.sul.t-online.de!t-online.de!news-spur1.maxwell.syr.edu!news.maxwell.syr.edu!cpk-news-hub1.bbnplanet.com!news.gtei.net!newsfeed1.cidera.com!Cidera!news.netcarrier.net!not-for-mail Xref: chonsp.franklin.ch alt.folklore.computers:91420 Anne & Lynn Wheeler might have said: >I have some vague recollection >that one of the terms was CUT and may refer to the 3272/3277 protocol >... but I can't remember the other term (and/or even sure CUT is one >of the terms). > >does anybody remember the two terms? CUT and DFT? in *very* broad terms, the first is dumb, the second smart. ###### Sender: lynn@LYNNPC Newsgroups: alt.folklore.computers Subject: Re: 3270 protocol References: <5874rtc9cnovjjjet9t6jt7pfn95n0dasf@news.fishead.meowkitty.com> Reply-To: Anne & Lynn Wheeler From: Anne & Lynn Wheeler Message-ID: Organization: Wheeler&Wheeler Lines: 44 User-Agent: Gnus/5.0808 (Gnus v5.8.8) Emacs/20.7 MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii Date: Wed, 26 Sep 2001 22:24:51 GMT NNTP-Posting-Host: 199.174.233.179 X-Complaints-To: abuse@earthlink.net X-Trace: newsread1.prod.itd.earthlink.net 1001543091 199.174.233.179 (Wed, 26 Sep 2001 15:24:51 PDT) NNTP-Posting-Date: Wed, 26 Sep 2001 15:24:51 PDT X-Received-Date: Wed, 26 Sep 2001 15:21:26 PDT (newsmaster1.prod.itd.earthlink.net) Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!news-ge.switch.ch!enews.sgi.com!newshub2.rdc1.sfba.home.com!news.home.com!newsfeed.direct.ca!look.ca!newsfeed1.earthlink.net!newsfeed.earthlink.net!newsmaster1.prod.itd.earthlink.net!newsread1.prod.itd.earthlink.net.POSTED!not-for-mail Xref: chonsp.franklin.ch alt.folklore.computers:91302 Don Quixote writes: > > > CUT and DFT? in *very* broad terms, the first is dumb, the second > smart. note while the 3277 CUT was a "dumb" terminal ... in that no microcode could be downloaded ... it was much smarter than the 3278. A lot of the head & keybarod function that was implemented in the 3277 was moved back into the 3274 controller for the 3278. we had done both a keyboard mod for fast cursor (actually control the repeat latency, plus the repeat rate for all keys). by appropriate selection of resister you wired inside the keyboard, you selected the rate that suited you. I had a keyboard set to the very short delay and fastest possible repeat. It did have the shortcoming that it was faster than the screen refresh rate ... so there was the effect of cursor "coasting" ... you held down a cursor motion key and then had to get use to when to let up on the key so that it would eventually stop at the desired location. the other modification was the addition of a keyboard FIFO that went into the display head ... you unplugged the keyboaard from the display, plugged in the keyboard FIFO box and then plugged in the keyboard into the FIFO box. the problem was that while 3270 could operate at speeds of kbytes, tens of kbytes ... they were actually half duplex devices and had a very unfortunate characteristic that if a screen update (from the system, as opposed to simple keystroke copy/record) occured just as key was being depressed ... the keyboard lost the keystroke and "locked". You then had to hit the keyboard reset button to get it back. for people used to full-duplex and nominal typing rate ... the keyboard locking was a frequent and unpleasant human factor characteristic. one justification that was given for this characteristic was that 3270s weren't designed for interactive computing ... they were designed for data entry, and data entry people didn't operate in full-duplex mode (there was almost no scenerio where screen would need updating by the system while data entry was going on). -- Anne & Lynn Wheeler | lynn@garlic.com - http://www.garlic.com/~lynn/ ###### From: jcmorris@mitre.org (Joe Morris) Newsgroups: alt.folklore.computers Subject: Re: 3270 protocol Date: 27 Sep 2001 12:56:44 GMT Organization: The MITRE Corporation Lines: 26 Message-ID: <9ov7mc$lli$1@top.mitre.org> References: <5874rtc9cnovjjjet9t6jt7pfn95n0dasf@news.fishead.meowkitty.com> Reply-To: jcmorris@mitre.org NNTP-Posting-Host: jmorris-pc.mitre.org X-Trace: top.mitre.org 1001595404 22194 128.29.251.13 (27 Sep 2001 12:56:44 GMT) X-Complaints-To: usenet@news.mitre.org NNTP-Posting-Date: 27 Sep 2001 12:56:44 GMT X-Newsreader: NN version 6.5.6 (NOV) Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!news-ge.switch.ch!newsfeed00.sul.t-online.de!newsfeed01.sul.t-online.de!t-online.de!fu-berlin.de!arclight.uoregon.edu!news.tufts.edu!blanket.mitre.org!news.mitre.org!jcmorris Xref: chonsp.franklin.ch alt.folklore.computers:91275 Anne & Lynn Wheeler writes: [half-duplex 3270 keyboard traffic could collide with an screen update] >one justification that was given for this characteristic was that >3270s weren't designed for interactive computing ... they were >designed for data entry, and data entry people didn't operate in >full-duplex mode (there was almost no scenerio where screen would need >updating by the system while data entry was going on). Sigh. The 3270 product line had a lot of very nice characteristics, but