* * * * ORK * * _ * * KK * * RIK * * RE OK * * eK O® * * * * * * * * OX * * x ROX RK * *% Xx KK * * * ORK * * O* * eK * RR OK * RK * * SER * * * * SEROK IK * * RK SR IRR ORO KR ORC KOK OK IE ICRC OR DAC EC KC EG AK KO EE DUNE Drivew * Repip van 63,4 * NOTICE * é * * att wu * This document may contain information covered * colo) ackage 65 2 f * by one or more licenses, copyrights, and * * non-disclosure agreements. Circulation of * & *x this document is restricted to holders of a * Ovdey wildy ORI% * license for the UNIX software system from * eauvals * Western Electric. Such license holders may * * reproduce this document for uses in conformi~ * * ty with the Unix license. * F spe \ : | | : ‘Nea. * All other circulation or reproduction is * S WW - * prohibited. * * * SHC CCR EO AR ERNE ER NC OR SK ORO A CE EE CCE EC EE EO OK OO CRC EK UNIX NEWS Due to the closing of facilities of the City University of New York at the end of May, there was no May issue of Unix News. The University is back in business and we would hope there will be no further closings. Our fiscal year and our current Board of Trustees both expire at the end of June. . The suggestion from Australia that the hotline information be captured is well taken. We will, in the future, reproduce all hotline information which is worthy of permanent recording in Unix News and remove it from the hotline file. The hotline will then contain only messages less than two months old. 7 . The response to the survey questionaire has been approximately 30% to date. It wilt be compiled and made available. I1f we can arrange for the keyboarding, it will be machine readable and available from the Chicago Circle Software Exchange as well as on line. A concise form of the "haves" and "wants" will be published here. SOFTWARE EXCHANGE Notice is reproduced below of the firet mailing from the Software Exchange. If you are on the reproduced List you will receive a tape. If you are not and want a copy send a magnegtic tape or 5 Dectapes. The Harvard software, with the exception of Basic js now at Chicago Circle. If you wish the Harvard software, sign the license agreement and mail it together with a tape to } Chicago. If you are not an academic institution, or if you want Basic, get in touch with Lew Law at Harvard. Ignore the part of the notice that asks for money if you are ordering from Chicago. If you attempt to use the new archiver, you will need the program be low: f A mk temp( name) ANG . char *name; OW t register char *ptr; Q cs register r3 ptr = name; % e r++ (gah Bea hes r = getpid(); do *--ptr = °'@’ + (r%1O); : while (r = r/10); while (stat(name) >= 6) xeeptr = ’a’s return( name); ny Number 6 May-June 1976. ~~~~SOSOS*~C=~‘“s*~S*~*~*~*~*~*~:~:~:”:””””S gees MEETINGS Notice is given in the letter from the University of New South Wales of the first Eastern Hemisphere "East Coast Meeting". Our Correspondent's notes have been misplaced and the report on the Larvard meeting will appear in a future issue. Lwrs our MANUALS The response to Harvard's offer to print the manuals was much greater than expected and the offer is repeated. The original notice, with a new closing date, follows! The Science Center at Harvard is willing to undertake the task of reproducing and distributing the manuais for UNIX. The enclosed order form may be returned no later than August 2, 1976. All orders received by that date will be part of the second printing. Closing dates for subsequent printings, if auy, will be announced. For logistical reasons, each of the manuals will be available only as indicated below, with 3&3 hole punching and no binding. Orders should be batched, one order per in~ stallation and will be mailed to a single address which must be at the users’ group’s ad- dress of record for the installation. The "UNIX PROGRAMMER’S MANUAL" Sixth Edition dated May 1975 will be reproduced in its entirety. Most installations will want to remove several pages which most users need not know about. "DOCUMENTS FOR USE WITH THE UNIX TIME-SHARING SYSTEM" Sixth Edition will be repro- duced omitting sections 1 (Setting Up UNIX) and 13 (Qn the Security of Unix). : The price of each manual will be dependent upon the total number of orders for that manual and will range between $8.00, if 10@ copies are printed, and $12.00, if as few as 10 copies are printed. The order form may be sent