Minutes of the SIPB Meeting of 1 November 1982 The meeting was called to order at 7:40pm by JR. In attendance were: Voting members: JR, CBF, KAM, RDZ, BCN, EJB, Alan Others: DCP, RK, GMP, Ed (briefly), Gumby, JIS (briefly) The minutes of the meeting of 25 Oct. 82 were read and accepted as corrected. The minutes of the meeting of 18 Oct. 82 will be mailed out in a corrected condition at the end of the meeting. People should read them & check for accuracy, etc. TREASURER'S REPORT Treasurer's Report: October Spending: 9,160.56 (999.72 spba/spb -- only 11% !!) Last October 7,780 YTD at end of October This year 24,988 1981-82 21,474 1980-81 32,588 YTD 27,077 We are some between last year's and the year before's spending. JR passed around a "fascist graph" showing all the above numbers pictorially. motion by CBF to buy a four-color plotter for the treasurer. Dies for lack of second. APPLICATIONS Nelson, Fernandez. $70.00 ->mtg for text editing. Given. Prizant, Jim. $100.00 -> mtg for text editing. Prev had $100.00 (also for text editing). RDZ warned him about the $200.00 'limit'. Given $75.00 Bendriem, Bernard. $20.00 ->mtg for text editing. He is a visiting scientist. Given. Koskie, Sarah. $100.00 ->mtg for text editing. Prev had $100.00. Given. Johns, Eric T. $30.00 ->mtg. Prev had $100.00. Text editing of papers and law school applications. Given. Martinez, Maria. $50.00 ->mtg by Jay. Prev had $50.00. text editing. Deferred. Ismail, Nassar. $50.00 ->mtg. Prev had $100.00. For FORTRAN work. Deferred until we find out what he's doing. Kwon, Young. Was deferred last week, deferred again this week as he is not here. Ruck, Harold. unspecified ->mtg, prev had $100.00. Text editing, seems to want as much as he can get. Given up to $50.00 at discretion of office. (This should be added into next week's autoed totals). Lewis, Mark J. $50.00 ->mtg, prev had $100.00. Text editing. Given. Brahm, David. $25.00 ->ec. Text editing. Given. CHAIRMAN'S REPORT At the CCE (Committee on the Computing Environment at MIT) meeting Thursday, we decided to have meetings every two weeks (we hope). We will be discussing the next five years of computing at MIT, trying to decide on the direction(s) the effort(s) should take. We discussed the CMU plan (with IBM), and mentioned that since DEC lost out when IBM won, MIT may be able to make a deal with DEC. MIT is working on it. MIT is also talking to IBM to see if IBM wants to set up a CMU-style deal with MIT. Nothing is at all definite, but they are all worried enough to really put effort into it. Joel Moses pointed out that IPC lost $200,000 last year, and that this was 10% of the Institute's deficit. Wes was not at all pleased with this statement. Joel and perhaps others think that IPC should go into the mass-storage, moby special program and network-keeping business, and that with so many people getting their own computers, this is just the beginning of a steady downward slide for IPC. There will be more discussion on that in later meetings, I'm sure. Although Provost Low wasn't there, I brought up the SIPB's desire for money to convert the new rooms we got in the Stu Cen into terminal rooms and a home for our concentrator. They thought I should go to Low, but after the meeting Wes told me he'll look into it, but he thinks he can give us the money. We'll see. I also told them now that we'll have more terminal access we'll be asking for either more funding immediately for use this year, or a pie-slice of Multics. I explained that we are the quickest and easiest route by which they can get more computing for students. By the end of the two hour meeting, they were all conditioned to think of SIPB the moment anyone mentioned student computing. There was also some random talk about networking the entire campus or of getting a hundred VAXen and distributing them around buildings and dorms. Of course, none of this has gotten anywhere yet, and we hope to discuss it more and make progress (dare I hope it?) in later meetings. By the time we have the next meeting (2-4 weeks), I believe the SIPB will have its budget increase, pie slice, renovation and future computing environment proposals done. And with any luck, Course VI will have made some progress with their small computer hunting. wer WMY has done some research into the idea of a pie slice on Multics. Anyone who is interested in this SHOULD TALK TO WMY BEFORE DOING ANYTHING about it. OFFICE REPORT The office phones are broken. Someone should call these in either tomorrow or sooner. Someone was in today and made them worse. Some people would like to get rid of the dataset. motion by CBF to replace the dataset with a normal phone. Fails. motion by KAM to table last week's tabled motion. Passes. ("Can we table w/ extreme prejudice?"