it was quite clear from the beginning that interactive computing was never going to be one of its strong points. IBM deserves kudos for designing a good terminal for business data-entry and -retrieval applications, but a rasberry for failing to understand the need for interactive devices. And before someone else points it out, I'll agree that the channel architecture on the S/360 made it for all intents and purposes impossible to have a full-duplex terminal session. (As I've noted in prior postings to a.f.c, IMHO it was the Yale Series/1 IUP that saved IBM's mainframe business in academic and research shops, by allowing generic, inexpensive ASCII dialup terminals to appear to the mainframe as locally-attached 3270s.) Joe Morris ###### From: ab528@FreeNet.Carleton.CA (Heinz W. Wiggeshoff) Newsgroups: alt.folklore.computers Subject: Re: 3270 protocol Date: 27 Sep 2001 16:57:10 GMT Organization: The National Capital FreeNet Lines: 14 Message-ID: <9ovlp6$l00$1@freenet9.carleton.ca> References: <5874rtc9cnovjjjet9t6jt7pfn95n0dasf@news.fishead.meowkitty.com> <9ov7mc$lli$1@top.mitre.org> Reply-To: ab528@FreeNet.Carleton.CA (Heinz W. Wiggeshoff) NNTP-Posting-Host: freenet10 X-Trace: freenet9.carleton.ca 1001609830 21504 134.117.136.30 (27 Sep 2001 16:57:10 GMT) X-Complaints-To: complaints@ncf.ca NNTP-Posting-Date: 27 Sep 2001 16:57:10 GMT X-Given-Sender: ab528@freenet10.carleton.ca (Heinz W. Wiggeshoff) Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!news-ge.switch.ch!newsfeeds.belnet.be!news.belnet.be!feed2.onemain.com!feed1.onemain.com!feeder.qis.net!nntp1.roc.gblx.net!nntp.gblx.net!nntp.gblx.net!xcski.com!freenet-news!FreeNet.Carleton.CA!ab528 Xref: chonsp.franklin.ch alt.folklore.computers:91329 Joe Morris (jcmorris@mitre.org) writes: > > Sigh. The 3270 product line had a lot of very nice characteristics, > but it was quite clear from the beginning that interactive computing was > never going to be one of its strong points. IBM deserves kudos for > designing a good terminal for business data-entry and -retrieval > applications, but a rasberry for failing to understand the need for > interactive devices. ... Pure slander. I spent most of the 69-70 school year at a 2741 terminal running APL\360, first on Queen's U.'s model 50, then later at I.P. Sharpe on their model . B-) ###### From: jcmorris@mitre.org (Joe Morris) Newsgroups: alt.folklore.computers Subject: Re: 3270 protocol Date: 28 Sep 2001 13:01:22 GMT Organization: The MITRE Corporation Lines: 36 Message-ID: <9p1sb2$ipe$1@top.mitre.org> References: <5874rtc9cnovjjjet9t6jt7pfn95n0dasf@news.fishead.meowkitty.com> <9ov7mc$lli$1@top.mitre.org> <9ovlp6$l00$1@freenet9.carleton.ca> Reply-To: jcmorris@mitre.org NNTP-Posting-Host: jmorris-pc.mitre.org X-Trace: top.mitre.org 1001682082 19246 128.29.251.13 (28 Sep 2001 13:01:22 GMT) X-Complaints-To: usenet@news.mitre.org NNTP-Posting-Date: 28 Sep 2001 13:01:22 GMT X-Newsreader: NN version 6.5.6 (NOV) Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!news-ge.switch.ch!news.imp.ch!news.imp.ch!news-spur1.maxwell.syr.edu!news.maxwell.syr.edu!arclight.uoregon.edu!news.tufts.edu!blanket.mitre.org!news.mitre.org!jcmorris Xref: chonsp.franklin.ch alt.folklore.computers:91262 ab528@FreeNet.Carleton.CA (Heinz W. Wiggeshoff) writes: >Joe Morris (jcmorris@mitre.org) writes: >> >> Sigh. The 3270 product line had a lot of very nice characteristics, >> but it was quite clear from the beginning that interactive computing was >> never going to be one of its strong points. IBM deserves kudos for >> designing a good terminal for business data-entry and -retrieval >> applications, but a rasberry for failing to understand the need for >> interactive devices. >... > Pure slander. I spent most of the 69-70 school year at a 2741 terminal > running APL\360, first on Queen's U.'s model 50, then later at I.P. > Sharpe on their model . The issue isn't the ability to communicate with the mainframe; I wound up writing my master's thesis using ATS through a 2741 terminal. (And I had fun with that, since it was the first computer-produced thesis at the university.) For many applications it worked quite well. And for many, many years I was chief sysprog and later manager of IBM mainframe systems, communicating with them through a 3270 interface. Again, where the user's requirements could be met with a half-duplex interface, the facilities were supurb. But the IBM systems, both by their channel architecture and by the field-mode design of the 3270, were still crippled by their inability to support anything more than what amounted to query/response user interface. There were band-aids for some interfaces -- such as the extra-cost "reverse break" feature (RPQ?) for the 2741 that allowed the CPU to force it into receive mode -- but the prime failure was the lack of full-duplex capabilities, and the apparent lack of understanding by IBM's public faces of why anyone would want anything more than half-duplex. Joe Morris ###### From: ab528@FreeNet.Carleton.CA (Heinz W. Wiggeshoff) Newsgroups: alt.folklore.computers Subject: Re: 3270 