@device(imprint10) @make(report) @definefont(Userfont, 1=, 2=, 3=, 4=, 5=, 6=) @modify(CopyrightNotice, Fixed -1 inch, Flushright) @Modify(Titlebox, Fixed 1.5 inches) @Style(Font computermodernroman11) @Style(spacing 1,indent 0) @Style(leftmargin 1.0inch) @Style(Linewidth 6.5inches) @begin(Titlepage) @Begin(Titlebox) @begin(MajorHeading,Flushright,Font Userfont, FaceCode 1, below .2inches) Project Athena Newsletter @end(majorheading) @begin(majorheading,flushright,font userfont, facecode 6,above 0) December 3, 1984 Volume 1, No. 3 @end(Majorheading) @End(Titlebox) @CopyrightNotice(Massachusetts Institute of Technology) @end(titlepage) @newpage @blankspace(5 inches) @begin(Heading,flushleft, font userfont, facecode 6) Editorial Policy @end(heading) @begin(quotation,indent 0) The @i[Project Athena Newsletter] is the official publication of Project Athena, a five-year experiment in the use of computer technology to improve the education of students at the Massachusetts Institute of Technology. The @i[Newsletter] prints articles of interest to members of the MIT community who participate, or who are interested, in the Project. These articles present the general purpose, philosophy, technical development, and direction of Project Athena, and specific news items about Project Athena facilities and projects. We encourage article submissions from the community and publish appropriate contributions whenever possible. @end(quotation) @tabset(.5inches) @\If you have any comments or questions on the newsletter-- @tabset(1.5 inches, 2.6 inches) @\Contact:@\Will Doherty @\@\Editor @\@\Project Athena Newsletter @\@\MIT E40-426 @\@\Cambridge, MA 02139 @\@\(617)253-1300 @newpage @set(page=1) @Begin(MajorHeading,Flushright, below .2inches) Project Athena Newsletter @end(MajorHeading) @begin(majorheading,flushright,font userfont, facecode 4,above 0) December 3, 1984 Volume 1, No. 3 @end(Majorheading) @pageheading(immediate,left "DRAFT", center "DRAFT", right "DRAFT") @Pageheading(left "12/3/84", center "Project Athena Newsletter", right "12/3/84") @pagefooting(odd,Immediate, left "Project Athena Newsletter", center "Volume 1, No. 3", right <@value(page)>) @pagefooting(even,Immediate,left <@value(page)>, center "Volume 1, No. 3", right "Project Athena Newsletter") @Section(Student Center Athena Cluster Opens) @i[Eva Tervo, Operations Coordinator] You may have noticed the recent flurry of activity in the library on the fifth floor of the MIT Student Center. A huge new Athena cluster opened in W20 in November. Containing five new Unix systems, one server machine, over 30 terminals including many of the new large-screen VAXstation 100s (VS100), in ????? square feel of space, the Student Center cluster is the largest facility Athena will open. The MIT student body donated the space for the cluster to Athena. The machines in the cluster are named after characters in the Larry Niven novel @i(Ringworld). The server machine is @i(ringworld). The user machines are @i(nessus, louiswu, speaker, teela, and prill). This breaks a short-standing Athena tradition that named Athena machines after characters in Greek mythology. Although the original five cluster machines will retain their Greek names, we were beginning to run out of "good" names like @i(hera) and @i(apollo) and into the obscure and unpronounceable names like @i(mnemosyne) and @i(pasiphae). We think the new heuristic will provide a fertile and fun "name space." To find the Student Center terminal room--enter the 5th floor library and turn left at the copy machines. The terminal room is open 24 hours a day, except when the Student Center is closed for the holidays. If you obtain an account at the Student Center facility, please be aware that: @begin(itemize, spread 0) You will have to show your MIT student ID at the door. You should not bring food into the cluster (as for all Athena clusters). @end(itemize) Students will compete for the limited resources available; thus, you should not play games on the only terminal available when others need to do course work. Don't be shy if you need a terminal for course work -- speak up! @Section(Students: Accounts Move To Student Center) @i[Cecilia d'Oliveira, User Services Manager] Students currently enrolled in classes using Athena resources should be aware that their accounts will become inactive on January 15, 1985. These students must therefore move any files or data they wish to keep from the current course machine to the disk space allocated for them in a new account on one of the Student Center machines. We are moving current accounts before this spring, because students enrolled in spring courses will need the course account resources present students now use on the course machines. We will establish accounts for people no longer in courses using Athena resources on the machines in the new Athena cluster in the Student Center Library (W20-5th floor). These new accounts will be available early in December so that students can transfer files or data to new accounts before their existing accounts become inactive and before files stored in their current disk space allocation are deleted. The new accounts in the Student Center will allow students with existing accounts: @begin(itemize, spread 0) Continued access to Athena resources in the Spring Term Limited storage of data and files created this term @end(itemize) Project Athena will set up new accounts in the Student Center automatically using the student usernames assigned this fall. We