slogin: ) : THE UNIX NEWSLETTER VOLUME 2 NUMBER 8 SEPTEMBER 1977 Page 1 NOTICE This document may contain information covered by one or more licenses, copy- rights, and non-disclosure agreements. Circulation of this document is restricted to holders of a license for the UNIX or Mini-UNIX software system from Western Electric. Such license holders’ may reproduce this document for uses in conformity with the Unix license. All other circulation or reproduction is prohibited, DISASTER ENDED As we informed everybody on the North American mailing list, the system used for preparing both the newsletter and the software distribution was hit by we know not what. Every controller on the bus had bit 5 blown. We were off the air for 4 weeks but, “hossanna", we are now (October 29) fully restored except for some memory and a Versatec. This newsletter should be mailed on November 3 and immediately thereafter we will resume processing software requests. The remaining issues for this year will be mailed at three week inter- vals. SOFTWARE DISTRIBUTION It will take a few weeks for us to handle the backlog of request for software distributions which has built up during the time we were down. ee We have received corrections for the Harvard magtape drivers and an enhancement of the Harvard TU10 driver for multifile tapes. This are now on the distribution tape and we will mail diff listings for these changes, within the week, to those who have already received D As the letter from Andy Tanenbaum indicates, we messed up the Pascal source on the tapes already sent out. We have made the corrections and all tapes sent from now on should be okay. If you have already received distribution 3 and need the Pascal, please write for it. The missing source is too large for you to keyboard from hard copy, so a tape will have to be sent. The easiest thing is probably to request distribution 3 again. This will also save your keying the tape driver corrections. NEW TORONTO RELEASE The tape containing the new Toronto release has been received. As is explained in the letter from Toronto, a new license form must be submitted. As before, the license form may be sent by educational in- stitutions to the software center with the request for the software. thers will have to negotiate directly with Toronto. MEETINGS Planning for the early summer meeting in New York has begun. The local committee would appreciate your returning the reservation form as soon as you can. PROGRAMMERS’ WORK BENCH The Programmers’ Work Bench has been released for licensing. Licensing details are available from Western Electric. A paper by T. A. Dolotta and R. C. Haight, "PWB/UNIX - Overview and Synopsis of Facilities", describes the system and may be requested of Ted Dolotta at Bell Laboratories, Piscataway. Address editorial material and software submission to Melvin Ferentz cla CUNY/UCC 555 West 57 Street New York, N.Y. 10019 Subscription requests, payments and address changes should be addressed to Armand Gazes Physics Department Brooklyn College Brooklyn, N.Y, 11210 Fl (es 5 | wy" Computer Systems Research Group The University of Toronto Toronto, Ontario, Canada = Melvin Ferentz efo CUNY/UCC 555 West 57 Street spot 10019 New York, N.Y. Dear Hel: Please Find enclosed with this letter the long-promised second University of Toronto Software Distribution. The newsletter once said that people should send everything which might be useful, Well, this tape contains over 1050 files, with over 8200 blocks. Maybe this will change your policy. (Actually, the bulk of the stuff is a few large packages, which people have actually been asking for, Deleting the trivial stuff would not cut the size very much.) Once again, we are distributing only to educational and other non-profit UNIX license holders, Other cases aust be decided by direct negotiation, Also, once again people must sign the baroque masterpiece which the University’s lawyers insist must accompany any software. One other problem to note: Your version of "stp" may not read this tape. (Our version has MDIRENT==2700, I think yours prob- ably has a value of 2000. It is hard to tell, since there are so many versions floating around.) Our version is on the front, in the "tp" readable section, along with the first bit of documenta- tion. I have verified that those instructions will allow you to read the taps. Our version does not depend on any special tape driver modifications, If there are any problems, please call. Sincerely yours, Lrihe Tebow Mike Tilson 416-978-6320 PS. plexve send The Fape beck with the cegvested Aster butrend on FP — berms enchseed. HARVARD SCIENCE CENTER HARVARD UNIVERSITY CAMBRIDGE, MASSACHUSETTS 02138 September 21, 1977 Professor Melvin Ferentz e/o CUNY/UCC 555 West 57 Street New York, NY 10019 Dear Mel: Here are bug fixes for the TM-11 driver provided. on the third distribution tape. None of the bugs are substan- tial, and their only impact is in extra features available on the raw device, such as the write eof and the seek backward Note that the first fix {on line 214) is just to remove a reference to B_PHYS, and replace it with an equivalent test for raw 1/0. My understanding was that you bracketed this portion with IFDEF/ENDIF CPU70 so that line numbers may not match the distributed copy. : In addition, I have included a listing of changes made to include a file positioning capability for the raw device. This new feature allows positioning by file, using the stty oall but with