protocol Date: 28 Sep 2001 13:52:24 GMT Organization: The National Capital FreeNet Lines: 20 Message-ID: <9p1vao$5e5$1@freenet9.carleton.ca> References: <5874rtc9cnovjjjet9t6jt7pfn95n0dasf@news.fishead.meowkitty.com> <9ov7mc$lli$1@top.mitre.org> <9ovlp6$l00$1@freenet9.carleton.ca> <9p1sb2$ipe$1@top.mitre.org> Reply-To: ab528@FreeNet.Carleton.CA (Heinz W. Wiggeshoff) NNTP-Posting-Host: freenet10 X-Trace: freenet9.carleton.ca 1001685144 5573 134.117.136.30 (28 Sep 2001 13:52:24 GMT) X-Complaints-To: complaints@ncf.ca NNTP-Posting-Date: 28 Sep 2001 13:52:24 GMT X-Given-Sender: ab528@freenet10.carleton.ca (Heinz W. Wiggeshoff) Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!news-ge.switch.ch!news-fra1.dfn.de!news0.de.colt.net!colt.net!dispose.news.demon.net!demon!feed2.news.rcn.net!rcn!news.voicenet.com!news-peer.gip.net!news.gsl.net!gip.net!nntp1.roc.gblx.net!nntp.gblx.net!nntp.gblx.net!xcski.com!freenet-news!FreeNet.Carleton.CA!ab528 Xref: chonsp.franklin.ch alt.folklore.computers:91327 Joe Morris (jcmorris@mitre.org) writes: > > The issue isn't the ability to communicate with the mainframe; I wound > up writing my master's thesis using ATS through a 2741 terminal. (And > I had fun with that, since it was the first computer-produced thesis at > the university.) For many applications it worked quite well. That would be Administrative Terminal System/360, right? I first ran across that product at a service bureau which didn't really push it as an offering. Another outfit, the late Alphatext did so, to the extent that they renamed it to MITS after major extentions were added. Believe it or not, it wasn't until the early 90's (when I worked there) that a stake was finally driven through MITS's heart. Shortly thereafter I rescued the last MITS terminal - a Trendata 4000, which dominates my livingroom yet. Speaking of the TD4000, in the latter 70's I had one in my office to communicate to a 370-based service bureau in both ordinary text (ASCII) and VS/APL mode (when I switched the daisy wheel). ###### Sender: lynn@LYNNPC Newsgroups: alt.folklore.computers Subject: Re: 3270 protocol References: <5874rtc9cnovjjjet9t6jt7pfn95n0dasf@news.fishead.meowkitty.com> <9ov7mc$lli$1@top.mitre.org> <9ovlp6$l00$1@freenet9.carleton.ca> <9p1sb2$ipe$1@top.mitre.org> <9p1vao$5e5$1@freenet9.carleton.ca> Reply-To: Anne & Lynn Wheeler From: Anne & Lynn Wheeler Message-ID: Organization: Wheeler&Wheeler Lines: 15 User-Agent: Gnus/5.0808 (Gnus v5.8.8) Emacs/20.7 MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii Date: Fri, 28 Sep 2001 14:18:05 GMT NNTP-Posting-Host: 199.174.224.135 X-Complaints-To: abuse@earthlink.net X-Trace: newsread2.prod.itd.earthlink.net 1001686685 199.174.224.135 (Fri, 28 Sep 2001 07:18:05 PDT) NNTP-Posting-Date: Fri, 28 Sep 2001 07:18:05 PDT X-Received-Date: Fri, 28 Sep 2001 07:14:40 PDT (newsmaster1.prod.itd.earthlink.net) Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!news-ge.switch.ch!news.maxwell.syr.edu!netnews.com!xfer02.netnews.com!newsfeed2.earthlink.net!newsfeed.earthlink.net!newsmaster1.prod.itd.earthlink.net!newsread2.prod.itd.earthlink.net.POSTED!not-for-mail Xref: chonsp.franklin.ch alt.folklore.computers:91317 ab528@FreeNet.Carleton.CA (Heinz W. Wiggeshoff) writes: > > Speaking of the TD4000, in the latter 70's I had one in my office to > communicate to a 370-based service bureau in both ordinary text (ASCII) > and VS/APL mode (when I switched the daisy wheel). actually we had translate tables for both ASCII apl as well as ebcdic apl (there was actually tables for two types of 2741, plus tty, with apl translate tables for all three). random refs: http://www.garlic.com/~lynn/2000f.html#6 -- Anne & Lynn Wheeler | lynn@garlic.com - http://www.garlic.com/~lynn/ ###### From: "Charlie Gibbs" Newsgroups: alt.folklore.computers Subject: Re: 3270 protocol Date: 01 Oct 01 12:06:35 -0800 Organization: http://extra.newsguy.com Lines: 21 Message-ID: <898.674T1787T7265120@nowhere.in.particular> References: <9ov7mc$lli$1@top.mitre.org> <9ovlp6$l00$1@freenet9.carleton.ca> <9p1sb2$ipe$1@top.mitre.org> <0NQt7.30658$x84.8655881@ruti.visi.com> NNTP-Posting-Host: p-623.newsdawg.com X-Newsreader: THOR 2.5a (Amiga;TCP/IP) Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!news-ge.switch.ch!newsfeed00.sul.t-online.de!t-online.de!news-spur1.maxwell.syr.edu!news.maxwell.syr.edu!pln-e!spln!dex!extra.newsguy.com!newsp.newsguy.com!news3 Xref: chonsp.franklin.ch alt.folklore.computers:91690 In article lynn@garlic.com (Anne & Lynn Wheeler) writes: >the original comment was that we had a list of complaints about 3270 >(like if you were typing at the wrong moment when the system updated >the screen), the keyboard would lock and you would have to hit reset >(and anything you happened to type in the period was lost). the >response was that 3270 was designed for data entry ... not interactive >computing. At least you had the sense to distinguish between the two. According to the marketroids at Sperry->Unisys, anything with a tube was automatically deemed "interactive" - and their tubes were in many