Alan J. Perlis

Preliminary report: international algebraic language

Editors Notes: Although this method is not novel, it has been printed here to summarize for the benefit of a new generation of computer personnel. I t should be noted that : 1) This method seems advantageous if only a few significant figures are required. Otherwise the normal method, Log-Multiply-Antilog, is more desirable and faster in particular for higher order roots. These subroutines are normally required for other purposes anyway and space is not lost. 2) One immediately notices many tricky ways of coding this method for a computer, via looping and the use of tables or converting instructions. Note that, as one proceeds, the contribution of the left-hand term becomes proportionately large enough such tha t it alone might be used within accuracy limits after a certain number of digits are developed. 3) Although the author states that this method used more memory space than other routines, it seems tha t the converse could well be true if advantage were taken of higher order differences in building up the subtrahend. This appears to be a natural method for a 256 memory machine, if it had good indexing and looping features. Remember that An(X ~) = a constant n[

The American side of the development of Algol

History is contextual. The Algol* development was a product, perhaps a miraculous product, of its time. All developments proceed, almost implacably, from the primitive to the rococco, e.g. from Algol58 through Algol60 to Algol68 with an appearance of a large number of offshoots (e.g., JOVIAL, MAD, NELIAC and Euler), extensions (FORMULA ALGOL and LCC), virile offspring (PASCAL), etc., occurring enroute. The earliest developments appear clean, surprising us in the new views they reveal. One frets over why the vision that prompted the beginning weakened during the course of future developments. Perhaps it is inevitable that, as unexpected complexity is uncovered, smoothness, equal value if you will, in solutions cannot be maintained. Trivia cannot be identified easily, special cases overwhelm the search for general patterns, custom and habit move performance into the realm of objective concept, experience warps both intuition and reason, fear of instability burdens insight with caution. The elegance of Algols offspring is a tribute to the grace and power of the original. Algol, a second generation language, was more graceful than any of its predecessors, for example FORTRAN, MATHMATIC and IT.

A visit to computation centers in the Soviet Union

The four authors spent a two-weeks period from August 27 through September 10, 1958, visiting Computation Centers in the Soviet Union at Moscow, Kiev, and Leningrad. The visit was the result of an invitation by the Soviet Academy of Sciences reciprocal to an earlier trip in June to the United States by four Soviet computer scientists. Their trip to the United States is described in the News and Notices section of the Communications, Vol. 1, December 1958, pages 23-25. The invitation was addressed to the four Americans by Academicians S. A. Lebedev, Director of the Institute for Precise Mechanics and Computational Techniques, and A. A. Dorodnitsin, Director of the Moscow Computation Center of the Academy of Sciences. The four Americans had all met and served as hosts to the Soviet computer visitors at the Engineering Summer Conference on Digital Computers at the University of Michigan in Ann Arbor, and on a side trip to Urbana, Illinois, where the Soviet scientists visited the Computer Laboratory of the University of Illinois. Mrs. Jane Robertson, wife of Professor Robertson, accompanied the group to the Soviet Union. The itinerary of the visitors to the Soviet Union may be outlined as follows: August 27-September 4. Arrived Moscow. Visited the Institute of Precise Mechanics and Computing Techniques, the Computing Center of the Academy of Sciences of the U.S.S.R., and the Computing Center at the Moscow State University. September 4-6. Visited the Computing Center of the Ukrainian Academy of Sciences at Kiev. September 8-9. Visited the Leningrad branch of the V. A. Steklov Mathematical Institute of the U.S.S.R. Academy of Sciences. September 10. Departed Moscow. Unassigned time was for travel. At each computing installation, each of the four visitors gave one or more lectures in computer-oriented areas such as numerical analysis, circuitry and logical design, and automatic programming. They talked with Russian and Ukrainian computer mathematicians and engineers working on comparable problems and were given very complete guided tours of the following Soviet equipments: 1. Strela, Computation Center, Moscow 2. BESM I, Institute of Precise Mechanics and Computing Techniques, Moscow 3. BESM II (under construction), Computation Center, Academy of Sciences, Moscow

LC 2 : a language for conversational computing

The purpose of time-sharing is twofold: to increase the efficiency of the computer system and, while attaining this increase, to permit efficient communication between a programmer and his programs. This communication we may call conversation. Prevailing programming languages like FORTRAN, PL/1, ALGOL, COBOL, etc., are poorly designed for such interactive programming. However, languages like JOSS [1], APL [2], and the to be described LC2 are much more suited to this task.