will post machine assignments, probably sorted by last name, in the Student Center facility. The new accounts will have limited disk storage of perhaps 0.5 MBytes. Please note that individual students must move the existing files they want to keep to the new account. You can either: @begin(quotation, indent 0) Transfer your files over the Athena network using the new @i(moveme) command that we will make available. @end(quotation) Or: @begin(quotation, indent 0) Download the files from the existing account to a floppy disk using an IBM PC/XT and the Kermit file transfer program, then upload the files to the new Student Center account at some later date by reversing the process. (Students who have access to a PC/XT may find this a reasonable alternative.) @end(quotation) You will find information on both approaches in all Athena terminal rooms by December 7. Student consultants will be available in the Student Center from 7-9PM on Tuesdays, Wednesdays, and Thursdays during December to provide assistance. @Section(Student Center Cluster: Holiday Schedule) @begin(format) @tabset(3inches, 5inches) Monday, December 24, 1984@\Closed at 6pm@\Christmas Eve Tuesday, December 25, 1984@\Closed all day@\Christmas Day Wednesday, December 26, 1984@\Reopens 7am Monday, December 31, 1984@\Closed at 6pm@\New Year's Eve Tuesday, January 1, 1985@\Closed all day@\New Year's Day Wednesday, January 2, 1985@\Reopens at 7am @end(format) @section(IAP 1985: Project Athena Activities) @i[Connie Donaghey, Athena IAP Coordinator, E40-443, 3-1300] @begin(format) @b[Project Athena Seminar Series] Tue. and Thu., Jan. 8-29, 3:30-5pm, 34-101 @end(format) Project Athena is a large-scale educational experiment aimed at developing and appraising ways in which convenient access to networked computer workstations can influence the structure and content as well as the teaching and learning associated with the educational programs of the Institute. Speakers will discuss educational software projects now underway at MIT that use Project Athena resources. Contact: Connie Donaghey, E40-443, (25)3-1300. @begin(format) @i[Project Athena Status Report] Prof. Steve Lerman Director of Project Athena Tue., Jan. 8, 3:30pm, 34-101 @i[Computer-Aided Electromagnetic Field Instruction] Prof. James Kirtley Electrical Engineering and Computer Science Thu., Jan. 10, 3:30pm, 34-101 @i[Computer Applications in Transport Phenomena] Prof. Julian Szekely Materials Science and Engineering Tue., Jan. 15, 3:30pm, 34-101 @i[Computer-Based Software for Foreign Language Instruction] Dr. Janet Murray School of Humanities and Social Science Thu., Jan. 17, 3:30pm, 34-101 @i[Computer Enhanced Curriculum for Fluid Mechanics] Profs. Earll Murman and Sheila Widnall Aeronautics and Astronautics Tue., Jan. 22, 3:30pm, 34-101 @i[Computer-Based Course in Modeling, Research & Design] Dr. Andrea DiSessa Educational Computing Group Laboratory for Computer Science Thu., Jan. 24, 3:30pm, 34-101 @i[Playing and Analyzing Sequential Prisoner's Dilemmas] Prof. Hayward Alker Political Science Tue., Jan. 29, 3:30pm, 34-101 @end(format) @begin(format) @b[Documenting Your Athena Project] Linda Merims, Athena Documentation Manager Wed., Jan. 23, 10-12am, 34-101 @end(format) This seminar is for people who are developing educational software as part of Project Athena and who know that--somewhere along the line--they will need to document it. The seminar has two main purposes: to provide a kit of short, practical tools for documentation, and to find out what documentation help you want from Athena. Specific topics include: documenting the internal workings of your programs for programmer maintenance; putting online help in your system; how to make a man page; and what ought to be in a document for the users of the system. Contact: Linda Merims, E40-431, (25)3-1368. @section(Consultant's Corner: When You Run Out of File Space) @i(Win Treese, Athena Student Consultant) Did you ever get the message: @blankspace(1line) @center( @b[WARNING - too many files (/mit)]) Or: @center(@b[WARNING: disk quota (/mit) exceeded]?) @blankspace(1line) Both of these messages warn users about limits on the disk storage available to them. Computer disks provide file storage on Athena machines. Since a lot of people share Project Athena machines, users must have a @i(quota), or portion, of the total disk space they can use. To find out what your quota is, type @b[quota]. You should see something like this: @begin(ProgramExample, font userfont, facecode 5) % quota disk quotas for student (uid 460): Filsys current quota limit #warns files quota limit #warns /mit 907 2500 3000 172 600 800 @end(ProgramExample) @blankspace(1line) The quota system limits two things: the number of disk blocks you may use and the number of files you may have at a given time. The first column of the example above shows the file system partition your account accesses, in this case @b[/mit]. In the second column, you find the number of disk blocks your files occupy. A disk block is a unit of storage on a system disk; in this version of UNIX, it is 2048 bytes. Under the first @b[quota] column is the ``soft'' limit on the number of disk blocks you can have. If you try to use more, and exceed the ``soft'' limit, you will get the message: @blankspace(1line) @center(@b[WARNING: disk quota (/mit) exceeded]) @blankspace(1line) You may continue to work, but you should try to remove some of your files to get under your quota before you logout. The next column, the first @b[limit] column, is the ``hard'' limit, the maximum number of disk blocks you can use. If you try to create more files at this point, you will get the message: @blankspace(1line) @center(@b[disk LIMIT REACHED (/mit) - WRITE FAILED]) @blankspace(1line) The program trying to write a file will be unable to do so and you may find yourself stuck in the program having lost some of your work. Essentially, the difference between the hard and soft limits gives you some space to use temporarily while you are logged in, but you should restrict yourself to the soft limit by the time you log out. The next set of numbers in the example gives information about the number of files you may have. Under the @b[files] column is the number of files you have. The columns @b[quota] and @b[limit] give soft and hard limits, respectively, on the number of files you may have, just like the soft and hard limits of your disk quota. If you exceed the soft limit, you will see: @center{@b[WARNING - too many files (/mit)]} @blankspace(1line) and you should get rid of some of your files before you log out. If you exceed the hard limit, you will see: @blankspace(1line) @center{@b[FILE LIMIT REACHED - CREATE FAILED (/mit)]} @blankspace(1line) and you may lose work and find yourself stuck. NOTE: If one of these errors occurs while you are using @b[emacs], the message: @blankspace(1line) @Center(@b[?Display size too big for terminal type "filename"]) @blankspace(1line) will appear at the top of the display. Admittedly, this error message is not very informative. You can do several things to correct the problem if you get one of these messages. Usually, you can delete some really large files that you don't need to keep around. Files you can delete include: @begin(itemize, spread 0) @i(core), a copy of the memory made when certain errors occur Your deleted mail messages, located in your @b[~/Mail] directory, acquire a @b[#] symbol when you @b(rmm) them. For example, @i(2) becomes @i(#2) when you type: @center(@b[rmm 2]) The system purges the @b[#] files periodically, but you can delete them yourself if you need the space. @end(itemize) It's always a good idea to do an @b(ls -l) to look for any really large files. For example, @blankspace(1line) @begin(ProgramExample, font userfont, facecode 5) %ls -l total 2086 -rw-r--r-- 1 student 2121728 Nov 5 17:08 core -rw-rw-rw- 1 student 2148 Nov 5 16:57 corner1.mss -rw-rw-rw- 1 student 1702 Nov 5 16:52 corner1.mss~ -rw-rw-rw- 1 student 0 Nov 5 17:08 log -rw-rw-rw- 1 student 324 Nov 5 16:18 quotalog -rwxrwxrwx 1 student 5120 Nov 5 17:08 test* @end(ProgramExample) @blankspace(1line) This listing shows that the file @i(core) is very large, taking up 2121728 bytes. It also shows the file @i(corner1.mss~), which is a backup file created by EMACS. If you are finished editing a particular file, you can remove the files ending in @b[~] to gain some space. If you have a legitimate need for more disk space, even after you clean up the extra files in your directory, you may want to contact your Account Administrator to increase your quota. @section(The Project Athena Staff) @i[Linda Merims, Documentation Specialist, and Will Doherty, Newsletter Editor] One year ago, Project Athena had no director, no offices, no faculty curriculum development projects proposed (much less funded), only one computer, and a few employees. Athena's early course was charted by a collection of faculty committees. Jim Bruce, MIT's Director of Information Systems, was our Acting Director. He brought together a small staff to translate the direction of these committees into a concrete plan of action. Doug Wilson, captured from the Joint Computer Facility, was "Systems and Operations." With the help of one staff programmer (Mike Gretzinger) and one student programmer (Jim Fulton), he built and equipped the first few Athena clusters. Cecilia d'Oliveira began as Assistant to the Director. She handled all the non-technical aspects of getting the project off the ground: How do we start course development projects? What will we do for staff? And for software? Beth Anderson planned user support. Ed Balkovich of Digital Equipment Corporation and Rip Parmlee of IBM represented their companies in the planning process. There was one Unix "wizard" in residence, a systems programmer from DEC. Project Athena now has over thirty full-time staff from MIT, Digital, and IBM. Our administrative offices are located in Building E40 on the east side of campus. We have six clusters open, with some 45 Digital VAX/750's and over 100 IBM PC's in use. Over fifty faculty curriculum development projects are underway. This fall the first one thousand MIT students are taking courses that teach with the aid of Athena resources. (Even with all this progress, we are technically still behind schedule.) Project Athena staff members now divide their work hours between administrative activities, user services, software development, and operations. A diagram of the staff organization appears on the last page of this @i[Newsletter]. Project Athena Director Steve Lerman, a professor of Civil Engineering at MIT, guides the administration of the project, assisted by Associate Director Rip Parmelee from IBM and Associate Director Ed Balkovich from Digital. Jim Bruce now serves as head of the Athena Administration/Operations Committee and actively confronts management issues related to the administration, operations, and networking of the project. Professor Jerry Saltzer of the Laboratory