ways similar to the 3270 (block-mode, polled, synchronous, half-duplex, keyboard locked from the time you hit the TRANSMIT key to the time the computer responded). -- cgibbs@nowhere.in.particular (Charlie Gibbs) I'm switching ISPs - watch this space. ###### Newsgroups: alt.folklore.computers Subject: Re: 3270 protocol References: <5874rtc9cnovjjjet9t6jt7pfn95n0dasf@news.fishead.meowkitty.com> <9ov7mc$lli$1@top.mitre.org> <9ovlp6$l00$1@freenet9.carleton.ca> <9p1sb2$ipe$1@top.mitre.org> X-Newsreader: NN version 6.5.6 (NOV) From: jot@visi.com (J. Otto Tennant) Lines: 78 Message-ID: <0NQt7.30658$x84.8655881@ruti.visi.com> Date: Mon, 01 Oct 2001 02:45:16 GMT NNTP-Posting-Host: 209.98.98.8 X-Complaints-To: abuse@visi.com X-Trace: ruti.visi.com 1001904316 209.98.98.8 (Sun, 30 Sep 2001 21:45:16 CDT) NNTP-Posting-Date: Sun, 30 Sep 2001 21:45:16 CDT Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!news-ge.switch.ch!news.maxwell.syr.edu!upp1.onvoy!onvoy.com!zeus.visi.com!news-out.visi.com!ruti.visi.com!not-for-mail Xref: chonsp.franklin.ch alt.folklore.computers:91661 jcmorris@mitre.org (Joe Morris) writes: >ab528@FreeNet.Carleton.CA (Heinz W. Wiggeshoff) writes: >>Joe Morris (jcmorris@mitre.org) writes: [... I delete lots of this exchange ...] >But the IBM systems, both by their channel architecture and by the >field-mode design of the 3270, were still crippled by their inability >to support anything more than what amounted to query/response user >interface. There were band-aids for some interfaces -- such as the >extra-cost "reverse break" feature (RPQ?) for the 2741 that allowed >the CPU to force it into receive mode -- but the prime failure was >the lack of full-duplex capabilities, and the apparent lack of >understanding by IBM's public faces of why anyone would want anything >more than half-duplex. "Crippled?" Doesn't that depend on the design of the application? I grant that "vi" was/is pretty spiffy on an ADM-3a, and, with "termcap", could support an almost infinite variety of dumb terminals. (On the other hand, I don't know of any termcap/terminfo entry which supports the full capability of a VT420. More years ago than I care to think about, I walked into my office one day to find a box from DEC; it contained a VT420 and a note from the chap I had been talking to at DEC saying, essentially, "Good Luck." I did update the terminfo entry (which probably never got outside a very local community.)) But there are _excellent_ text editors easily working within the half-duplex nature of the 3270. EDGAR (for VM/370) had to have been available in the mid 70-s, followed by XEDIT. When one combined XEDIT with REXX, it left "vi" in the dust. I definitely remember a time in (well, I don't really remember the year) when I was dragged kicking and screaming from my 327x (it was actually a Lee Data Systems clone) to a Sun workstation. I'm sure that you are willing to admit that, capability-wise, "vi" isn't even in the same class as XEDIT naked, much less than when garbed with REXX. After about six weeks, I had reprogrammed my spinal cord, to deal with "vi". ^H^J^K^L became, however unintuitive, , , , . At about that time, I had become recently (and shortly) married. After what, despite later awkward and expensive experiences, was at the time a pleasant night, I learned what "vi" could do if one accidentally allowed "J" to auto-repeat. I no longer remember how deep XEDIT's "undo" was; heck, I'm not even certain that it had an "undo." You only need "undo" if your full duplex terminal obeys you too quickly. And "vi" had only one level of "undo", which wasn't of too much use if one drifted off, either in memory or in sleep, accidentally touching the "J" auto-repeat key. So I switched to EMACS. I had to reprogram myself again, but I had high hopes of many more such evenings. (In the event, I should have stuck with 'ed'.) The Amdahl Computer Corporation, in the very early 80's, produced a version of Unix, called UTS, under VM/370. They offered an editor (I forget what it was called (possibly "ni"?) which adapted "vi" to the half duplex nature of a 3270. It worked swell. I recognize that increasingly cheap hardware makes concerns such as mine irrelevant. Right now, I'm using "vi" to compose this message, and every single character I type must cause a flurry of context switches on my own machine, a flurry of network traffic, and a flurry of context switches on my ISP. I'm an old guy, and, to this day, I have not really understood how cheap all of this really is today. FWIW, which is not much. >Joe Morris -- J.Otto Tennant jotto@pobox.com Forsan et haec olim meminisse juvabit. Charter Member of the Vast Right Wing Conspiracy ###### Sender: lynn@LYNNPC Newsgroups: alt.folklore.computers Subject: Re: 3270 protocol References: <5874rtc9cnovjjjet9t6jt7pfn95n0dasf@news.fishead.meowkitty.com> <9ov7mc$lli$1@top.mitre.org> <9ovlp6$l00$1@freenet9.carleton.ca> <9p1sb2$ipe$1@top.mitre.org> <0NQt7.30658$x84.8655881@ruti.visi.com> Reply-To: Anne & Lynn Wheeler From: Anne & Lynn Wheeler Message-ID: Organization: Wheeler&Wheeler