--RDZ DCP: We should table it to a committee that's not allowed to meet.). RK points out that a). the terminal concentrator IS in the office, and that b). there are more terminals in the office than we know what to do w/, and we should store them somewhere. CBF on the other hand thinks the terminal concentrator is in a very bad place, so he thinks RK's point is irrelevant. COMPUTER SERVICES motion by GMP to terminate all services. Fails. TELECOMMUNICATIONS REPORT Student Center Report Steve talked with Slyvia McDowell last week. She has already given us three typing rooms (552, 554, 556,) keys for which are in the strongbox. We can use these rooms now for storage of some of our stuff. She has also offered us two conference rooms( 555 and 557) for use as the new SIPB terminal room, in exchange for our present room. If we want, we can take just one of these rooms and keep our present room. Steve and Wendy also talked to Wes Burner. He may have the money we need to rennovate these rooms from his discretionary budget. It is important that we get an estimate from physical plant soon for the work that needs to be done. Zigurd is working on ( or has completed ) a request to PP for this work. The estimate will be requested in two parts, creation of a new terminal room, and making a second terminal room out of the equipment in our present room. This way we can figure what we want to do and what we can get money for. The cost for all work will probably be something over $20,000. Maybe nearer to $30,000. No one should deal officially with Sylvia McDowell without first talking to Wendy or Steve. In general members shouldn't have to bother the library staff too much as we now have our own keys for the three storage rooms. If a member is doing things in the library, he should identify himself to the desk worker as a SIPB person and show his membership card. We want to train these people so that they don't let randoms walk in and fool with our equipment. Gill and Naha seem to be taking care of our antenna and its connection. In general, Steve is hoping to have work finished in the student center sometime in Feburary. This requires that a lot of arrangements be made now. One thing we can do almost immediately is get the terminal concentrator running with our new terminals. This can be done in the office, with public terminals out in some IPS space. If DCP or others could make this happen soon, it could be a big help to us in convincing people to give us money. Everyone take care, Steve BCN wants to know if there is a catch. Gumby points out that Physical Plant is twice as expensive as anything else. GMP points out that we have little choice. RK points out that the Institute is paying for it. DCP gave us a terminal concentrator report. For details, see DCP. motion by CBF to buy the cable for the terminal concentrator. Everyone agrees. Eric also said he would like about $1500.00 for the Q-Bus chaos card. The motion was discussed but never voted on. The vadic arrived. After the dataset is fixed, RK will hook up the vadic. How do you sell an AJ? If someone comes in and wants to buy an AJ, take $100.00 or their offer if it is higher, & make sure you give them the right AJ (DO NOT SELL THEM THE VADIC W/ SERIAL #89798 AS IT HAS ALREADY BEEN SOLD). TERMINAL REPORT There are 8 TTR's, which BCN read to the meeting. 1). 3 August 82 (!). ADP's Z19 caps lock key is broken. 2). 5 Oct. 82. MacGregor's vadic modem is dead. This form was marked solved, but later was pulled out again as still a problem. 3). 13 Oct. 82. ZBT'S terminal also has a broken caps lock key. 4). 14 Oct. 82. New House printing terminal is dead. 5). 23 Oct. 82. East Campus LA36 is missing the right shift key. 6). 28 Oct. 82. Burton LA36 dataset works only in local; will not talk to any computer. The dataset works O.K. as a telephone. (RK thinks the person may not be hanging up the phone properly). 7). 29 Oct. 82. #PN22734 in the Student Center Library intermittently fails self-test. 8). 31 Oct 82. ZBT's terminal can't dial up computer. Alan thinks this is more boring than television. motion by RK to take Heath19's to the places w/ broken keyboards & bring the broken ones back here so that eventually we can take them to Wellesley. CBF wants to know how they got broken. This motion applies to ADP and ZBT. Passes w/o a vote. CBF thinks the stack of TTR's is getting hard to deal w/, and someone should think of a place in the office where they can be spread out. There are five keys to the terminal room in the Student Center which should be tested. The 5 vt100's we have don't have the advanced video option. CBF thinks we should get it. PUBLICATIONS REPORT WER is working on the new NETS for Process_paper. ANNUAL-REPORT REPORT R/O REPORT IAP REPORT OTHER RDZ read & totalled the autoed applications. Total appropriations: $ (without Coop) Autoed: Student: $2355.00 370: $75.00 EC: Student: $25.00 370: $-- To Meeting: Student: $395.00 370: $-- Coop: $-- OTHER OTHER DCP --randomness w/ wrong numbers for Symbolics. They have nearly the same number as foo. The meeting was adjourned at 9:00pm. Minutes taken and submitted by Katharine.  