Upgrade to Pro — share decks privately, control downloads, hide ads and more …

All of this has happened before and it will hap...

Jade Allen
September 18, 2014

All of this has happened before and it will happen again

A brief discussion of the history and origins of 4 early programming languages. (N.B.: A lot of the papers in the bibliography can be found outside the ACM paywall with a search on your favorite web search engine.)

Jade Allen

September 18, 2014
Tweet

More Decks by Jade Allen

Other Decks in Technology

Transcript

  1. Agenda •  Background •  FORTRAN (1957) •  LISP (1959) • 

    ALGOL (1960) •  COBOL (1960) •  Conclusions •  Bibliography
  2. Jean Sammet, [2] “Virtually everyone agrees that today programming is

    an art, not a science. Many -- although somewhat fewer -- people would contend that programming can (and should) be made into a science.”
  3. State of the Art <1954 •  UNIVAC Short Code (William

    Schmitt) •  1st compiler "A-0" developed for UNIVAC in 1952 at Remington-Rand (Grace M. Hopper)
  4. Grace Murray Hopper [1] “The programmer has been supplied with

    a "code" into which he translates his instructions to the computer...until the novelty of inventing programs wears off and degenerates into the dull labor of writing and checking programs. This duty now looms as an imposition on the human brain.”
  5. The goal of early programming languages was to overcome the

    limitations of early computers and their (octal) op codes, such as: •  Lack of floating point computation •  Lack of sufficient (hardware) registers •  Primitive input/output operations •  Restrictive (hardware) instruction sets (Logical AND support but not OR) Backus, [4]
  6. What did "automatic programming" mean to contemporaries in 1954: • 

    Providing mnemonic operational codes (instead of octal) •  Symbolic addresses (instead of physical memory locations) •  Library subroutines •  Floating point operations Backus [4]
  7. 1954 •  Dwight Eisenhower is the President of the United

    States. •  First children given polio vaccine. •  USS Nautilus (1st atomic submarine) commissioned. Library of Congress
  8. John Backus •  Hated "programming" •  Invented "Speedcoding" •  Lead

    FORTRAN team 1954-57 •  Invented "Backus Normal Form" (BNF) to formally describe ALGOL grammar, 1959 •  National Medal of Science, 1975 •  ACM Turing Award, 1977
  9. John Backus, [4] “Early systems were slow and expensive. Through

    experience, programmers began to doubt writing efficient programs could be automated. Also some marketing departments claimed their systems could have human-level lexical understanding of programmer intentions.”
  10. John Backus [4] “One was accustomed to finding lots of

    peculiar but significant restrictions in a system when it finally arrived that had not been mentioned in its original description.”
  11. John Backus [4] “[I]t is difficult to convey … the

    strength of skepticism about 'automatic programming' in general and about its ability to produce efficient programs in particular as it existed in 1954.”
  12. Fortran Design Goals •  "Virtually eliminate coding and debugging..." • 

    Make programming "faster, cheaper [and] more reliable." •  Serve as an implementation for others to copy.
  13. John Backus [4] “It will suffice to say that [the

    compiler] produced code of such efficiency that its output would startle programmers who studied it… The degree of optimization performed… was not equaled again until optimizing compilers began to appear in the middle and late 1960s.”
  14. John Backus [4] “It was an exciting period; we were

    often astonished at the amazing transformations which made the program efficient but which we would not have thought to make as programmers ourselves.”
  15. • Weak debugging facilities • Poor design choices • Project timeline estimation problems

    • Other flaws included whitespace issues (ignoring blanks, even in the middle of identifiers), disallowing mixed integer and floating point expressions. • Cost savings and efficiency gains are eaten up by increases in program size and complexity.
  16. John McCarthy •  Invented LISP, 1959 •  Worked on ALGOL,