in advance of, or with purchase orders or checks. Assume the $12.00 figure. Correct invoices and refunds will be sent. SEVENTH EDITION? Conversations with Dick Shahpazian and Dennis Ritchie indicate some ambivalence within Bell-Western with respect to a new edition of UNIX. There are a number of changes that exist at Murray Hill and Ken Thompson is bringing many more back from his year at Berkeley. It seems clear that there will eventually be a new edition incorporating the non-incremental changes. The question that is still unresolved is whether a “totally de~ bugged" edition six will be formally, or informally, released to the users before the quantum jump to edition seven. Mike O’Brien and I have expressed the desire that such a_ release be made to the software exchange, if only for archival purposes. The argument against, of course, is that time taken by Ritchie and Thompsou to produce such a version is time taken away from new goodies and the preparation of the next version for distribution. We don’t know how influential they will be, but your comments are welcome and if sent to Unix News will be vead by the powers that be at the vendor. ' EXODUS FROM THE INNER CITY As part of a major move of Western Electric from New York, the Patent Licensing func- tion ‘at Western is moving to the "sun belt". Effective July 1, the address is: Richard G. Shahpazian Patent License Manager Western Electric P. 0. Box 25000 Greensboro, NC 27420 telephone 919-697-2861 UNIX NEWS ADDRESS Prof. Melvin Ferentz Brookyln College of CUNY Brooklyn, NY 11210 HRSTSS Software - SS. The following software has been developed or modified by Harvard Me for use with the Bell Unix operating system on a PDP 45: PPL, BASIC, LISP 1.5, DDT,CDDT, MACRO 11, LINKER, HARVARD SHELL. This software is available to other academic institutions under the following conditions: : 1) The Institution or recognized representatives thereof shall sign an agreement that: a) the software provided shall be used only by the insti- tution signing the agreement and shall under no circumstances be further distributed; b) the software provided shall be used for academic, non- commercial purposes only. 2) For BASIC, proof of purchase of a Digital Equipment Corp. License to use BASIC/PTS V@l Software ($500) or its equivalent must be furnished. A copy of the DEC invoice for payment for purchase of the license is sufficient. 3] No support for the software will be provided. 4) Additions and improvements to the software provided shall be made available to Harvard University on a reciprocal basis. $} Payment will be made for DECtapes and documentation* plus a $10 handling fee for each DECtape. Requests and inquiries should be sent to: Lewis A. Law Director of Technical Services Science Center, Harvard University 1 Oxford Street ° Cambridge, Massachusetts 02138 617-495-2627 ‘Documentation, other than Users Manuals, will be available hopefully by mid-1975. (0 sein teat CONTENTS OF FIRST MAILING . UNIX spc : 1) Motorola M6800 assembler (C) 2) “em” (ed for mortals) ~ for faster terminals. A superset of ed. (C) 3} “setup” - general management package for downline loading of satellites (Cc) Above from Jon Rowson, Queen Mary College, University of London. 4) Ken Knowlton's EXPLOR and MINI~EXPLOR in FORTRAN, plus MINI-EXPLOR in C - for easy, fancy line-printer graphics. With examples. 5) DL-11lE driver which handles dialups. Good for CENTREX phone systems which do not have long-space disconnect or second-party disconnect in the hardware. (C) Above from Lou Xatz, Columbia College of Physicians and Surgeons. 6) "“xtpip! - file-hand@ling package which can transfer DECtape and RKOS files between UNIX and RT-1l formats. Well commented and documented. 7) PDP 11/40 UNIX FORTRAN - completely rewritten runtime system, uses 11/40 FIS and runs on no other kind of machine (except Cal Data's with FIS, of course). Runs fast, but supports single precision only. 8) "convert", an attempt at a program to convert 11/45 Floating Point assembly language to 11/40's by inserting a “jsr" to a special entry in “fptrap". Each floating instruction thereby grows by two words, but no signalling takes place. System overhead is greatly lowered. May fail if assembly source is in very weird format. Rbove from Chemistry Department, University of California at San Diego. fete ate Sse saree lee |» New archiver - includes “ar.c", new "ld.c" which reads the new style, and | “arcv.c" which converts from old to new formats. The stinker is that \ “ar.c” uses long (double-word) integers, which can only be compiled by the new C compiler whith we can't distribute yet. -Therefore, executable modules are included. Instructions included. Above from Bell Telephone Labs. 7 oh a pst. nen 10) Modified ‘ac.c* - takes “-t" Flag to do accounting over ports as well as people. Useful for finding out which terminals or lines are used most often. 