Date: 1 November 1982 19:20-EST From: Zigurd R. Mednieks Subject: text processors. To: GJC at MIT-MC cc: SIPB at MIT-MC, GILL at MIT-CCC Date: 1 November 1982 14:47-EST From: George J. Carrette The IBM PC is too expensive, it is priced so as to not compete with IBM's line of minicomputers. I would have thought this would have been obvious? And what about all the other MS-DOS machines? The numbers work out roughly the same. The point is that CP/M, floppies and crufty, cheap, 8-bit micros do not a service make. Also, by the time we install such a system (get a room, remodel it, install A/C, etc.) 8 bit micros will join 370s, RSX11, and COBOL as things we'd just prefer not to think about -- if they haven't already. That we might wind up using a 780 as a hardware base is bad enough. Could we agree not to live in the past any more than we have to? The motivation for using micros should not be that we could set out n minimal systems that do text processing minimally well. The reason micros are attractive at all when a VAX is no more expensive is that there is a large software base. Dozens of software houses will be cranking out sexy things like Visicalc and TK!Solver. Further, they are going to be writing these things for the 8086 based machines more and more, and Z80 machines less and less. Roughly the same argument militates for using Unix on whatever minicomputer we get, except that we wouuld be getting compatibility with all the people doing system hacking, compiler hacking, and big subsystems, as opposed to commercial applications. Twenex and VMS have a design heritage of DOS-like systems and will forever be saddled with design flaws that used to be features when the economics of computer systems was completely different from today. Chaining ourselves to five years of prepetuated obsolescence is the very height of folly. We could just as well use a pie slice of the IPCs big new IBM machine as buy (or beg) our very own brain damage. Don't think that Unix is the Ultimate Answer either. If MIT is to stake out the #1 spot in computer education with renewed vigor, it is going to have to involve original ideas, systems hackery, and projects like the LISPM and Nu. What the SIPB is trying to do is build a bridge to that future -- a temporary fix. That bridge should not be made of slag heaps of software that was out of date before it was even finished. On Horseback, Zig  PMF@MIT-MC 11/01/82 19:11:41 To: CBF at MIT-MC Sometime soon, but not sure when.  Date: 1 November 1982 14:47-EST From: George J. Carrette Subject: text processors. To: ZRM at MIT-MC cc: SIPB at MIT-MC, GILL at MIT-CCC The IBM PC is too expensive, it is priced so as to not compete with IBM's line of minicomputers. I would have thought this would have been obvious?  Date: 1 Nov 1982 0923-PST From: DKING Subject: Re: a couple of terminals To: CBF at MIT-MC In-Reply-To: Your message of 29-Oct-82 1348-PDT I checked with Zubkoff's company. Indeed, they sell the PROM's, not HDS. They say they haven't done the Ambassador yet. Thanks for the info. Dick -------  Date: Sunday, 31 October 1982, 21:37-EST From: Howard D. Trachtman To: cbf at MIT-MC BTW some of the messages I did send to bug-hdt as test seem to not have gotten thru. Could you please fix whatever part of this you may have caused? Thanks. --Howard  Date: Sunday, 31 October 1982, 21:36-EST From: Howard D. Trachtman To: cbf at MIT-MC, hdt at MIT-OZ NO!!! I frequently have been mailing myself messages from a lispm machine to myself in order that I can remember things to do that I don't have time to do now. Also, if I find a bug in something that I am working on, it is useful to have that information inserted for me. The facility proved by default sends bug reports to mc. That's why I added the bug-hdt. What made you srccom names > anyway. There were no errors in the file.  MSG: *MSG 1402 DISTRIB: *BBOARD EXPIRES: 11/08/82 05:36:24 Return-path: @USC-ISID,@UCL-CS,bundy@edxa Via: USC-ISID ; Monday, November 1, 1982 01:34:01-PST Via: Ucl-Cs; 1 Nov 82 09:32-GMT Via: ERCC ; Monday, 1-Nov-82 09:30:14-GMT Date: Monday, 1-Nov-82 09:30:14-GMT From: BUNDY HPS To: arpanet-bboards%mit-ai@ucl Reply-To: bundy@rutgers Subject: IJCAI-83 Programme Committee Remailed-date: 01 Nov 1982 0536-EST Remailed-from: Arpanet-BBoards-Request at MIT-ML -------- IJCAI-83 PROGRAMME - FURTHER DETAILS IJCAI-83 will take place in Karlsruhe, Germany from 8th-12th August 1983. Four copies of submitted papers should be sent to: Alan Bundy Programme Chairman IJCAI-83, Department of Artificial Intelligence, University of Edinburgh, Hope Park Square, Edinburgh, EH8 9NW, Scotland. to arrive by 1st February 1983. Maximum length, 5500 long papers, 2250 short papers. The following information should be included with submissions: author's name(s), address(es), telephone number(s) and netmail address(es); whether submitted as long or short paper; which subfield(s) paper comes under; a 100-250 word abstract; and the length in words. Further details about the IJCAI-83 conference can be obtained from: Graham Wrightson (49-721)608-3975 IJCAI-83 Local Arrangements Institut fuer Informatik I telex: uni d 07826521 Universitaet Karlsruhe Postfach 6380 ArpaNet: WRIGHTSON@RUTGERS D-7500 Karlsruhe 1, W. GERMANY The programme committee for IJCAI-83 has now been formed and is listed below. Subfield of AI Committee Member Responsible Automatic Programming Yves Kodratoff Cognitive Modelling Aaron Sloman Expert Systems Lee Erman Knowledge Representation Ray Reiter Learning and Knowledge Acquisition Tom Mitchell Logic Programming Bob Kowalski Natural Language Aravind Joshi Planning and Search Randy Davis Robotics Takeo Kanade System Support Guy Steel Theorem Proving Joerg Siekmann Vision Harry Barrow Alan Bundy IJCAI-83 Programme Chairman --------   MSG: DOVER TROUBL DISTRIB: *RTS, *VX, *BBOARD EXPIRES: 11/05/82 11:48:10 OAF@MIT-MC 11/01/82 11:48:10 Re: still dead Repairs will involve some time (which you don't have) or lots of smarts (which I don't have). Print your papers elsewise or elsewhere if you need them this week. Next week there may be a better chance.... Oded  Date: 27 Oct 82 16:03:45-EDT (Wed) From: Mark Reinhold To: Charles Frankston Subject: C Itoh cit101 Via: UPenn; 31 Oct 82 16:37-EST Technical Magic Inc. 17742 Mitchell N Irvine, CA 92714 (714)-556-2380 Mark.  