for Computer Science has just joined Athena as Technical Director. Assistant Director Cecilia d'Oliveira heads the User Services group that provides consulting and documentation services for Athena users and curriculum development projects. Within ``User Services,'' Beth Anderson coordinates a dedicated team of some 40 student consultants, crowned with owl-insignia caps, who provide on-the-spot help in the terminal rooms. Applications consultant Ed Moriarty devotes his technical expertise to current courses, while application consultants Mark Levine and Jennifer Pinkus work with the development projects. Linda Merims organizes the documentation of the Project. Doug Wilson is Athena's Manager of System Development, heading a team of system programmers, employed by Digital (4 programmers), IBM (4 programmers), Bolt, Beranek, and Newman (1 programmer), and Mike Gretzinger, Ron Newman, and Joachim Martillo are MIT staff members. Project Athena is in the process of recruiting a permanent Manager of Operations, but in the meantime, Doug Wilson acts as Manager of Operations in addition to his system development position. He directs Operations Coordinator Eva Tervo, System Programmers Dave Grubbs, Bertha Hoskins, and Barbara Pease, Technician Larry Peatfield, and User Accounts Administrator Connie Correia. Digital Field Service personnel aid in installation and service of Digital equipment. Fifteen student operators help Eva Tervo with operations tasks, like machine installation and maintenance, that range over the entire campus. Codex, a subsidiary of Motorola Corporation, has donated the services of a manager of the Codex Networking Group, along with two System Programmers who work on the network. Jeff Schiller and Dennis Baron, both of whom work in Telecommunications Systems of MIT's Information Systems department, are telecommunications experts who set up communications for the Athena network. Administrative Secretary Connie Donaghey and Secretary Beverly Pottinger provide important additional administrative support for the staff and committees of the Project. @section(New System in January--Many Changes) @i(Linda Merims, Documentation Specialist) During the first week of January, Project Athena will release a new version of its system to user machines. This release will contain several major new software systems, upgrades to some existing packages, and many minor improvements. A few of these changes, particularly the new Scribe, will require you to change some things that you have been used to doing because the new version is not fully "upward compatible" with the old version. This article will describe the new software and changes, and will tell you where you can find more information. @subsection(Schedule) We plan to release the new system during the first week and a half of January. Athena now has some thirty-five user machines spread across five user clusters. Because there are so many machines, the release will not occur at the same time on all systems. Instead, the release will be made as a kind of "rolling blackout" from cluster to cluster. The "blackout" is scheduled to begin on January 2, 1985, and last until about January 10. The planned order of conversion for the clusters is W20 (Student Center), 66, 38, 1 (includes 4 and 6), and 11. You can tell that your cluster is being done because there will be a sign on the terminal cluster door. If you dial-in, your machine will not answer. We expect, in the best case, that each cluster will be out of service for about one day, from about 9 a.m. to 9 p.m. If we encounter difficulties, the cluster may be down longer. This is more likely to occur in the first clusters we convert than in the later ones. You can tell that the new system has been installed because the banner you get at after logging in will say something like "Athena Unix #2-something" instead of the "Athena Unix #0-something" or "1-something" it displays now. If you have problems with the new system or its software, tell the student consultant on duty. If no consultant is available, <> Please, do NOT call the main Athena administrative number. @subsection(Caveat) As of this @i(Newsletter's) deadline, not all of the features mentioned in this article had completed final testing. Some may be found wanting and will be withdrawn before the actual release. The January issue of the @i(Newsletter) will report any such last-minute deletions. @subsection(Hanging Up Will Log You Out) The system will now automatically log you out if you hang up your phone when dialing in to Athena. This will solve a major problem we have had with dial-ups, namely people hanging up without logging out because they assume (or perhaps just fervently hope) that they will be logged out automatically. They aren't. Instead, people who dial-in later frequently find they have acquired somebody else's leftover terminal session. The system will also now force a hang-up signal upon logout. This will solve another, subtler problem wherein people logout but their modems keep the phone connection to Athena, tying up lines unnecessarily. You must still explicitly logout of hardwired terminal sessions. Turning off a hardwired terminal will still NOT log you out. @subsection[Flow Control (CTRL-S) Comes Back On] Project Athena turned off flow control on its network when we installed the Login Server in September. (The "Login Server" is the program that asks you what machine you want to talk to on both dial-ups and terminal cluster terminals and