11) Modified “kl.c™ which handles DL~11E dialup lines. handle second-party and long-space disconnects, but is smaller than #5. Expects hardware to [ 1 I f Above from University of Illinois at Chicago Circle. The modified versions of C, nroff, and UNIX cannot be distributed until Bell Laboratories has released them. If Bell decides that they have been "substantially changed", any or all of them may require a separate license from Western Electric. . The Harvard HRSTSS software is separately handled. This edition of the newsletter should include an order form for the software. The manuals must still be purchased from Harvard. The following should be a complete list of those slated to receive a tape. “If you want your name added to the list, send a magtape or several DECtapes to: Michael T. O'Brien Department of Information Engineering University of Illinois at Chicago Circle P. 0. Box 4348 Chicago, Illinois 60680 THE LIST H.J. Thomassen Katholieke Universiteit Nijmegen, Netherlands Hernan Suérez-Flamerich Universidad Simon Bolivar Caracas, Venezuela Professor E. Milgrom : Université Catholique de Louvain », Louvain-la-Neuve, Belgiua Bill Mayhew The Children's Museum Boston, Massachusetts John Cornelius uesD San Diego, California a Martin Tuori Computer Systems Research Group Toronto, Ontario, CANADA Steve Holmgren Centex for Advanced Computation Urbana, Illinois Lou Katz Columbia College of Physicians 6 Surgeons New York, New York Jon Rowson Queen Mary College University of London Lew Law Harvard University Cambridge, Massachusetts Mel Ferentz Brooklyn College, CUNY New York, New York Ernie Chang University of Waterloo Ontario, Canada Pat Fitzhenry Aviation Research Lab Savoy, Illinois Dennis M. Ritchie Bell Telephone Laboratories Murray Hill, New Jersey Harold Solow RLG Ausociates Reston, Virginia THE UNIVERSITY OF NEW SOUTH WALES P.O. BOX 1 + KENSINGTON + NEW SOUTH WALES + AUSTRALIA + 2033 TELEGRAPH: UNITECH, SYDNEY * TELEPHONE 663 0351 EXTN. 2805 PLEASE QUOTE Jk. 2ist June, 1976 Professor Kelvin Ferentz, Physics Department, Brooklyn College of Cuny, Brooklyn, N.Y. 11210 Dear Professor Ferentz, We trust the recent news of the fortunes of CUNY has not been as graphic as portrayed In the newspapers here, and that the future of "UNIX News" is still as bright as ever. We are writing to inform you that the Department of Computer Science at the University of New south Wales has decided to convene a UNIX Users Meeting on August 29, on our “East Coast". Details are available from the undersigned. All are welcome. So far besides ourselves, we understand that the University of Sydney and the University of Newcastle are also UNIX Ticencees, and we understand that the Australian Atomic Energy Commission, and maybe others, are negotiating currently with Western Electric. So we do not expect to be short of attendees. We would be grateful if you could help us publicise this meeting through "UNIX News’. We expect to send you a report in due course. . Also is there any way we can receive copies of material disseminated via the UNIX Hotline without actually dialing up? (Our administration rather frowns on international calls and besides we don't currently have any terminals connected to dial-up modems.) Finally would you enter another subscription to “UNIX News’ for the undersigned. A cheque for $10.00 is enclosed. Yours sincerely, Chie L-SOHN LIONS Department of Computer Science, University of New South Wales, Kensington 2033 AUSTRALIA. THE UNIVERSITY OF NEW SOUTH WALES P.O. BOX 1» KENSINGTON = NEW SOUTH WALES + AUSTRALIA > 2033 TELEGRAPH: UNITECH, SYDNEY = TELEPHONE 663 0351 extn. 2805 Jt DEPARTMENT OF COMPUTER SCIENCE 21st June, 1976. Professor Melvin Ferentz, Physics. Department, Brooklyn College of CUNY, Brooklyn, KEW YORK 11210 Dear Professor Ferentz, Please find enclosed the survey you