arguments very similar to seek. The address of the routine tmstty must be added to conf.c (or c.c if you prefer). Sincerely, Vachet. Tucker Taft { Systems Programmer ‘Harvard Sclence Center Rm 110A Cambridge, Ma 02138 PLS. Also Find inladed bug foes for WINE. C, as mace ay they art Known, : KATHOLIEKE UNIVERSITEIT FACULTEIT DER SOCIALE WETENSCHAPPEN r 4 Prof. M. Ferentz Brooklyn College of CURY Brooklyn NY_{1210 USA RIE and fil ipping software sought, t pr Who has RJE and/or file shipping software between UNIX and IBM OS-HVT/VS1/VS2/MVS systems, NUMBGEN. September 6, 1977 A 2780 emulator for UNIX would be great to start with, va. $700/GR/wh Please contact: George Rolf Psychological Laboratory Erasmuslaan 16 . Nijmegen Dear Professor Ferentz, . att . The Netherlands As you might have noticed, I have switched departments. About 3 months ago, I called you for information about % heavy loaded UNIX systems, and file shipping between UNIX and IBM 360/370 machines, You directed me to Lew Law ai Harvard, and co Chuck Prenner st Berkeley. From them, I got very valuable information for a follow-up on my initial proposal, The proposal is still in the mill, and may never get our, but my new department is interested very much in the RJE/file shipping part of my proposal, Therefore, I have included a@ note to be posted in UNIX News, At the same time I wrote to Rick Haight at Bell Labs, a pointer that I didn’t foliow ac the time I called you. Thanke a lot. Sincerely, George Rolf Bulagen: ERASMUSLAAN 16 NOMEGEN _ TELEFOON (880) 31. 2633 VALE UNIVERSITEIT SUSFACULTEIT WIBKUNDE WISKUNDIG SEMINARIUM 1007 perder 2 . tae becbieae 108 Prof. Malyin Ferentz, powata 7181 c/a CUNY/UCC, wafetees SAR HED 10 555 Wast 57 Street, New York, N.Y. 10019, U.S. As Um kenneth vw beiat wan one kenmerk deum 26.09.77 Dbiltage(nt ernterweep Oeor Mel: Apparently someone laft part of the sources ta aur PASCAL system off the third distribution. ¥ If this wes marely an oversight, covld you correct it? Since we are at the end of the alphabet, I have a sneaking suspicion thet maybe the tape filled up while copying our stuff onto it. If so, I would suggest you remove the binary files and tibrartes and replace them with the sources. People can maka their own binarias from the sources, but the reverse is much trickier, Candidates for elimination aret PCOPPILER.B, PALIB.O, PC, PCLIST, PI, PILIB.O, PIPCLIS8.0, PPASS 1, PPASS 2, SHARE. Yours truly, Qn dy Andy Tenenbaum. The University of Calgary ; 20 24 AVE. NW, . FACULTY OF ENVIRONMENTAL DESIGN: een: CANAOA + TELEPHONE: (403) 284-6601 September 2], 1977 Melvin Ferentz c/o CUNY/UCC 5% West 57 Steeak New York, N.Y, 16919 Ovac Ne. Ferentz: I have 4 modified version of UNIX in which the large 1/0 buffers have been moved out of kernel space. This has several obvious advantages, particularly for machines without separated I/D space. The memory management design dictated a reduction of the buffer size to 512 bytes (from 514); the only dependency on this that I have found is a bug in iget.c/itrune. All the standard device drivers are completely compatible with this mod (although two of cur own had to be altercd). In the assembler suppore I have only touched m40.s, so a small amount of work might be re- quired to get it working on, say, an 11/68. Unfortunately, we can't write tapes at the moment {only cead_~\ them!), but should be able to within a couple of months. I tho: ‘da let you know now in case anyone has simi . : - By the way: a possible answer to D. W. Taylor's problem AN] (UNIX news, Januacy, 1977). There is a bug in tty.c/ttwrite, The last call to ttstart should be bracketed by an SPL5/SPL@ pair. This can cause occasional loss of characters with the KL/DL/DC drivers, most ‘obviously when the user program does sin- gie chacacter write requests, The DH driver doesn't have the problem (and if the above mod is made, then the priority fiddling in dh.c/dhstart could be removed). Fe. tiie R. Sidebotham reneuld, yor ie University of Illinois at Urbana-Champaign Graduate College CENTER FOR ADVANCED COMPUTATION Urbana, Iilinois 61801 September 28, 1977 TO THE UNIX COMMUNITY (1.e., attendees at Urbana meeting in Hay of 15977) Has anyone out there started to use the Qume Sprint Micro 3 Series of daisy wheel printers? We'd be interested in a hardware interface to it and/or a Unix filter for nroff output to it. Especially if anyone has done a filter for the twintrack model we'd like to hear about it. Failing that it seems likely we'll have to do one ourselves {with our highly competent but overworked and underpaid staffl}. Karl C. Kelley, Sr. Research Programmer Computing Services Office 344 Advanced Computation Building University of Illinois at Urbana-Champaign Urbana, Illinois 61681 THE UNIVERSITY OF NEW SOUTH WAL P.O. BOX 1 + KENSINGTON + NEW SOUTH WALES + AUSTRALIA » 2033 TELEGRAPH: UNITECH, SYDNEY + TELEPHONE 663 0351 is ey. Lys SCHOOL OF ELECTRICAL ENGINEERING . o PLEASE QUOTE ° SL:AG Se Professor Melvin Ferentz Brooklyn College of CUNY August 17h, 197? Dear Professor Ferentz, The third Australian UNIX Users Meeting was held again at the University of Hew South Wales on Monday July 4, We were very pleased to welcome our [ret overseas visitor, Jeff Rottman, a graduate student from Berkeley. Jeff was able to provide us with much relevant and first hand information on the UdIX scene. (We were delighted rhat he was able to spend a whole month with us. More so since he has fixed some remaining