then connects you to it through the network.) "Flow control" is a protocol that helps slow links in the computer network--such as graphics terminals and people trying to read screens--regulate the rate at which the fast links in the network send them data. Its usual manifestation is the control-S character to pause output, and the control-Q character to start it up again. On VT100-type terminals, the NO SCROLL key can perform the same function. We turned off flow control because the Emacs text editor uses CTRL-S as its search command. The ambiguous situation could arise that a person sitting in the Emacs editor would type CTRL-S to search for a character string, and the network would intercept the CTRL-S before it got to Emacs, interpret it as "turn off output" and the person would find that his/her terminal seemed to have gone dead for no apparent reason. To avoid this, Athena told its network to ignore all CTRL-S, CTRL-Q flow control commands. This caused a different set of problems. Graphics terminals that need flow control would garble their graphic output if used over the network. People who had been transferring files between Athena and their micros found the process no longer worked reliably. People sitting at terminals typing CTRL-S/CTRL-Q or NO SCROLL would find that, although it appeared to be working, they were actually losing large chunks of data out of the middle of their files. Flow control would only work if the terminal a person was using was actually hardwired to the machine he/she was logged into. We have fixed this problem now, removing the possibility of ambiguity. When the new system goes up in January, flow control will come back on. In the meantime, if flow control is vital to your application, as it is for all graphics, you should always use a terminal whose Login Server banner shows it is hardwired to the system you actually want to use. @subsection(CTRL-S, Emacs, and VT125, VT240, and VT241 Graphics Terminals) Although Athena has fixed the general problem "who gets the CTRL-S" between Emacs and flow control for January, a related problem will @p(always) remain for graphics terminals. Graphics terminals tend to like to grab CTRL-S's themselves and interpet them as "stop sending output to the screen." This causes additional confusion when using Emacs on an Athena Graphics terminal. (VAXstation 100's and Pro 350's do not have this problem.<>) When using Emacs on a VT125, VT240, or VT241 terminal set-up for graphics, follow these tips: @itemize[ NEVER enter CTRL-S while in Emacs. If you do, your terminal will seem to have gone dead. Emacs has equivalents for all its commands that contain CTRL-S and CTRL-Q characters. These were published in the chart on page 5 of the November 84 Athena @i(Newsletter). We will also paste these charts on all graphics terminals. If you do enter CTRL-S while in Emacs and your terminal seems go dead, try to start it up again with CTRL-Q. This will almost always work on VT240 and VT241 terminals. Sometimes it will not work on VT125 terminals. If a VT125 seems to have gone dead, try SET-UP RESET. If that doesn't work, you will probably have to kill your Unix session from another terminal. ] <> @subsection(The Electronic Laboratory Notebook: RS/1) The January release will include RS/1, a major new software package. RS/1 is a powerful, integrated system for data management and analysis developed by BBN Research Systems. It is basically an electronic laboratory notebook designed to meet the information-handling needs of research scientists, engineers, and students of these diskiplines. Based on a simple English-like command language, RS/1 enables you to enter data into two-dimensional tables, and then visualize this data easily in graphs, perform a variety of statistical tests, and create models of complex systems. Users can extend and tailor the RS/1 system with RPL, the PL/1-like Research Programming Language that comes with it. RS/1 has an extensive online help facility. The package is part of the resource donation BBN has made to Project Athena. Along with RS/1, Athena will be installing RDE, Research Data Entry System, a forms oriented software system that makes it easy to enter, edit, review and report data. <> RS/1 is documented in a three-volume User's Guide. The Guide will be available for sale at the IS Publications Office in 11-209 toward the end of December. The three-volume set (not sold individually) will cost about $50. RS/1 also has an easy pocket reference card that sells for about 50 cents. The RDE system is documented in a separate User's Manual, that will cost about $20. @subsection(New Version of Emacs with Elisp) We will install a new version (162.45z) of the CCA Emacs text editor. The following is a list of the major new features and changes: @itemize[ This new version has one significant incompatibility with the old Emacs. People who have created their own .emacs_keys files are the only people truly affected. All of the old Emacs functions that were also bound to keystrokes began their names with ^R (control-R), for example, ^R Forward Character. The reason for this is an obscure historical artifact dating back to the origins of Emacs that can be summed up by saying, "It seemed like a good idea at the time." CCA has removed the ^R from the names of @i(all) Emacs functions. The only people who will really notice this are people who have created their own custom key bindings in a .emacs_keys file. Those keys files will no longer work. You must remove