Date: 30 Oct 1982 2100-EDT From: Phil Budne Subject: PCRTSTY.1243 To: Bug-Crtsty at MIT-OZ Dies with inferior fork termination trap for the command 'PCRTSTY VT52'. -------  Date: 28 Jul 1982 1224-EDT From: James Haleblian Subject: The IBMPC terminal program and CRTSTY... To: cbf CBF, I was discussing using the IBMPC to dialup OZ with CRTSTY, and the subject of your terminal program came up again. Howard Trachtman said tht he would speak to you about the program, and he told me to send you mail about it again. I would * really * like to get a copy, and, if necessary, pay for it (if it doesn't cost too much). I am an MIT student (Howard told me to stress this fact). Thanks, James Haleblian  From: Rowe.SIPBADMIN at MIT-MULTICS Subject: Re: current events To: LSP at MIT-ML cc: cbf.spba at MIT-MULTICS In-Reply-To: Message of 29 October 1982 04:41 edt from LSP Well, I don't really know what you planned to accomplish by your mail, since many of us have already realized exactly what you said and have been trying HARD to start doing something about it for the past couple months. And it seems to be working. The SIPB tends to oppose change, and tends to move along like a huge, lumbering, cumbersome beast -- which is why some of us have been organizing to get things done around the Board, when necessary, instead of waiting for the entire organization to make a single-minded decision. Your opinion is in agreement with mine and with the opinions of several other "do-ers", although we disagree on some points. I appreciate your support for the ideals toward which I am working, although I found the "g.d. sipb won't move its collective duff and do anything!" tone disturbing. Especially since we ARE attempting to do exactly what you said we should. Why couldn't you have flamed before, when nothing was getting done and when we desperately needed some prodding?? BTW, I am already working on a Technology Review article, although I haven't focused it well enough. I'm taking a tech writing course and making contacts at all the appropriate places, which is useful. Wendy  Date: 28 October 1982 15:13-EDT From: HBB at MIT-MC To: BUG-CRTSTY at MIT-MC WE HAVE A TERMINAL HERE HP2621P WHEN I CONNECT THIS TERMINAL TO MIT SYSTEM I GET BLOCKS ALL OVER THE LINES. COMMANDS "TCTYP HP,TCTYP HP2621,TCTYP HP2621P,TCTYP 2640,CRTSTY HP2621, CRTSTY HP2640,CRTSTY HP2621P"DO NOT WORK. I SHALL APPRECIATE ANY HELP. THANK YOU  Date: 28 Oct 1982 11:33:51-EDT From: naha at mit-vax To: S.LEBLANC@mit-eecs, sipb@mit-mc Subject: What needs to be done. We are not goining to let physical plant run the wires from the roof. We are in no way going to suggest to them that this even needs to be done. When the time comes for this to happen (when the stars align themselves properly) this will happen as if by magic. The little elves will sail from the westernmost seas, snap their fingers at the stroke of midnight and the cables will appear. Clap your hands if you believe in fairies. -naha  Date: 25 Oct 1982 0759-EDT From: Stephen J. LeBlanc Subject: [Wendy E. Rowe : rough draft] To: ls.psr at MIT-EECS, Wendy at MIT-MULTICS, cbf at MIT-MC, Zabih.spba at MIT-MULTICS, zrm at MIT-MC, sjl at MIT-CCC Greetings, You have been drafted into the Chair's task force to deal with this proposal. People should start thinking how to extend, revise, and rewrite this into something presentable. Any chunks you come up with should be sent out, and I will try to collect them into something coherent. Specific ideas and suggestions and whatever should be sent to the task force via netmail, and (OH GOD) discussions in person. We could perhaps get together later this week if there is any reason. Steve From: Wendy E. Rowe Subject: rough draft To: s.leblanc@ee@mc Here you go, Steve. "Inconvinving" is a really neat word! Can we keep it?? I have come to believe that this is about what should be said, though maybe it needs some substantiation a bit. The real trouble is that all the statistics I have checked don't seem to help. See you soon. Steve The SIPB ha projected its budget. We realize that at current demand and spending we will probably run out of money sometime in march or April. The only we can see to avoid this is to severely limit the computer access and services we provide to the MIT student community. Since the SIPB is the only access most of these student have to a computer system on campus, we are very reluctant to do this. The reason for this is that the demand for our services has grown steadily over the past years. Our present budget simply cannot meet the demand for the computing resources by