Lines: 32 User-Agent: Gnus/5.0808 (Gnus v5.8.8) Emacs/20.7 MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii Date: Mon, 01 Oct 2001 12:35:42 GMT NNTP-Posting-Host: 209.245.11.162 X-Complaints-To: abuse@earthlink.net X-Trace: newsread1.prod.itd.earthlink.net 1001939742 209.245.11.162 (Mon, 01 Oct 2001 05:35:42 PDT) NNTP-Posting-Date: Mon, 01 Oct 2001 05:35:42 PDT X-Received-Date: Mon, 01 Oct 2001 05:32:14 PDT (newsmaster1.prod.itd.earthlink.net) Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!news-ge.switch.ch!enews.sgi.com!paloalto-snf1.gtei.net!paloalto-snh1.gtei.net!lsanca1-snf1!news.gtei.net!newsfeed2.earthlink.net!newsfeed.earthlink.net!newsmaster1.prod.itd.earthlink.net!newsread1.prod.itd.earthlink.net.POSTED!not-for-mail Xref: chonsp.franklin.ch alt.folklore.computers:91658 jot@visi.com (J. Otto Tennant) writes: > > "Crippled?" Doesn't that depend on the design of the application? > the original comment was that we had a list of complaints about 3270 (like if you were typing at the wrong moment when the system updated the screen), the keyboard would lock and you would have to hit reset (and anything you happened to type in the period was lost). the response was that 3270 was designed for data entry ... not interactive computing. That doesn't imply that it wasn't better than many things that had been supposedly designed for interactive computing (sometimes raw speed can count) ... it just that the 3270 could have been better (like the FIFO keyboard buffer in 3277 to overcome the keyboard lockup problem). Not only was it designed for data entry, that who would possible want to be typing when the system had something to do. Another was trying to reposition the cursor on the screen, the speed was ... chug, ... chug, ... chug, ... So does anybody remember the up/down, scroll-up/scroll-down, page-up/page-down editor wars with edgar/red/xedit/etc (from the early to mid 70s)???? aka in effect, "were the commands done with respect to the program or the human"? edgar was with respect to the program, "scroll-up", in effect moved the document "up" (viewing moved down). misc refs: http://www.garlic.com/~lynn/2001i.html#1 http://www.garlic.com/~lynn/2001k.html#33 -- Anne & Lynn Wheeler | lynn@garlic.com - http://www.garlic.com/~lynn/ ###### From: "GerardS" Newsgroups: alt.folklore.computers Subject: Re: 3270 protocol Date: Mon, 1 Oct 2001 16:17:01 -0500 Organization: Posted via Supernews, http://www.supernews.com Message-ID: References: <9ov7mc$lli$1@top.mitre.org> <9ovlp6$l00$1@freenet9.carleton.ca> <9p1sb2$ipe$1@top.mitre.org> <0NQt7.30658$x84.8655881@ruti.visi.com> <898.674T1787T7265120@nowhere.in.particular> X-Priority: 3 X-MSMail-Priority: Normal X-Newsreader: Microsoft Outlook Express 5.50.4807.1700 X-MimeOLE: Produced By Microsoft MimeOLE V5.50.4807.1700 X-Complaints-To: newsabuse@supernews.com Lines: 30 Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!news-ge.switch.ch!news.tele.dk!small.news.tele.dk!171.64.14.106!newsfeed.stanford.edu!sn-xit-01!sn-post-01!supernews.com!corp.supernews.com!not-for-mail Xref: chonsp.franklin.ch alt.folklore.computers:91689 | Charlie Gibbs wrote: | (Anne & Lynn Wheeler) writes: |>the original comment was that we had a list of complaints about 3270 |>(like if you were typing at the wrong moment when the system updated |>the screen), the keyboard would lock and you would have to hit reset |>(and anything you happened to type in the period was lost). the |>response was that 3270 was designed for data entry ... not interactive |>computing. | | At least you had the sense to distinguish between the two. | According to the marketroids at Sperry->Unisys, anything with | a tube was automatically deemed "interactive" - and their tubes | were in many ways similar to the 3270 (block-mode, polled, | synchronous, half-duplex, keyboard locked from the time you | hit the TRANSMIT key to the time the computer responded). On the VM/CMS systems, as soon as the user pressed enter (or some other function key for example), the VM system immediately sent back an unlock, thereby minimizing the amount of time the keyboard was locked. On channel-attached 3270s, this was almost instantaneous, so that I could hold down the ENTER key (making use of the typomatic key), and the CMS (VM) system would be able to process all the ENTERs. I would do this while in XEDIT mode, or plain CMS mode, and the system would be able to keep up with the multiple (if not infinite) ENTERs). One could also XEDIT a very large file and just keep the PF8 (down) depressed and watch XEDIT scroll down the file. Gerard S. ###### Sender: lynn@LYNNPC Newsgroups: alt.folklore.computers Subject: Re: 3270 protocol References: <9ov7mc$lli$1@top.mitre.org> <9ovlp6$l00$1@freenet9.carleton.ca> <9p1sb2$ipe$1@top.mitre.org> <0NQt7.30658$x84.8655881@ruti.visi.com> <898.674T1787T7265120@nowhere.in.particular> Reply-To: Anne & Lynn Wheeler From: Anne & Lynn Wheeler Message-ID: Organization: Wheeler&Wheeler Lines: 70 User-Agent: Gnus/5.0808 (Gnus v5.8.8) Emacs/20.7 MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii Date: Tue, 02 Oct 2001 01:10:15 GMT NNTP-Posting-Host: 209.245.9.114 X-Complaints-To: abuse@earthlink.net X-Trace: newsread2.prod.itd.earthlink.net 1001985015 209.245.9.114 (Mon, 01 Oct 2001 18:10:15 PDT) NNTP-Posting-Date: Mon, 01 Oct 2001 18:10:15 PDT X-Received-Date: Mon, 01 Oct 2001 18:06:44 PDT (newsmaster1.prod.itd.earthlink.net) Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!news-ge.switch.ch!newsfeed00.sul.t-online.de!t-online.de!news-spur1.maxwell.syr.edu!news.maxwell.syr.edu!cpk-news-hub1.bbnplanet.com!lsanca1-snf1!news.gtei.net!newsfeed2.earthlink.net!newsfeed.earthlink.net!newsmaster1.prod.itd.earthlink.net!newsread2.prod.itd.earthlink.net.POSTED!not-for-mail Xref: chonsp.franklin.ch alt.folklore.computers:91771 "GerardS" writes: > > On the VM/CMS systems, as soon as the user pressed enter (or some other > function key for example), the VM system immediately sent back an > unlock, thereby minimizing the amount of time the keyboard was locked. > On channel-attached 3270s, this was almost instantaneous, so that I > could hold down the ENTER key (making use of the typomatic key), and > the CMS (VM) system would be able to process all the ENTERs. I would > do this while in XEDIT mode, or plain CMS mode, and the system would be > able to keep up with the multiple (if not infinite) ENTERs). One could > also XEDIT a very large file and just keep the PF8 (down) depressed and > watch XEDIT scroll down the file. STL was bursting at the seams ... to they needed to relocate about 300 people from the IMS group to remote, leased building about half-way between STL and the main plant site. Rather than make them endure the horrors of remote 3270s, they were going to get local 3270s with (HYPERchannel) as channel extenders. There was some issues getting the support to run over a dedicated T1 line since it was only 1.5mbytes/sec ... even tho the aggregate data rate didn't tend to exceed the line rate, the channel attach 3270s had a higher burst rate and so there were some speed-matching and pacing issues. One of the unexpected outcomes of the whole thing was that the mainframes back in STL started running 10-15% better thruput. Standard operating procedure up until then had been to spread the 3274 control units across all available channels on the same channels with disks controllers (just because the disks had been spread across available channels for load-balancing leaving a lot of unit addresses available on each channel). The HYPERChannel support that I wrote used a single HYPERChannel A220 on the mainframe end to drive the T1 and the 3274 controllers (for 300+ 3270s) and misc. other channel controlers at the remote site. The problem was that 3274 controller could burst data transfers at channel speed to the 3270s ... it had significant control overhead that resulting in significant channel busy. It turned out that the bad 3274 channel busy overhead was causing signficiant interface for disk activity. The HYPERChannel A220 controller hung on the mainframe channel had significant better (tremendously less) channel busy overhead compared to the 3274s. Remoting all the 3274s out to HYPERChannl A51x channel simulators (and performing the actual channel operations on an A220) turned out to not degrade the thruput perceived by all of the 3270s AND signficiantly reduced overall channel busy across all the mainframe channels significantly improving disk thruput resulting in a 10-15% overall system thruput increase (hows that for a run on sentence). The above helped overall system thruput and helped mask the poor channel busy characteristics of channel attach 3274 controllers. After that, there started being a lot more recommendations about 3274 controller placement to minimize interference with disk thruput. SNA 3274s couldn't really be considered an option since the resulting 3270 performance characteristics were really terrible for interactive computing. Those controller characteristics were orthogoanl to keyboard lockup issues ... I kept my trusty modified 3277 (fifo keystroke buffer and modified repeat key operation) up until almost 1990 (as backup, long after I had 3270 emulation on PC and could program around the short comings). misc. hyperchannel & hsdt refs: http://www.garlic.com/~lynn/subtopic.html#hsdt -- Anne & Lynn Wheeler | lynn@garlic.com - http://www.garlic.com/~lynn/ ###### Message-ID: <3BBB9D1B.5E11D18B@yahoo.com> From: Peter Flass X-Mailer: Mozilla 4.75 [en] (Win98; U) X-Accept-Language: en MIME-Version: 1.0 Newsgroups: alt.folklore.computers Subject: Re: 3270 protocol References: Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit Lines: 10 Date: Wed, 03 Oct 2001 23:26:46 GMT NNTP-Posting-Host: 24.169.5.42 X-Complaints-To: abuse@nycap.rr.com X-Trace: typhoon.nyroc.rr.com 1002151606 24.169.5.42 (Wed, 03 Oct 2001 19:26:46 EDT) NNTP-Posting-Date: Wed, 03 Oct 2001 19:26:46 EDT Organization: Time Warner Road Runner - Albany NY Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!news-ge.switch.ch!news.maxwell.syr.edu!cpk-news-hub1.bbnplanet.com!news.gtei.net!newsfeed1.cidera.com!Cidera!cyclone.nyroc.rr.com!cyclone-out.nyroc.rr.com!typhoon.nyroc.rr.com.POSTED!not-for-mail