    1958-1963 (if/then/ else) •  Invented the term and technique of "garbage collection", 1959 •  Coined the term "artificial intelligence." •  Predicted "computing utilities" (aka cloud computing), 1961. •  ACM Turing Award, 1971 •  National Medal of Science, 1990
  17. John McCarthy [7] “LISP seems to be the second oldest

    surviving programming language after Fortran, so maybe we should plan on holding one of these newspaper interviews in which grandpa is asked to what he attributes having lived to 100.”
  18. LISP Design Goals •  Enable symbolic reasoning (instead of explicit

    numerical computation) •  Representation of data and expressions as lists in memory •  Functional composition and recursion as programming idioms •  Automatic memory management
  19. •  Used Church's λ notation as a means to give

    a function a name which could be passed as an argument to other functions. [6] •  Developed garbage collection to deal with the "erasure" problem. (This is the answer to the question of how to reclaim memory occupied by the remnants of a list after applying a car or cdr function.) •  Steve Russell hand coded the eval function (defined in [5]) in machine code for the IBM 704 which served as an interpreter for the language.
  20. John McCarthy [6] “Writing eval required inventing a notation representing

    LISP functions as LISP data... with no thought that it would be used to express LISP programs in practice.”
  21. •  Interpreted programs ran about 60 times slower than compiled

    programs. •  No lexical scope for variable bindings (that came in LISP 1.5) •  McCarthy planned that the awkwardness of the S- expression syntax would be solved by the use of "M-expressions" in an ALGOL like syntax (but that goal "receded into the indefinite future." [6])
  22. John McCarthy [6] “[W]e had the intention of producing a

    compiler. But the FORTRAN people were expressing their shock at having spent 30 man-years on [a compiler]... and it didn't seem like these graduate students would hold still long enough that we could get 30 man-years of work out of them.”
  23. E.W. Dijkstra [8] “LISP has jokingly been described as 'the

    most intelligent way to misuse a computer'. I think that description a great compliment because it transmits the full flavour of liberation: it has assisted a number of our most gifted fellow humans in thinking previously impossible thoughts.”
  24. ALGOL Design Goals •  Create a language "as close a

    possible to standard mathematical notation and be readable with little further explanation." [11] •  "Use [this language] for the description of computing processes in publications." [11] •  Advance a "universal" agreed-upon language without worrying about specific implementation details.
  25. Alan Perlis •  Original member of the U.S. ALGOL delegation.

    •  Considered to be a pioneer in promoting computer science as a discipline distinct from mathematics. •  President of ACM, 1962 •  ACM Turing Award, 1966
  26. Alan Perlis, [10] “There are two ways to write error-free

    programs; only the third one works.”
  27. Peter Naur •  Editor of ALGOL Bulletin •  Refined Backus

    Normal Form grammars (as part of the ALGOL 1960 report) •  One of the 1st ALGOL compiler implementations •  ACM Turing Winner, 2005
  28. Edsger W. Dijkstra •  Best known for his graph traversal

    algorithm •  Foundational paper in concurrent computing (invented the mutex) •  "GO TO considered harmful" •  1st implementation of ALGOL 1960 •  ACM Turing Winner, 1972
  29. ALGOL Firsts •  Recursion •  Code blocks with lexically scoped

    variables •  Call by value and call by name (an early form of "lazy" evaluation) •  An intrinsic boolean type •  Keywords "modern" readers find familiar •  Formally specified grammar Sammet, [12]
  30. Alan Perlis, [9] “Neither [Backus'] nor any other adequate formal

    semantic definition technique was available to the ALGOL 60 designers. We now know that semantics is considerably more difficult to treat than syntax.”
  31. •  ALGOL was never very popular in the United States.

    •  Failed to replace FORTRAN (as indeed, it is still around today!) •  Semantic meaning of the formal grammar was open to interpretation. •  Opened new avenues of inquiry in computer science. (Such as [13])
  32. COBOL Design Goals •  Desired a single programming language that

    worked across multiple manufacturers' computers. •  Maximum use of English (instead of mathematical notation.) •  "Easy to use" even if that meant the language would be less expressive. •  Broaden the base of who can state problems to computers.
  33. Grace Murray Hopper •  PhD, Mathematics, Yale, 1934. •  Retired