distributed earlier. It has been completed with much enthusiasm but perhaps with not as much care as you would prefer. (ive Lat at rertyprd af Gs ate!) Our principal local extensions of UNIX (almost entirely due to lan Johnstone) include: « CRIT Card reader handler « LPI Block oriented printer handler = XY Calcomp plotter handler ~ L¥EL Versatee plotter handler « a remote/local batch subsy: terete \ The Femote batch aubaystam emulates the U-200 terminal for communication with KRONOS on a Cyber-72. Features | full spooling, full accounting, job name invariance \ (KRONOS users will know what this means) and operator~ | less operation. { The local batch subsystem uses the same card reader ond \ | printer as remote batch, and local and remote jobs may | \ be freely Intermed neem ern A Shared data segments. This feature was implemented originally along with the block line printer driver, to Improve the efficiency of the batch subsystem. Uses a modification of the shared text segment mechanism. Requires changes to both operat~ . ing system and loader. Interlocking provided by "tp" and “v" system calls. Yours sincerely, sum LIONS - “Senior Lecturer. Department of Computer Sclence. FACULTEIT DER WISKUNDE EN NATUURWETENSCHAPPEN Toemooiveld, Nipmegen Telephone 080-55 88 33 KATHOUEKE UNIVERSITEIT Telex 4.82 28 winac ni NIJMEGEN THE NETHERLANDS May 25, 1976 Professor Melvin Ferentz Physics Department Brooklyn College of CUNY SrGOR LYN s New York 11210 U.S.A. Dear Professor Ferentz, Enclosed you'll find our answer to the UNIX poll. We appreciate the entry on “need any packages or drivers ete”. May we assume that all forms will become available to all: installations? If so, we hope for a response to our wishes, and we'll do our best for other people's needs.. However, we would like to get _soffe hetp_ona shorter term for some of our problems. Thérefore please add this letter to(@ next UNIX newslette: orextract our desires as a classified ad: pee te words)+ address”. x ho has support for’ data areas shared between’ processes ~ (on 11/45); who ha&s support for DECnet or other coupling to remote RT-i1 snstaliations- fie : Thank you very/ much for all your UNIX efforts. Ve H.J. Thomass, ane at ; HJT/ab Enclosures ed 9 D2 a a ¥ Waterloo, Ontario, Canada NA GL University of Waterloo Computer Systems Group 519/885-1711 June 11, 1976. Professor Melvin Ferentz UNIX University Users Society Department of Physics Brooklyn College of CUNY Brooklyn, N.Y. 11210. Dear Professor Ferentz: Here at the University of Waterloo we are faced on jes between UNIX and cther PDP-11 operating arasenn: (1s Most DEC systems have a utility program called a d from the DOS-1] file structure. Is it possibl {lable for transferring files between WIX 1d tell me whom to contact, problem cf moving fi RSX-11D, RT-Il, etc)- to convert these files to an that there is such a utility ava and DOS-l1 formats? If so, maybe you cou Set vipp algbore cwthis plusletey , Yours aa? T.A. Wilkinson TAW:ed University of Technology LOUGHBOROUGH LEICESTERSHTRE LEt1 3TU Tal: ojo9 43171 Telex 34y19. Telegrems Techeslogy Loughboroagh DEPARTMENT OF COMPUTER STUDIES Head of Department o Profesor D. J. EVANS DWI/ JMB 12th April, 1976 Prof. Melvin Ferentz, Physics Department, Brooklyn College of CUNY, : Brooklyn, N.Y. 11210, U.S.A. Dear Professor Ferentz, Will you please enrol us as a member of the UNIX users' group. I should be extremely grateful for any available literature on UNIX that you could supply me, back issues of newsletters, program write ups, etc.... In particular I should like to contact anyone who has a ‘standard’ BASIC or BASIC-PLUS interpreter ‘available, or who has interfaced D.E.C. software onto UNIX (perhaps through an RT-11 or DOS-il emulator}. In addition I have heard rumours that Steve Bourne ia implementing ALGOL 68C at Bell, do you have any further information? If there is not a U.K. UNIX sub-group in existence then we should be very willing to initiate such a group. For your information our UNIX system will be running on a PDP 11/40 with - 60K words of memory, 2 RKOS disc, a dual floppy disc drive (D.E.C. model - anyone with @ acivert 7) abo: 5 eletypes and V.D.U.s, a teletype 40 R.O.P. printer as 2 eprinter and varioug odds and ends. It is used for undergraduate eines postgraduate resdarch. We are currently running the 11/40 under RSTS V4 - hence the desire fbr BASIC or BASIC PLUS - and hope to cut over to/UNIX completely in October 1976. 