Xref: chonsp.franklin.ch alt.folklore.computers:91851 I just joined this newsgroup, so I may have missed the answer: CUT = Control Unit Terminal Anne & Lynn Wheeler wrote: > ... I have some vague recollection > that one of the terms was CUT and may refer to the 3272/3277 protocol > ... but I can't remember the other term (and/or even sure CUT is one > of the terms). > > does anybody remember the two terms? ###### Sender: lynn@LYNNPC Newsgroups: alt.folklore.computers Subject: Re: 3270 protocol Reply-To: Anne & Lynn Wheeler From: Anne & Lynn Wheeler Message-ID: Organization: Wheeler&Wheeler Lines: 61 User-Agent: Gnus/5.0808 (Gnus v5.8.8) Emacs/20.7 MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii Date: Mon, 05 Nov 2001 04:11:40 GMT NNTP-Posting-Host: 64.156.37.25 X-Complaints-To: abuse@earthlink.net X-Trace: newsread2.prod.itd.earthlink.net 1004933500 64.156.37.25 (Sun, 04 Nov 2001 20:11:40 PST) NNTP-Posting-Date: Sun, 04 Nov 2001 20:11:40 PST X-Received-Date: Sun, 04 Nov 2001 20:07:35 PST (newsmaster1.prod.itd.earthlink.net) Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!news-ge.switch.ch!newsfeed00.sul.t-online.de!t-online.de!news.tele.dk!small.news.tele.dk!199.60.229.5!newsfeed.direct.ca!look.ca!newsfeed1.earthlink.net!newsfeed.earthlink.net!newsmaster1.prod.itd.earthlink.net!newsread2.prod.itd.earthlink.net.POSTED!not-for-mail Xref: chonsp.franklin.ch alt.folklore.computers:93907 finally stumbled across a reference. ANR was the 3272/3277 protocol ... where most of the smarts were in the display head. DCA was the 3274/3278/3279/etc protocol ... where most of the smarts had been moved back in the controller. The problem when they were coming up with the 3270PC (and various other pc 3270 coax) was that with things like file transfers (and/or large data movements), ANR had three times the thruput of DCA. Part of the problem was that since all the smarts had been moved back into controller for DCA ... there was huge amounts of protocol chatter with DCA (for instance controller had to constantly poll the keyboard for each key up/down operation). There was then an effort to revise DCA in order to try and improve things to be compareable to ANR thruput. random 3270 refs: http://www.garlic.com/~lynn/94.html#23 CP spooling & programming technology http://www.garlic.com/~lynn/96.html#41 IBM 4361 CPU technology http://www.garlic.com/~lynn/98.html#49 Edsger Dijkstra: the blackest week of his professional life http://www.garlic.com/~lynn/98.html#56 Earliest memories of "Adventure" & "Trek" http://www.garlic.com/~lynn/99.html#26 IBM S/360 http://www.garlic.com/~lynn/99.html#28 IBM S/360 http://www.garlic.com/~lynn/99.html#60 Living legends http://www.garlic.com/~lynn/99.html#61 Living legends http://www.garlic.com/~lynn/99.html#69 System/1 ? http://www.garlic.com/~lynn/99.html#108 IBM 9020 computers used by FAA (was Re: EPO stories (was: HELP IT'S HOT!!!!!)) http://www.garlic.com/~lynn/2000.html#6 Computer of the century http://www.garlic.com/~lynn/2000.html#90 Ux's good points. http://www.garlic.com/~lynn/2000b.html#49 VM (not VMS or Virtual Machine, the IBM sort) http://www.garlic.com/~lynn/2000b.html#50 VM (not VMS or Virtual Machine, the IBM sort) http://www.garlic.com/~lynn/2000c.html#63 Does the word "mainframe" still have a meaning? http://www.garlic.com/~lynn/2000c.html#65 Does the word "mainframe" still have a meaning? http://www.garlic.com/~lynn/2000c.html#76 Is a VAX a mainframe? http://www.garlic.com/~lynn/2000d.html#12 4341 was "Is a VAX a mainframe?" http://www.garlic.com/~lynn/2000e.html#32 Tektronics Storage Tube Terminals http://www.garlic.com/~lynn/2000e.html#53 Why not an IBM zSeries workstation? http://www.garlic.com/~lynn/2000e.html#56 Why not an IBM zSeries workstation? http://www.garlic.com/~lynn/2000g.html#17 IBM's mess (was: Re: What the hell is an MSX?) http://www.garlic.com/~lynn/2001b.html#49 PC Keyboard Relics http://www.garlic.com/~lynn/2001b.html#83 Z/90, S/390, 370/ESA (slightly off topic) http://www.garlic.com/~lynn/2001f.html#8 Theo Alkema http://www.garlic.com/~lynn/2001f.html#9 Theo Alkema http://www.garlic.com/~lynn/2001f.html#28 IBM's "VM for the PC" c.1984?? http://www.garlic.com/~lynn/2001f.html#57 any 70's era supercomputers that ran as slow as today's supercomputers? http://www.garlic.com/~lynn/2001i.html#18 History of Microsoft Word (and wordprocessing in general) http://www.garlic.com/~lynn/2001k.html#30 3270 protocol http://www.garlic.com/~lynn/2001k.html#33 3270 protocol http://www.garlic.com/~lynn/2001k.html#38 3270 protocol http://www.garlic.com/~lynn/2001k.html#44 3270 protocol http://www.garlic.com/~lynn/2001k.html#46 3270 protocol http://www.garlic.com/~lynn/2001l.html#14 mainframe question http://www.garlic.com/~lynn/2001l.html#62 ASR33/35 Controls http://www.garlic.com/~lynn/2001m.html#1 ASR33/35 Controls -- Anne & Lynn Wheeler | lynn@garlic.com - http://www.garlic.com/~lynn/ ###### Sender: lynn@LYNNPC Newsgroups: alt.folklore.computers