students. We find that even now we must refuse to fund several students whose applications fall within our policies because of monetary limitations. As the term progresses, we will certainly have to more severely limit our services in order to stay within the limits of our budget. This will probably include refusing to consider any new applications sometime next semester, which we have been forced to do in past years. This may leave a period of two or three months where a student without a computer account will not be able to obtain access anywhere to a computer system. We feel that there are two possible actions which could be taken now that would help eliminate or decrease this loss of basic services to students. The simplest immediate solutions would be an increase in our budget. In order to provide a reasonable level of student access to our services we would certainly need half again the money we have now. While any increase would be beneficial, maintaining a decent level of service throughout this fiscal year would require a substantial increase in our budget, assuming that current rates and arrangements with the Information Processing Services are maintained. A more long term, and possibly less costly solution would be to modify our agreements with the IPS so that more students can work for the same amount of money. This might be in the form of a modified rate structure, or in the SIPB purchasing a percentage (pie slice) of IPS Multics for student use. We feel that such a request is reasonable for the SIPB to make of IPS, because our student users don't require many of their services such as User accounts and consulting. The SIPB administers all of our 1000 plus users ourselves, thus greatly reducing the workload of the User accounts office. Also we provide nearly all the consulting that our users require as well as a substantial amount for other IPS users thereby being a definate addition, rather than drain on IPS consulting rescources. These meaures are in no way the final answer to MIT's student computing problems, but if implemented immediately they could show immediate results, and help to greatly improve the student computing enviroment this year. (copies should be sent to people like Peter Richardson with a note saying that any encourage he could giv econvinving people of the importance of this would be very helpful.) ------- -------  Date: 25 October 1982 01:15-EDT From: Charles Frankston To: SIPB at MIT-MC Remember that Steve Ott from DSI in New Hampshire will be dropping a Cermatek modem off in the office for us to test Tuesday.  Date: 23 October 1982 14:26 edt From: Pattin at MIT-MULTICS (Jay Pattin) Subject: guide to computing at MIT Sender: Pattin.PDO at MIT-MULTICS To: SIPB at MIT-MC Peggy Richardson from the UASO (Dean's office person in charge of dealing with freshman) called me yesterday wantin to know if I knew anyone interested in writing a guide to MIT computing for freshman and freshman advisors. Evidently, the freshman advisor meeting yesterday brought out a lot of complaints from advisors that they don't know what to tell their advisees to do when they say "I want to take a computer course, but I don't want to major in CS." I explained lack of 6.030, and that I thought 1.00 was designed to take its place. I also mentioned amazing varieties of lossage, and gave her a rough idea of the sorts of things that were going on - no details and few names. Is the SIPB interested in writing such a guide? -- Jay  Date: 22 Oct 1982 1131-PDT From: Ted Anderson Subject: RMAIL problems To: rem at MIT-MC, cbf at MIT-MC, bug-rmail at MIT-MC CC: "#SPACE.REQ[SPA,OTA]" at S1-A It seems that RMAIL does not recognise the increasingly common headers that come out of UUCP land that have a blank or broken from field that is continued on the line. RFC733 appearently permits this if the continuation line starts with whitespace. How about fixing up RMAIL? Thanks, Ted Anderson  Date: 22 October 1982 0023-edt From: Charles Frankston Subject: To: cbf@mc &print Due to high rates, usage of the computer between 9AM and 5PM weekdays is &print restricted to people who need to use it. Each user is allowed $10 per month &print (about one hour's usage) for emergencies. You have &[minus [user shift_limit] [user shift_spending]] left. Contact &print the SIPB if you feel you need to use the computer during the day regularly.  JPG@MIT-MC 10/19/82 15:20:17 To: CBF at MIT-MC CC: JPG at MIT-MC CBF@MIT-MC 10/19/82 13:10:06 I have a report that the ODE2 function does not exist on Multics macsyma. Is there anyone that can look into this? This was answered in MULTICS-MACSYMA-BUGS@MC : (Hopefully some day someone will install a better way.) ELLEN@MIT-MC 17 September 1981 01:06-EDT To use ODE2 on Multics, do batch(">lib>macsyma>share>ode2.macsyma"); and when it has been batched in, you will be able to use it.  