7 sf Software t we hope to implement under fx (timescale unknown at present} include the DECuet protocols, a logic patch-board simulator, possibly a LISP interpreter, a Computer usage accounting package, UNIQUE - a machine independent JCL soapiler for use on networks, afloppy disk driver, graphics packages, etc. eae Looking \forvard to bearing from you. : a sinceret: Teel foe D.W. TAYLOR Ofr.) S OEPARTMENT OF UNIVERSITY OF SASKATCHEWAN . COMPUTATIONAL SCIENCE March 4, 1976 Professor Melvin Ferentz, Department of Physics, Brooklyn College of CUNY, Brooklyn, N.Y., 11210, U.S.A. Dear Sir: I have enclosed two items which may be of interest to UNIX users. ei To our knowledge Bell Labs is the only other instal- lation which has the hardware necessary to run the Graphics pe: Satellite described on the next page. Therefore, we do not plan to distribute the software. since it fs unlikely anyone else can use it. Sincerely, PA Had, P.A. Hardie, . Manager, Mini-Lab. PAH/ gw encl. . A Graphics Satellite for the UNIX Time-Sharing System The satellite hardware consists of a PDP-11/20 with 16 Kwords of memory, a PC 11 paper tape reader/punch, a Tektronix 611 storage tube with DEC AAI1-D interface and a console teletype. Data and control information pass bidirectionally between this satelTite and a PDP-11/40 system (running UNIX) over paired DLII-E serial and DR11-C parallel interfaces. From a logical point of view the PDP-11/20 can be bootstrapped into satellite mode by loading a small paper tape program which establishes some preliminary nandshaking with and subsequently accepts a core load from UNIX. Thereafter UNIX can access and manipulate the 11/20 paper tape, console teletype and graphics equipment. In the latter case, a rather complete "display processor” was designed such that UNIX can send high-level "display files" to the satellite and have these files executed interpretively to draw pictures. The system we designed and implemented has several objectives: i. Two modes of the graphics satellite should be permitted; in one case C programs have previously executed and produced as their output display files which are executed in the satellite in an “off line" 'mode, while in the other a currently executing (in the PDP-11/40) C program can interact in real time with the satellite. In this latter mode the user can view inter- mediate graphics output and return data to the UNIX resident program that presumably contro] further output. At the moment all such interaction is from the console device keyboard or from the 11/20 switch register; in the future we plan to add a “joystick” device for graphical input. di. Given that the DEC AA-1] D storage tube controller is a very restrictive device, it is desirable that none of its weaknesses are reflected back to UNIX programs. This means that since the satellite has little else to do while pictures are being drawn, as much of its processing power as possible fs brought to bear on making the graphics interface to UNIX as high a level as can be attained. : fii. Since the satellite communicates bidirectionally with UNIX, other devices such as the paper tape reader/punch are made available to UNIX as well. ? ce iv. The satellite console teletype should appear identical to any other UNIX terminal. When no graphics work is being done the satellite console device thus becomes just another UNIX terminal. By coincidence the designers of UNIX seem to have a similar (from the hardware point of view) satellite. Their system however did not meet the above objectives to our satisfaction and thus we have implemented a completely different system. We must emphasize that we did not go out and purchase the above equipment specifically for this purpose - the hardware simply happened to be available and we chose to iniplement a better graphics system on it. If one was starting from scratch there are better choices of hardware, particularly for displays and/or display interfaces. Initially our system used a serial link to UNIX exclusively, and it proved itself over one term of an undergraduate course in computer graphics. More recently we have coupled the pair of parallel interfaces and are currently re-implemen- ting the system such that only the console teletype communication is carried over the