ACM Publication Policies and Plans

was mainly due to his efforts that the ACM Journal was established as an important contribution to the growth of computer technology. It was under his direction that the Journal grew impressively in size and, more importantly, in the quality of the technical papers. With former ACM President, Dr. John W. C a r l and others, Dr. Alt gave impetus to the establishment of the Communications in recognition of the need to have timely information given to ACM members on important topics more qmckly than is possible in the Journal. Dr. Alt has made an important contribution to the computing field; ACM members and computing people everywhere owe him a real debt of gratitude. Many have asked about the objectives of the Journal and the Communzca-tions, and the similarities and differences of these two publications. The Journal will generally accept papers of a research nature, papers marking significant advances in computer technology and papers which are less temporal in nature and more generally applicable as reports of research and developments to the computing field. The Communications is designed to bnng to the attention of ACM members information in the computing field of a more immediate or current nature. More precisely, it will contain preliminary reports of research m progress, information on computer standards, news and notices, and articles on techniques and applications. The establishment of the Communwat,ons pays tribute to the fact that the computing field is young and growing rapidly and that much information of a current and immediate nature should be made available to ACM members on a monthly basis. Certainly there will be some overlap in the kinds of papers appearing in these two publications. Neither publication will purposely emphasize certain kinds of articles. For example, papers on programming research and development will appear in both the Journal and the Communicatwns depending on the particular orientation of the paper. In general, the publications will accept papers on computer methods, applications and design. Every effort will be made to continue and hasten the trend toward a better balance of papers. Probably because numerical analysis is so closely identified with the well disciplined field of mathematics, there has been a preponderance of papers in that field. Every effort will be made in the future to publish good papers on computer programming and business applications. In addition, of course, papers on computer logic, system design, and hardware will 121

Two Thousand Words and Two Thousand Ideas-The 650 at Carnegie

Prior to the first commercial drum computers (see Carr and Perlis 1954), only a few universities building computers under defense contracts had access to electronic digital computers. Other equally good commercial drum machines were available, but it was the IBM 650 that opened the tool of digital computation to American universities. A major contributing cause was the generous and farsighted grant of 60 percent rental provided by IBM. Along with the computer came opportunities and problems: How, where, and by whom would it be administered? What were the important problems the computer would help to solve? Where did it fit into the educational and research environment? How would it be paid for? Who would use it, and how would it be used? At Carnegie Tech (now CMU) the 650 arrived in

Compiling matrix operations

It is unfortunate that almost all of the presently used algebraic languages do not provide the capability of linear algebra. Operations such as the inner product of vectors, the product of two matrices, and the multiplication of a matrix by a scaler must inevitably be written out in detail in terms of the individual components. The reasons usually given for avoiding linear algebra in these languages are (1) the difficulties which would arise in scanning linear algebraic expressions, and (2) the uncertainty involved as to the amount of temporary storage needed during the evaluation of linear algebraic expressions when the program is executed. The purpose of this paper is to show how these two types of difficulties can be overcome. Although suggestions have been made for even further increasing the general capability of ALGOL such as including the ability to form a matrix from a collection of vectors, we shall be content here to consider the ordinary operations of linear algebra. Even if this much becomes available in algebraic languages, considerable progress will have been made. The following remarks constitute a suggestion for the addition to ALGOL of linear algebraic expressions.

A new policy for algorithms

Debate is under way concerning the Algorithms section of the Communications. Specifically, there is considerable sentiment in favor of also publishing algorithms in languages other than ALGOL.

A format language

One of the most primitive parts of a formula language is its specification of input-output actions within the framework of the language. While the specification is intrinsically more complex, say, than the evaluation of an arithmetic expression, most of the difficulties associated with input-output specification arise from the fact that the desired operations have not been properly defined using the framework of a programming language. Indeed, the complexity largely disappears when a programming language is constructed to specify input-output actions. The point to be made here is that the definition of an appropriate programming language makes more rational and simpler all three phases of the input-output programming cycle: (i) source program construction, (ii) object program construction, (iii) object program execution.

Identifying and developing curricula in software engineering

One basis for developing an education program is the recognition of a continuing need for a certain class of professionals in our society.