Date: 19 October 1982 04:11-EDT From: Robert Elton Maas Subject: 1R in RMAIL bug? To: BUG-RMAIL at MIT-MC cc: littauer at BBN-UNIX Here's the header of the message I received and tried to reply to: Date: 18 Oct 1982 18:38:41 EDT (Monday) From: Ben Littauer Subject: Re: TAC/ITS telnet bug? In-Reply-to: Your message of 14 Oct 1982 17:23:47 EDT (Thursday) To: Ben Littauer Cc: moon at SCRC-Tenex at MIT-MC, EAK @ mit-mc, REM @mit-mc, frye at BBN-UNIX, ditmars at BBN-UNIX, herman at BBN-UNIX, JNC at mit-mc Here's the reply header I got from that when I used the 1R command in RMAIL. Where on Earth did it get that first Cc line from, the one with a null netaddress inside <> ? It looks like a bug in RMAIL? To:littauer@BBN-UNIX Subject:Re: TAC/ITS telnet bug? Header-Force:RFC733 Cc:Ben Littauer <> Cc:moon@SCRC-Tenex@MIT-MC, EAK@mit-mc, REM @mit-mc Cc:frye@BBN-UNIX, ditmars@BBN-UNIX, herman@BBN-UNIX Cc:JNC@mit-mc  Date: 19 October 1982 03:59-EDT From: Robert Elton Maas Subject: RMAIL created a bad interface-to-COMSAT file To: littauer at BBN-UNIX cc: BUG-RMAIL at MIT-MC I used the 1R command in RMAIL to reply to everyone, and edited out myself since I didn't want a copy. But RMAIL somehow created something that COMSAT didn't like, see below. I don't know if there was something wrong in the original syntax of the message (it looked ok to me, but I didn't look carefully) that RMAIL failed to complain about, or a bug in RMAIL. COMSAT@MIT-MC 10/19/82 03:31:18 To: REM at MIT-MC CC: MAIL-MAINTAINERS at MIT-MC Error in input request file. Parsing error: No recipient name was specified? Line stopped at is: RCPT:( (R-OPTION CC)) Message not sent and not queued; text of bad file follows: ------- FROM-PROGRAM:RMAIL FROM-XUNAME:REM FROM-UNAME:REM AUTHOR:REM RCPT:(littauer@BBN-UNIX) SUBJECT:Re: TAC/ITS telnet bug? HEADER-FORCE:RFC733 RCPT:( (R-OPTION CC)) RCPT:(moon@SCRC-Tenex@MIT-MC (R-OPTION CC)) RCPT:(EAK@mit-mc (R-OPTION CC)) RCPT:(frye@BBN-UNIX (R-OPTION CC)) RCPT:(ditmars@BBN-UNIX (R-OPTION CC)) RCPT:(herman@BBN-UNIX (R-OPTION CC)) RCPT:(JNC@mit-mc (R-OPTION CC)) TEXT;-1 Aha! This time I think you got it fixed! [Patch on SU-TIP] I tried my simple test in lisp, namely (LISTARRAY OBLIST) ^S and that worked correctly where it hadn't before. Then I did a lot of aborting output in LISP, and that all worked correctly where it hadn't before. Finally I did some aborting of printing out files from DDT, and that too worked each time, where it had been intermittant before. Thanks and congradulations. I bet the CC's of this message will be pleased when their TACs get the patch tomorrow and start working correctly too like the SU-TAC does now.  Date: 18 October 1982 1210-EDT (Monday) From: Guy.Steele at CMU-10A To: Charles Frankston Subject: Re: Jargon file In-Reply-To: Charles Frankston's message of 18 Oct 82 08:20-EST Thanks. She tracked me down, and gave me the spiel. She thinks it would be "a useful trade paperback". Well, I'll send her a copy. --Guy  Date: 18 October 1982 09:20-EDT From: Charles Frankston Subject: Jargon file To: GLS at MIT-MC, steele at CMU-20C cc: CBF at MIT-MC A Harriet Rueben from Harper & Row just called. Apparently they saw the Jargon file excerpted in Science '82 and are interested in publishing it as a book! She apparently managed to get the MIT computer center, who had never heard of you, so the call got forwarded down to the SIPB office. Her phone number is (212) 597-7233. I gave her your office number from MC inquire. Have fun and remember Psychology Today.  REM@MIT-MC 10/15/82 09:32:23 To: (BUG RMAIL) at MIT-MC Damn, that's the second time I've been burned by this. I have been saving my message edit in REM;RMAIL TMP. I want to pull in some material from a reference file, and the easiest way (RMAIL doesn't allow multiple windows because it's not the top level) is to M-X Insert File. Ok, I do that, edit out the parts I don't want, reformat the parts I do want, and then say C-X C-W to save my edit under the same name as before *** EXCEPT RMAIL SAVES MY EDIT UNDER THE NAME OF THE REFERENCE FILE I INSERTED, NOT UNDER THE NAME I'VE BEEN SAVING *** SO THE REFERENCE FILE IS NOW TOTALLY CLOBBERED *** Yes, I know if I had waited after C-X C-W before pressing , giving it 15 seconds to clear my output thru the network buffers and print the file name it was going to use, and then pausing to let me see if it was the right file name, I would have seen the bad file name and changed it back to what it was supposed to be. But changing it to the wrong file was such a dumb thing to do that I didn't expect RMAIL to do it, and unless there's some reason to wait I usually type ahead as much as possible, in this case wanting to save my edit before a system crash, and so I didn't wait, and my reference file got burned. So, can anybody think of a case where one would want to save one's edit under one file name, then insert another file by a different name, then edit some more, then save the result not under the original file name you've been saving things under but under the inserted file name?? It seems to me it's always doing the wrong thing here.  