serial line and all other communication (including the bootstrap of the satellite) is mediated over the parallel Vink. ; Floating Point on the PDP-11/40 A trivial modification to the floating point interpreter (fptrap) can significantly reduce the amount of system cpu time consumed by programs containing floating point instructions. The modification adds a “look-ahead” capability to fptrap such that, if after interpreting one instruction it finds that the next one is also floating point, then it interprets this next one too without having to go through the illegal instruction trap mechanism. An editing session which installs the two necessary additions-in fpl.s is shown below. Zed fpl.s /badins/a lookahead: add $2, spc / increment pc mov ssp, sp sub $8., sp / reset stack clr trapins / clear error indication Jretz/a tst trapins / error? bne 1f / yes. reset registers, then bomb. mov *spc,r5 / check to see if next mov rs, r4 / instruction is also floating bic $7777,r4 / point and if so go cmp $170000,r4 / back to lookahead to interpret jeq lookahead / it too before exit from fptrap. 1: w 4 Some tests,with C programs indicate that the system cpu time can be cut by at least 75% for very heavily cpu bound programs. The amount of time saved is naturally-deépendent upon whether the floating point instructions in a@ program occur consecutively. Ae Pals Had Peter Hardie, Manager, Mini Lab aywintsa TIAA DD IAYINA THE HEBREW’ UNIVERSITY OF JERUSALEM COMPUTER SCIENCE LABORATORY IMSTITUTE OF MATHEMATICS June 14, 1976 Professor M. Ferentz Physica Department Brooklyn College of CURT ti xed! Brookl k 12210 Wasco Brie Mom over hives pallor (Hardloeve) Dear Professor Ferentz, 4) August Vourle ty. We here at the Hebrev University have been actively working on UNIX for the past mooth or sc. We have an 11/45 with 48 K core and (temporarily) only ope RKOS disk. We do not have floating point, but we have a programmable clock. Unfortunarely, we have been unable to locate an intermittent but wery persistent bug which crashes the system and often messes up the file sys- tem. In the hope that you might know of someone who has run into this problem, I will try to describe it. The prevalent cause of crashes is an incorrect value of Register 5 in U.URSAY(1) when a program ie “returned to" from SWICH, causing wild transfers. We put in checks in SAVU and RETU which indicate that correct values are stored initially, but later they ate somehow altered. We have as yet been unable to ascertain whether the problem is related to swapping. Usually tha incorrect RS value 1s actually the SP value {i.e. U.U_RSAV [1] =~ U.U_RSAV {0]). fs Although the problem is very persistent (example: about 90% of C compil- ations using CC either fail or crash the system) virtually any change in the system is likely ro influence its manner of interference (1.e. programs that once worked reliably start to fail or vice versz). Incidentally, our, system works faultlessly on a nearby 11/45 with more ‘memory. At first we helieved our hardware was at fault, but after a full two days of installation’tests, the NEC man pronounced our hardware healthy. of course ve can’t be sure but ve believe ve have oftware timing bug driven by our particular figuration. — oe > If you have! any suggestions for « course of action we vould greatly eppreciate them. Also if you could refer us to anyone who might knew about our problea ve would be very heppy. We continuing our eearch for this bug but will /amxiously avait your reply. Help!! Sincerely, Toe Lief 4 Moel Kropf David Karach B Reaistrar's office Foom 4) "ssg» a Yale "niversity : New Haven, Conn, 06520 June 16, 1976 . Prot. Melvin Fere n yayeies Dept. te Tooklyn Colleae Brooklyn, N.Y.) PEN CONS tl21¢ Dear or, Ferentz: T wonder if m ’ ost lowing useful tactic: rae users Rave dls, feeding tree the D 7*50 we can Clear ALL Set top to edd, at the stare o¢ OC "ated space or use ton=sbrk(0) execution int ton; = extern char *allocs S(2r¢ extern char **alloco, ttalloce; “* Throw awa ‘ Y al atLace and shrink the core trae Mee pbseke Clearali() ! allocs(o} Salloc SCl)3 allocs(}} Sallocs(0}; RPISEE = Galloesti); = &allo ; brk(top); She w I hope tnis ro ' Feed utine wlil neip s n several Programs, with no nian oF ease: ens, Ver truly yours, HARVARD SCIENCE CENTER HARVARD UNIVERSITY CAMBRIDGE, MASSACHUSETTS 02136 June 24, 1976 Prof. Mel