the ^R prefaces from all of your key binding definitions for the new Emacs. @multiple{ The long-awaited Elisp extension language is included in the new version of Emacs. Elisp is an interpretive programming language, a subset of Common Lisp, that lets you write Lisp programs that manipulate files of text wherein the functions that you call are the familiar Emacs commands. It is like getting a programming language with a very rich, powerful set of subroutines for manipulating character data. The usual application of this is to program your own Emacs super-commands instead of using the more clumsy keyboard macro technique. However, Elisp can be invoked outside of Emacs at the shell level and can be used as a general text file programming tool. The Elisp manual will be available at the IS Publications Office by early January. It is up to the usual standard for Emacs documentation. However, the manual cannot stand by itself. In order to use it, you must also have the book @i(Common Lisp) by Guy L. Steele, Jr. We are trying to get this book stocked. } The new Emacs supports file version numbers. This feature is turned off by default. Emacs now has automatic file name completion. This means that when you enter a filename in response to a prompt (for example, CTRL-W Write File:), you don't need to type the whole name of the file. Instead, you can just type a few letters then the ESCape key, and Emacs will find the file that matches your pattern. C Mode has been changed to make it easier to modify the way Emacs automatically indents your C programs. See the C Indent variable documentation. You no longer have to type CTRL-U to preface a numeric argument to a command. Auto Arg Mode (on by default) will take any number you type that is followed by an Emacs command and interpret it as a numeric argument. For example, 45 CTRL-F will move forward 45 characters. You used to have to type CTRL-U 45 CTRL-F. Emacs now starts up somewhat faster, and scrolls to new pages considerably faster. ] <> There are other, minor changes. The new, updated Emacs manual will be for sale in the IS Publications Office by early January. Because eliminating ^R changed virtually every page in the manual, CCA will @i(not) publish update packets for old manuals. Your old Emacs manuals are still mostly accurate if you remember to ignore the ^Rs. @subsection(New Version of Scribe) We will release Scribe Version 4 as the default version of Scribe, replacing version 3C. There are many significant changes in the new Scribe, including one incompatibility that will affect all Scribe users. Incompatibilities @itemize[ The new Scribe has a bug in the way it handles file names. (The bug has been reported to Unilogic, but will not likely be fixed before June, 1985.) You can no longer give your Scribe input files simple filenames such as @i(myfile). If you do, Scribe will look for a file called @i(myfile.mss). It will not look for a file called just @i(myfile). Not finding a file with the .mss file type Scribe will give an Input file myfile not found error message. On the other hand, if you give your input file any name that does contain a . (dot), and you give Scribe its full name, such as @i(hi.there), Scribe will find the file. The simplest way to deal with all of this is to always name your Scribe input files the .mss file type. @multiple{ We will change the default Scribe output device. The default output device type has been "line printer," producing output files with file type .lpt, suitable for printing on a line printer. This normally works all right on both the LN01 laser printer and the LA26 impact printer. It works--until you try to change the right margin in an @i(@@style) command. Suddenly, instead of the text being narrower and leaving a wider right margin, you find the text running all the way off the page. What you didn't realize is that Scribe assumes a line printer is loaded with 132-column wide computer output paper. Your "wider" right margin is now some five inches off the edge of the page. For this reason, and because of other complaints about the way .lpt files print on our laser printers, we have decided to change the default output device to @@device(ln01). This will produce files with a file extension .ln0. (Alas, although the device is known colloquially is an "el en oh one" that @i(is) a zero.) The LN01 is no great improvement over .lpt. There are still no super nor subscripts, no boldface, nor fonts. The device does allow some space at the top of the page before the page number, and has somewhat more pleasing default margins. } Any user-written .mak files that include an @i{@@Libraryfile(math.lib)} line must be changed. @i(math.lib) has been renamed @i(common.lib). The @i(@@file) command is obsolete. Replace it in any .mss files with @i(@@include). There have been numerous minor changes to the Scribe database, each having a small effect on the way some documents will appear. ] New Features Unfortunately, the two major new features in Scribe Version 4 are not too useful to Athena users because we do not have a fully-functional laser printer that can take advantage of them. The new features are: @itemize[ A much-improved, much-expanded facility for typesetting mathematical equations. See Chapter 11 of the Scribe manual diskussed below. A facility, @@picture, for including graphics in the midst of text. This only works on Santec S-700, Imagen Imprint-10 8/300 and 5/480, Xerox 9700 and 8700, Lasergrafix 1200, and DEC GIGI Color Graphics