    US Navy as Rear Admiral, 1986. •  Famously visualized a nanosecond [14] •  Popularized the term "debugging" •  Lead "automatic programming" team at Remington-Rand •  "FLOWMATIC" was a major source for the design of COBOL [15]
  34. Jean E. Sammet •  M.A. Mathematics, Illinois, 1949. •  Worked

    at Sperry Rand 1955-58 on UNIVAC I •  Worked on COBOL for Sylvania 1958-61 •  Joined IBM 1961 and developed FORMAC
  35. Jean Sammet, [15] “It was quite clear that the disputes

    between people from the same organization were often as great (and sometimes greater) than between people from different organizations, and this applied to virtually all the organizations involved, whether manufacturers or users. ”
  36. Jean Sammet, [15] “[T]here was a strong anti-IBM bias in

    this committee from me and some (but certainly not all) of the others. Since I was not working for IBM at the time, I can freely (although not with pride) admit that in some cases suggestions or decisions were made on the basis of doing something differently than IBM did it.”
  37. “Experience with real users and real compilers is desperately needed

    before 'freezing' a set of language specifications. Had we known from the beginning that COBOL would have such a long life, I think many of us would have refused to operate under the time scale that was handed to us…” Jean Sammet, [15]
  38. •  Introduced the notion of a typed record or data

    definition distinct from executable code. •  Was at one point the single most widely used programming language. COBOL still in use at many companies today. •  Met its design goal of implementation independence (largely due to influence of US government contract requirements.)
  39. •  Many early practitioners expected problems related to programmer error

    and efficiency to be solved by high level computer languages. •  Even when it was an explicit goal to reduce what today we might term "cognitive load" in a particular programming language, it proved difficult to achieve satisfactory results in practice. •  I would strongly recommend browsing through the bibliography and reading one or two of these papers.
  40. Bibliography •  [1] Hopper, Grace Murray, The education of a

    computer, Proceedings of the ACM National Conference, 1952. http://dl.acm.org/citation.cfm?id=609818 •  [2] Sammet, Jean, Programming languages: history and future, Commications of the ACM, July, 1972. http://dl.acm.org/citation.cfm?id=361454.361485
  41. Bibliography •  [3] FORTRAN, The Pioneering Computer Language, http://www-03.ibm.com/ibm/history/ibm100/ us/en/icons/fortran/

    •  [4] Backus, John, The History of FORTRAN I, II, III, History of Programming Languages I, ACM. http://dl.acm.org/citation.cfm? id=800025.1198345
  42. Bibliography •  [5] McCarthy, John, Recursive functions of symbolic expressions

    and their computation by machine, Part I, Communications of the ACM, 1960. http://dl.acm.org/citation.cfm?id=367177.367199 •  [6] McCarthy, John, History of LISP, History of Programming Languages I, ACM, 1978. http://dl.acm.org/citation.cfm? id=800025.1198360
  43. Bibliography •  [7] McCarthy, John, LISP - notes on its

    past and future, Proceedings of the 1980 ACM conference on LISP and functional programming, ACM, 1980. http://dl.acm.org/citation.cfm?id=800087.802782 •  [8] Dijkstra, Edsger, W., The humble programmer, Communications of the ACM, 1972. http://www.cs.utexas.edu/users/EWD/ewd03xx/ EWD340.PDF
  44. Bibliography •  [9] Perlis, Alan, The American side of the

    development of ALGOL, History of Programming Language I, ACM, 1978. http://dl.acm.org/citation.cfm? id=800025.1198352 •  [10] Perlis, Alan, Epigrams on programming, ACM SIGPLAN Notices, September 1982, http://dl.acm.org/citation.cfm? id=947955.1083808
  45. Bibliography •  [11] Naur, Peter, The European side of the

    last phase of the development of ALGOL 60, History of Programming Languages I, ACM, 1978. http://dl.acm.org/citation.cfm? id=800025.1198353 •  [12] Sammet, Jean, Programming Languages: History and Fundamentals, Addison-Wesley, 1967.
  46. Bibliography •  [13] Irons, Edgar, A syntax directed compiler for

    ALGOL 60, Communications of the ACM, 1961. http://dl.acm.org/citation.cfm?id=366062.366083 •  [14] Hopper, Grace Murray, video clip explaining the length of a nanosecond and a microsecond. https://www.youtube.com/watch?v=JEpsKnWZrJ8
  47. Bibliography •  [15] Sammet, Jean, The early history of COBOL,

    History of Programming Languages I, ACM, 1978. http://dl.acm.org/citation.cfm?id=800025.1198367