Ferentz Physics Department. Brooklyn College of CUNY Brooklyn, New York 11210 Dear Mel: The Science Center is upgrading its present computer system which runs UNIX from an 11/45 to an 11/70. As a result we wish to sell the i ae } Present replacenent value (1) 11/45 CPU with memory management KT1IC {serial number 1147} (2) Hardware bootstrap . @) KWL - line frequency clock (4) DL11 - single asynchronous serial line interface (5) 24K non-parity DEC core (6) FP11B floating point processor . ) 96K non-parity core - Cambridge Memory Ec Expandacore 11 13, (8) S04 controller only for fast swapping disc 5 $35,800 Iteas 1 to 6 are to be sold as. a package. Items 7 and 8 could be sold Separately. All DEC equipment was purchased 7/1/74 and has been under maintenance contract gince installation. Bids would be acceptable, but a reasonable price would be in excess of 50% Of the present replacement value. Availability, FOB Science Center, Harvard University, will be 9/1/76. Inquiries should be made of Lew Taw, 617-495-2627. Would it be possible to insert this information in the next UNIX news ae 3 "i _ ( : ; 6-10 Prof. Mel Ferentz June 24, 1976 page 2 Printing and shipping of the UNIX documents seems to have gone quite well — we have ordered over 200 Programmers Manuals and 170 Documents. Most of these have already been shipped. It would be a good idea, I think, to include a second order form in the next UNIX News and I will place a second order with the copy service one month after the date of issue of the News. : Sincerely, fe Lewis A. Law Director of Technical Services _ TAL: nds Paga 2 of 2, af! Page 1 of 2 1 \ “) 7 Me. ‘aa ' BELL CYSTEM SOFTHARS_PRESE ‘ . ‘ BELL CYSTEM SOFTMARD PRESENTLY AVAILABLE FOR ¢ DELL SYSTEN SOPIVARE PRESENTLY AVATLABIE FOR . EXCSISING FRGL KE 5 ee ¥ EECENGIG FROM WESTERN ELECTRIC COMPANY LICENSING FROM WESTERN BLECTRIC COMPANY “ Program Tape Descrivtfon bndratieal / ; Hardware : : Kini ALTRAN ALTRAI is a languare and system for symbolie Portable . “Program Name Descrintion 3 Hardware commutation on alzevrai iy Highly portable snd 36 dented tee eee eae pero port is a library of Fortran subprograms Portable ‘ a C bendla rery lance prgbiens Win considerable i fee Meee Pee rreeabe nunsoee aris we " is wa 2 3 3 4 re = : computers, and is characterized by care- AAPAC Facilitates the analysis of parametric TBM 360 ful error checking, and dymmic storace test data on interrated circuits. Wafter Honeywell, allocation. ‘ograms ani statistical ses B . tical analyses ROFF A BIL publications formatting program. Honeywell. FE The input to ROFF is intermixed text €00/60c0 EFLIC A set of r oar Lines which contain the information to end monisaie Pr claerte pe an preuee Tmt 360 he Tormntted and request lines which pitrofiln plotter. ~ a Se ee HoneyweJ1 é contain instructions about how to,formt EESUFF x : the information. Output from ROFF isa ESUFEOX ECUERM is a scientifi or : paginated, formatted document produced Eee ntific program utilizi - se ; * the method of succcrsive vorkadinn tlons.te A 2 y on a time-sharing terminal. inize solutions & Fr s ‘ pele EAGAes nae Aeuiatcna cme STATLIB Ivrary of statistical progrags for rm 360 Solutions are caledirtcd cucces- —- performing statioticel anelysis end * = an Sig? a manner as to provide any computation. selected design param ‘ ‘ BREST ers TROFF, EQN ae Type setting packace which includes text (UNTK #) C COMPILER A version of the C Language Ccupiler which LEM : : . Tommasi and mathematical equation POP UfhG 45 runs on the IPM 360/370 computer line- 360/370 ; ; processing. eS GAPS, ROGAPS. * procrem: a FY . TSO-QED Power and flexible interactive text TR 360. eccrine ° Repke Ait es se rac to's fertabde weitor for evening fn notifying text vestizeting Proposed device ceomctries. OAK bytes . from a time-sharing terminal. ¥ SLOS-ED 1 ible 4 UNIX A general purpose multi-user time-sharing FDP 11/40,45 Shin Tee ead ro Gece Honeywell 1 eae, she systen operates in an inter~ and 70 from 2 tineerharing neaieel nese 600/6000 active made (canmnd-response), receiving ; p pr , ; comands to execute from a teletypewriter = a terminal. MEA Mobile Radio Antenna program performs an ; Tm 360 | i Sie aon. desi function fora seven - ~ aMote: Type setting package runs under - reule 7 — : UNIX. E . % Yr array antenna. Text formatter requires Phototypesctter equipment. 5/13/76 . 6-1