output devices. There are none of these devices within the Athena environment, but there might be elsewhere. Only certain file formats are supported. ] Documentation The complete Scribe Version 4 @i(Release Notes), including a thick list of changes to the Scribe database, will be placed in the Athena terminal cluster manual racks next to the new Scribe Version 4 @i(User Manual) for the sake of Scribe aficianados who want to know the gory details. The IS Publications Office in 11-209 has been selling Scribe Version 4 @i(User Manuals) since September. Unilogic has not yet published Version 4 editions of its Scribe @i(Pocket Reference) and @i(Database Administrators Guide). Unilogic Ltd. itself is offering a unique deal for people with Scribe Version 3 manuals (with the monk on the front) who want to upgrade to Version 4 manuals. If you send your old Scribe Version 3 manual @b(plus $5) to: @display[ Unilogic, Ltd. 160 North Craig Street Pittsburgh, PA 15213 ] they will send you a new Scribe Version 4 manual. Since the new manual normally costs about $20, you will be saving some $15. @subsection(New Version of Franz Lisp) A new version, Opus 38.91, of the Franz Lips compiler and interpreter will replace the existing version 38.79. This version of Franz has two new functions that will make it much easier to edit your Lisp programs. The two functions are @i(emacs) and @i(emacsl). As with the existing @i(vi) function, when you enter either of these functions while in the Lisp interpreter, you will immediately be transferred to the Emacs editor. With @i(emacs), you will simply start an Emacs session without any "current" file. When you exit Emacs, you return to Lisp. If you enter @i(emacsl), you will come up in Emacs with your current Lisp file in the main buffer. Changes to the file will be read back into the Lisp workspace. <> <> @Subsection(New Version of Penplot with QPICTR) Penplot will now support the QPICTR routine that lets you draw annotated two-dimensional graphs in one simple call. This implementation of QPICTR differs slightly from the Joint Computer Facility version of QPICTR. Also, Penplot has a new routine called TERMINAL that acts as a generic output device, causing Penplot to automatically detect the type of terminal that you are on when you compile a program. Lastly, we have fixed some bugs that prevented you from using all of the different line textures that are available. Support for Penplot graphics on VAXstation 100 (VS100) terminals is NOT included in this release. We are preparing a new edition of the @i(Athena Guide to Using the Penplot Graphics Library) that will explain these new features. @subSECTION(Small Changes to mh Mail System) A new mechanism, an @i(.mh_suppress) file in your home directory will filter out the message header lines you specify when you view your mail with the @i(show, next,) or @i(prev) commands. Type @i(man mh_suppress) on the new system to find out about it. The @i(repl) (reply) command has a new option pair, @i(-ccall/-noccall) that regulates whether everyone who has been Cc'd (copied) will @i(automatically) be Cc'd when you reply to a message. The default will be @i(-noccall), i.e., by default, only the people on the To: line of a message will receive copies of your reply unless you explicitly ask that everybody else receive copies. This is to prevent a situation where, because you were copied on the original message, you receive each and every reply to that message, no matter how trivial and occasionally personal, because the replying party forgot to delete you from the message header. You can now tell the @i(inc) command to @i(-noupdate) your current message when you incorporate new mail. (This actually works already, it is just undocumented.) The @i(pick) command has a new option (@i(-rmm)) that will delete the messages it finds from folders. This is off by default. Subfolders are now fully-supported. <> @subsection(New @i(clear) Command) A new version of the @i[clear(1)] command makes it possible to selectively clear just the text or graphics portion of a screen. @i(clear) also clears remote terminals of a different terminal type. @subsection(New Spelling-Checker Program) We will install @i(ispell), a new interactive spelling correction program. @i(Ispell) interactively asks you about spelling anomalies it finds in your files, instead of just listing out words the way the existing @i(spell) program does. @i(Ispell) also allows you to create your own private spelling dictionary to cope with your idioglossary. See the man page for @i(ispell) on the new system. @subsection(/usr File Partition Standardized) This change need not concern most users. Athena will be changing the way it distributes its software to user machines. User software (those programs whose names you can enter as "commands") usually resides under the @i(/usr) branch of the file system tree. When we wished to release a new program (perhaps a bug fix to the Fortran compiler, for example), a programmer would simply copy that new file to all the user machines. But because this was a manual process, because there might be several programmers using different clusters as testbeds, and because there are so many machines, it was very easy for the user machines to get "out of phase". Sometimes it is impossible to tell what versions of what software are running on any given system. Athena will use a new utility @i(rdist) that will periodically check each user machine's @i(/usr) file partition. Using an Athena