Date: 15 October 1982 17:31-EDT From: Michael E. Dove To: BUG-CRTSTY at MIT-MC Hello, I am interested in finding out how crtsty is structured and used to accurately work with all these terminals. Specifically the graphics capabilities, etc of some of these terminal. I find it very impressive that a driver using data can accurately emulate as many terminals as crtsty and tctyp can. I would appreciate any help you can give me Mike  Date: Wednesday, 13 October 1982, 12:45-EDT From: Bernard S. Greenberg Subject: Boredom To: INFO-COBOL at MIT-MC, FUN at SCRC-TENEX From the new "American Heritage Dictionary," second College Edition: boring (bo-ring, bor-), *adj*, Uninteresting and tiresome, dull; -boringly, *adv*. Synonyms: *boring, monotonous, tedious, irksome, tiresome, humdrum, dismal, dreary*. These adjectives all mean to be burdensome in the sense of inducing discontent and mental weariness. *Boring* implies dullness that causes listlessness and lack of interest. *Monotonous* stresses unchanging lack of variety. *Tedious* suggests slowness, longwindedness, and dullness. *Irksome* describes what is demanding of time and effort, and yet is dull, distasteful, or otherwise unrewarding. *Tiresome* intensifies the sense of *irksome*, and *humdrum* refers to what is commonplace, trivial, or characterized by unexciting routine. In this comparison, *dismal* and *dreary* apply to what is depressingly dull. Isn't that interesting?  Date: 13 Oct 1982 11:26:34-EDT From: naha at mit-vax To: LS.PSR@mit-eecs, ZRM@mit-mc Subject: Tuesday night prob set/beer parties Cc: SIPB@mit-mc, ls.grp@mit-eecs YOW! Bring my dolphin another Vodka Gimlet!  PAK@MIT-MC 10/13/82 09:29:19 To: CBF at MIT-MC I just tried crtsty vt102 at 9600 baud and it seems to work fine. Thanks for putting it in! Paul karger  Date: 11 October 1982 23:44 edt From: CFrankston-Prime.SoftArts at MIT-MULTICS Subject: more detail on VT102 Sender: COMSAT.SoftArts at MIT-MULTICS To: info-terms at MIT-MC *from: CBF (Charles Frankston) Local: info-terms@mc Original-date: 11 OCT 1982 15:11:56 It seems somewhat faster than VT100. Probably most visible improvement should be scrolling (on Line feed, or index) about twice as fast in jump scroll mode. Fill characters for VT102 vs. VT100 from manuals: VT102 VT100 at 1200 9600 baud at 9600 ---- ---- --- IND/LF/NEL/RI (Smooth) 20 160 192 IND/LF/NEL/RI (Jump) 2 16 30 DECCOLM (changes 132/80) 12 96 243 DECALN (fils screen with E's) 12 96 243 ED 132 9 72 not given ED 80 7 42 45 EL 132 3 not given EL 80 2 2 DECINLM (set interlace mode) 4 1 Others 1 1 Unfortunately, there are no manual timings for the insert/delete commands, which are standard on the VT102. The only real info I have so far is that they seem a little faster than the same commands implemented by a VT100 printer port option. Of course the insert/delete commands implemented by the VT100 printer port option (or by a VT132) were incredibly slow. ESC [ 4 h Insert mode ESC [ 4 l Replace mode ESC [ Pn L Insert n lines at the cursor ESC [ Pn M Delete n lines at the cursor Note that the above jumps, but takes a count whereas the scroll region stuff works on just one line at a time. Also, this is unlike the VT100 commands, which took the same count argument (Pn), but executed in a time proportional to the count! (Ie. they iterated shifting the entire screen by one line Pn times). Depending on just how slow they are, these might finally be useful for screen moves of more than a line or two.  Date: 11 October 1982 2324-edt From: Charles Frankston Subject: pmf-mode To: sipb@mc I took Jay's make-pmf-entry function out of his start_up and put it in >udd>spba>jjs>s>pmf-mode.lisp along with a mode hook to bind it to ^Z-p if what you're reading in is a pmf. I would suggest that members put the following line in their start_up.emacs: (putprop 'pmf-mode ">udd>SIPBADMIN>SIPB_admin>pmfs>pmf-mode" 'autoload) Feel free to flesh out more of a pmf mode. I just felt silly copying the function out of Jay's start_up. A word of warning: If you use this function, continue to check for exceptions like non-default cutoff dates!  Date: 13 Oct 1982 0836-EDT From: HES at MIT-AI Subject: Re: Yow! To: BLIV at MIT-OZ at MIT-MC cc: info-emacs at MIT-OZ at MIT-MC In-Reply-To: Your message of 12-Oct-82 1952-EDT ok now tell me why all of a sudden my emacs gives me cretinous mm instead of babyl which has been my setting since I've been on oz. -------  Date: Wednesday, 13 October 1982 06:48-EDT Sender: KLOTZ at MIT-OZ From: Leigh L. Klotz To: Richard Kenner Cc: H19-PEOPLE at MIT-MC Subject: Odd H19 (H89) problem I may be being naive, but it sounds like a bad filter capacitor to me. Leigh. -------  Date: Wednesday, 13 October 1982 05:31-EDT Sender: PGS at MIT-OZ From: PGS at MIT-MC To: DCP at MIT-OZ, bug-its at mc cc: twenex-haters at MIT-OZ AI is being backed up these days, and its disks seem fairly stable. The processor hardware is not currently flaking out, although we have no idea how long that will last. When I connect to it, often in the daytime, I am usually the only user using it. If we do indeed reconnect AI to the Chaosnet, then it'll make a pretty good mail computer, so I see no reason not to leave BUG-RANDOM-PROGRAM on it. Something we should take note of here is that Eric Ostrom, in his capacity as AI Lab Facilities Coordinator or some fancy title, has tracked down a KL Model A CPU (a la MC) that is being sold for $7500, supposedly in working order. The idea is to provide a machine of MC's power to AI lab members who prefer ITS to Twenex. If we bought one it would probably go where AI is now, both physically and spiritually (including network