bugs in a Fortran compiler which he once sent us, and which Ls in regular use here.) We were also able to welcome representatives from the Computer Section of the Australian Atomic Energy Commission who have recently become UNIX licensees. Our fourth meeting will be held at the Australian Atomic Energy Commission, Lucas Heignta, N.S.W. next February. . The University of Wollongong has had a free standing version of UNIX running now for nearly three months on their Interdata 7/32 but there is stiil much to be done before the transfer is complete. They have been able to adopt the Interdata Fortran and Basic compilers to run under UNIX. Progress slowed recently when Richard Miller decided to take a long overdue and well-deserved vacation, At the University of New South Wales we are still digesting various sottware imports, including the second Chicago software distribution. We recently made a concerted effort to update all the documentation prior to reprinting the manuals. We recently discovered by accident that a much improved "NROFF Users Manual" is available as Bell Laboratories Computing Science Technical Report #54. Is the existence of this generally known? When in March I sent you a letter regarding the availability of documentation on the UNIX Operating System, I little realised the full extent of the task ahead. I now understand how publishers earn their share af the customer's dollar! After correspondence with Western Electric last year I had understood that they would not object to distribution in the manner proposed. In March however Western Electric suddenly revived its interest and re-opened the correspondence. Thus when the firsc of some eighty requests reached me after the middle of May, I had to send out letters explaining that there had been a delay. Subsequently Western Electric agreed to the distribucion provided that the prospective purchasers, being coyered by a UNIX license, also entered an agreement with my University to treat the documentation received from ua as if it had been obtained directly from Bell Laboratories. More lettera! So far I have received back many completed agreements, and have despatched the books as requested, with the initial batch being posted on July 22. However agreements have still not been returned by a sizable fraction of the people who wrote originally. Some of the letters have taken a very long time to reach me, and I am concerned that some may not have arrived even yet. For instance two letters which had been despatched promptly at the end of April by surface mail did not arrive here till the end of July. If anyone feels that a reply is long overdue would they please write again (by airmail‘). Yours 7 Joh ona. 0 Z 0 2 Sept 6, 1977. Prof. Melvin Ferentz, c/o CUNY/UCC 555 West 57 Street New York, N.Y. 10019 Dear Prof. Ferentz, We are currently exploring the possibility of using UNIX for in- house timesharing and word processing. Soma of us are intimately acquainted with the advantages of UNIX, but there ara some that feel that it is not appropriate. Particular criticisms include : "the text editor is written for programmers, not secretaries," and hence that secretaries,could sot learn to use it; "the formatter is too obscure and needs too much to work it", also apparent trouble for secretarie user interaction With the intent of backing up the sseertions we are making to our people about the performance of UNIX in a "secretaries' workbench” environment, I am soliciting opinions from other users on their experiences. If you have the time, pleass write to me on how UNIX has performed for you the role of a non-programmer system for word, processing (text editing and formatting), meseage handling and/or special (administrative and managerial?) funcrions, I don't know what I can offer in exchange -- parheps a copy of the report I'm preparing from all this? Thank you for your a Yours, Rick rarlane, Asst. Systems Analyst. (416) Se0-0188 ‘Wha Chay (Uhiathye aieltigy Cott Whoa? CANT 58 vet 57 ster, New exh 11 OW Tagene 22 977:9000 September 19, 1977 Prof. Melvin Ferentz CUNY/University Computer Center SSS West 57 Street New York, N.Y. 10019 Dear Professor Ferentz: This lettér is to announce that preliminary arrangements are being made for the UNIX Users Group meeting in New York. We are currently in the process of making arrangements to hold the meeting at Columbia University from Wednesday, May 24 through Saturday May 27, 1978. I would also like to announce that we are trying to arrange accommodations at Columbia for ail of those participants in the convention who would prefer not to pay high commercial rates for a room in a downtown hotel (as high as $40/night). Columbia has single and double rooms available, with the rooms arranged in two-room suites sharing the same shower; each of the two rooms is locked separately. Linens, towels, and soap are provided, and’beds are made daily. The approximate charge is $10/night for a single room and $6/night for a double room. The rooms are not luxurious, but they are liveable and convenient {five minutes walk from the convention center); downtown hotels can be more. than a half-hour away by bus, Subway, or cab. To help us make further arrangements, I would appreciate it if each UNIX installation would return to us a copy of the enclosed form to give us some idea of the number of people we can expect. Further information will be published in ;login: as it becomes available. Sincerely, dim Bis Steve Eisen