machine in the E40 building specifically designated to be the "standard system" as its guide, it will discover what software a user machine is missing and copy it over, what non-standard versions are running and copy over standard versions, and diskover what files are there that don't belong and--in the words of our System Manager --"blow them away." @subsection(File Space Partitions, Symbolic Links, and Worm Holes) This change is quite esoteric, yet it will likely be visible to increasing numbers of Athena users. Unix manages its file space as a set of separate "partitions." Basically, one disk partition is one disk drive. Partitions cannot span disk drives; a partition has to fit on one drive. You can see what partitions are defined by entering the command @i(df). You would see something like: @verbatim[ Filesystem kbytes used avail capacity Mounted on /dev/ra0a 7429 6139 547 92% / /dev/ra0d 7429 665 6021 10% /tmp /dev/ra0f 106976 46147 50131 48% /usr /dev/ra0e 146347 29069 102643 22% /site /dev/ra0h 138688 116229 8590 93% /u3 /dev/ra1h 138688 81929 42890 66% /u1 /dev/ra1g 261295 159465 75700 68% /mit /dev/ra2c 424614 14000 368152 4% /u2 ] That last column is the name of the file partition. If you have some familiarity with the Unix file system structure you will begin to realize that what we have just said is, effectively, that all user files (all stored under /mit) are on one disk pack, and most applications software (stored under /usr) are on a different disk pack, etc. The question arises, "What if one disk pack isn't big enough to hold all user files on a machine? What if /mit gets too big? Where will they go?" The answer is that they have to go on a different disk pack (partition). Athena is using /u1 as its overflow pack. It may well be that, say, everybody whose username begins with the letter @b(m) or later in the alphabet will have their files stored on a different disk pack named /u1. New large applications software packages that overflow /usr may also be placed on /u1. Why are we telling you this? Who cares? In the usual run of events you may never notice that this has happened. The notation ~username will always find that user's HOME directory when entered to the shell. /mit/@i(u/s/username) will always find that user's HOME directory when given in a program. (Programs cannot interpret the ~ notation.) This is because we will have partially bridged the gap using a Unix system facility called a "symbolic link" that creates a kind of worm hole from one place on the Unix file system tree where a file "logically is" to where the file "really" is. <> (See @i(man ln) if you are curious.) The problem is that you can be happily crawling around the Unix file system tree and suddenly find yourself not at all where you thought you were. You can @i(cd /projects) to see what projects there are, but if you @i(pwd), you will see that you are really in @i(/mit/projects). This is not so bad. What becomes really confusing is if you @i(cd /projects), take a look, then try to get back to the root level (/) with the "pop me up one level" command @i(cd ..). You won't be in /, you'll be in /mit. This can be very confusing unless you realize that you have taken a ride on a symbolic link "worm hole" to another part of the file system. @subsection(System Built from Sources) This January release of the Athena system will have been built, from scratch, by recompiling the sources to all system software. It was not built by the usual "snapshot copy" method. This ensures that we will be able to duplicate this running version of the system at a later date should this be required. << There are a few other things. I'm not going to put in a newsletter article about this, but consultants at least should be aware that notes will no longer network conversations to other systems within the Athena network. Exchanging notes material was beginning to consume significant portions of CPU time on several systems. Thus, this exchange is being deliberately removed. Does this mean that notes stuff on apollo won't get out to usenet land? I really don't know. I'd have to ask Dave Grubbs. Also, be aware that notes has never been released to users. There are no notes managers on each machine to create or manage discussions. The whole notes situation is a mess. Second, I'm going to put something in about the window manager. It is undoubtedly going to be the most popular piece of software out there, even though it is going out in /usr/unsupported and again, even though it is unsupported, I don't think I can ignore it since users see it in use in every cluster. I don't want to hold up the draft any longer, though, so expect a late supplement to the draft. Then there is tapedd. This program works, and there is now a user-accessible tape drive in W20, but it is connected to the server machine. The distributed filesystem library is NOT going out with this release. Backwards more is not going out. It doesn't work with pipes and hence not in many usual cases, like man. Also has bugs. If you know of any other significant change that's going out, please come and tell me or Will.>> @section(Newsletter Deadlines) @b[The deadline for submission of articles and article ideas for the January 7, 1985 issue is 5pm on Friday, December 7.] Deadlines for this issue is earlier than usual because of the MIT vacations occurring during the next month. @b[The deadline for submission of articles and article ideas for the February 4, 1985 issue is 5pm on Monday, January 14.]