ports). Extra space would need to be allotted for memory; presumably 921 (Marty's office) would go away. Everyone who would like to see AI have a KL Model A running ITS should send a message to the effect to Eric@ee (not at OZ, where he doesn't read his mail), and CC it to PHW. Something strong along the lines of "I understand you're considering getting a KL to run ITS, well, do," is probably appropriate. -------  MSG: REVSEM 4 DISTRIB: *CMU, *EE, *MULTICS, *OZ, *RTS, *VX, *XX, *DM, *MC, *ML, *AI EXPIRES: 10/20/82 13:40:17 AGRE@MIT-MC 10/13/82 13:40:17 Re: Ken Kahn at the AI Revolving Seminar On Tuesday 19 October at 4PM in the 545 Tech Sq 8th floor playroom, Ken Kahn of Uppsala University in Sweden with speak on "Lisp Machine Prolog". Abstract Prolog is the first "logic programming language". It is very popular in Europe and Japan. I've implemented Prolog and various extensions on the Lisp Machine. First I'll talk about what's interesting about Prolog and how it compares with Lisp and Micro-Planner. Simple examples of Prolog programs will be presented. The implementation and the unique features of LM-Prolog will also be discussed. All come. Want to hear about coming AI Revolving Seminars but don't read a *MIT bulletin board? Add yourself to the Revolving-Seminar-Distribution@OZ mailing list or ask me to. Who'd like to give an AI Revolving Seminar in the spring? Who should be drafted to give one? Coming: 26 October - Dave Levitt 2 November - Demetri Terzopoulos 9 November - Gerry Sussman 16 November - Rod Brooks 23 November - Dan Weinreb 30 November - Jay McLelland from UC San Diego 7 December - Laurie Miller  MSG: VAX730 1 DISTRIB: *CMU, *EE, *MULTICS, *OZ, *RTS, *VX, *XX, *DM, *MC, *ML, *AI EXPIRES: 01/22/83 23:22:52 GJC@MIT-MC 10/13/82 23:22:52 Re: Usage of VAX730 at MIT's JCF. I've just read in the "JCF Newsletter" that they have a VAX-11/730 with 4 megabytes of memory with 16 (count 'em) terminals attached which is to support the students taking course 2.10 (intro to fortran). Yow!  MSG: *MSG 1367 DISTRIB: *BBOARD EXPIRES: 10/20/82 23:43:53 GSB@MIT-ML 10/13/82 23:43:53 Re: [Arpanet-BBoards: Apology] Date: 13 October 1982 23:37-EDT From: Glenn S. Burke Subject: [Arpanet-BBoards: Apology] To: ARPANET-BBOARDS at MIT-ML Remailed-date: 13 Oct 1982 2343-EDT Remailed-from: Arpanet-BBoards-Request at MIT-ML I apologize to those of you inconvenienced or offended by an earlier message to arpanet-bboards, which came under header from WMARTIN at OFFICE-8, and appeared to be an arpanet-bboards archive. He is in no way responsible for the screwup. The offending sofware will be attacked with a hatchet. On the lighter side, what went out was not exactly an archive; it was partly "new material", including things not intended to be posted...  Date: 12 October 1982 20:49-EDT From: David C. Plummer Subject: My Ethernet Address Resolution protocol To: CBF at MIT-MC, DCP at MIT-MC, EAK at MIT-MC, jek at SCRC-TENEX, moon at SCRC-TENEX, rwk at SCRC-TENEX, geoff at SRI-CSL, Westcott at BBNF, Tappan at BBNG, Burchfiel at BBNA, DClark.INP at MIT-MULTICS, Mishkin at YALE, avb at SU-SCORE, pratt at SU-SCORE, schur at DTI-VMS, rashid at CMU-10A, dalal at PARC-MAXC, JNC at MIT-XX, POSTEL at USC-ISIF, rogers at USC-ISIB, dos at BBN-UNIX, gateway at BBN-UNIX, gurwitz at BBN-UNIX, hinden at BBN-UNIX, iccb at BBN-UNIX, macg at BBN-UNIX, sam at UCB-C70 It has been pointed out (by Burchfiel) that my ethernet address resolution protocol could be made more general, with the practical application of using it in Packet Radio Networks as well as the Ethernet. The generalization is to make the hardware medium's address be a variable inside the packet. For the DEC/Intel/Xerox Ethernet this would be 6. I think this is reasonable, and unless I hear any objects, I will merge it into the protocol. (The length field for the protocol address currently takes up two bytes, so to make this as compatible as possible with existing implementations I will probably use the byte of the length field that is normally zero.) I will put some time into this and try and get it in the US Snail to Xerox for a type field by Friday, and will try to get a verbal allocation over the phone (since several people have started asking for the value of the type field). At the same time I will send it to Jon Postel for pulication as an RFC. (Jon, I don't know how things work, but is it possible to get a number for it now that I can include on the copy I send to Xerox?)  Date: 12 Oct 1982 1256-PDT From: Jeff Broughton Subject: Query To: cbf at MIT-MC Why don't you call in collect, I need to pick your brain about MarkIIA disk controllers?  MSG: *MSG 1366 DISTRIB: *AI, *DM, *MC, *ML, *XX, *CMU, *MULTICS EXPIRES: 10/26/82 18:04:13 JCM@MIT-DMS 10/12/82 18:03:59 Re: Important Things Just for fun, I'd like to collect a list of "essential papers in computer science" -- papers every computer scientist should be at least casually familiar with. If you have a minute, please send me a few titles and I'll post the results. Feel free to list your own papers if you like, but please try to limit the size of your response. Thanks.  Date: 11 Oct 1982 1552-PDT From: Hon Wah Chin To: cbf at MIT-MC hey you! how about it?