Subject: Re: 3270 protocol References: Reply-To: Anne & Lynn Wheeler From: Anne & Lynn Wheeler Message-ID: Organization: Wheeler&Wheeler Lines: 100 User-Agent: Gnus/5.0808 (Gnus v5.8.8) Emacs/20.7 MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii Date: Wed, 07 Nov 2001 05:18:54 GMT NNTP-Posting-Host: 209.245.9.102 X-Complaints-To: abuse@earthlink.net X-Trace: newsread2.prod.itd.earthlink.net 1005110334 209.245.9.102 (Tue, 06 Nov 2001 21:18:54 PST) NNTP-Posting-Date: Tue, 06 Nov 2001 21:18:54 PST X-Received-Date: Tue, 06 Nov 2001 21:14:48 PST (newsmaster1.prod.itd.earthlink.net) Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!news-ge.switch.ch!newsfeed00.sul.t-online.de!t-online.de!news-spur1.maxwell.syr.edu!news.maxwell.syr.edu!feed2.news.rcn.net!rcn!newsfeed1.earthlink.net!newsfeed.earthlink.net!newsmaster1.prod.itd.earthlink.net!newsread2.prod.itd.earthlink.net.POSTED!not-for-mail Xref: chonsp.franklin.ch alt.folklore.computers:94001 Anne & Lynn Wheeler writes: > The problem when they were coming up with the 3270PC (and various > other pc 3270 coax) was that with things like file transfers (and/or > large data movements), ANR had three times the thruput of DCA. Part > of the problem was that since all the smarts had been moved back into > controller for DCA ... there was huge amounts of protocol chatter with > DCA (for instance controller had to constantly poll the keyboard for > each key up/down operation). The attached is from a report on how 3274 controllers make it impossible to achieve a reasonable "response time" objective of an avg. of .25 seconds for trivial interactive response. Avg. trivial interactive response in the '70s and '80s was frequently measured in terms of avg. system service time; i.e. from the time the hardware interrupt was presented to the system until the system serviced the request and wrote a response. In the 3270 case, block transmission of complete screens and the 3272/3277 was pretty insensitve to partial or complete screen block transfers. The difference in transmission times between 1byte, 1000 bytes, or 2000 bytes was relatively inconsequential at several hundred kilobytes per second. Hardware terminal service time was dominated by controller processing, not bandwidth transmission. At some point during this analysis I received an email from an MVS system programmer telling me that I should stop making comments about how bad TSO response was; that he had tuned a lightly loaded MVS/TSO system and was capable of (outstanding) avg. 1 second average response for trivial interactive operations. By comparison, it wasn't too unusual for well, tuned, large, heavily loaded VM/CMS systems to avg. .25 seconds for trivial interactive response. In addition, I was able with some additional special tuning to cut that in half. For nearly identical workload that many systems were getting .25sec avg. trivial interactive response, I was able to show .11sec avg trivial interactive response (similar hardware and workload). Note that that is just the system service time, the terminal hardware service time would have to be added to get the actual end-user perceived response. Using the 3272 and 3274 hardware service numbers and adding them to typical system service times for the end-user perceived response. hardware TSO 1sec. CMS .25sec. CMS .11.sec 3272/3277 .086 1.086 .336 .196 3274/3278 .530 1.530 .78 .64 And then if you are talking about SNA attached controllers the hardware service time will start to dwarf even TSO service times. =================== append ==================== Applications developed for the 3272 will run un-modified on the 3274 but with substantially longer response times. Service time is 3-8 times that of the 3272. Quarter second response time objectives cannot be achieved using 3274 model D control units in the local environment. If the terminals are remote transmission time to the TP link will add to the system response time. Some simple data streams Data 3272 3274 Stream 3277 3278 A .086 .283 C .086 .530 If the new 3274 functions (color, highlighting and larger character formats) are used then response times are increased still further. ==================== end append ============================ A little discussion of the CERN CMS/TSO benchark: http://www.garlic.com/~lynn/98.html#28 Drive letters http://www.garlic.com/~lynn/2000f.html#61 360 Architecture, Multics, ... was (Re: X86 ultimate CISC? No.) http://www.garlic.com/~lynn/2001h.html#11 checking some myths. http://www.garlic.com/~lynn/2001i.html#30 IBM OS Timeline? Some aspects of MVS contributing to TSO performance http://www.garlic.com/~lynn/2001l.html#40 MVS History (all parts) Some HYPERChannel/3270 discussion http://www.garlic.com/~lynn/94.html#23 CP spooling & programming technology http://www.garlic.com/~lynn/2000c.html#65 Does the word "mainframe" still have a meaning? http://www.garlic.com/~lynn/2000d.html#12 4341 was "Is a VAX a mainframe?" http://www.garlic.com/~lynn/2001k.html#46 3270 protocol -- Anne & Lynn Wheeler | lynn@garlic.com - http://www.garlic.com/~lynn/