whatiscs.ppt

1
whatiscs.ppt

• version 20101117

2
What Is Computer Science?

• William J. Rapaport
• Department of Computer Science Engineering,
• Department of Philosophy, Department of
  Linguistics,
• and Center for Cognitive Science
• rapaport_at_buffalo.edu
• http//www.cse.buffalo.edu/rapaport

3
Why Ask?

• Academic / political purposes
• Should UBs CSE be in CAS? SEAS? SOI? SCS?
• Pedagogy What should we teach in Intro CS?
• programming?
• theory?
• ?
• Do CSists merely program?
• Publicity for prospective students?

4
Why Ask?

• Intellectual / philosophical motivation
• What is CS really?
• Is it like something else?
• math?
• electrical engineering?
• Is it sui generis

5
Is CS sui generis?

• Computer science has such intimate relations
  with so many other subjects that it is hard to
  see it as a thing in itself.
• Marvin Minsky
• Computer science differs from the known sciences
  so deeply that it has to be viewed as a new
  species among the sciences.
• Juris Hartmanis

6
Fundamental Principle of What Is Questions

• There are no sharp boundaries in nature
• only continua spectra
• We carve nature at joints (Plato)of our own
  devising (Kant)
• ?There may be no good answer beyond
• CS is what CSists do!
• But What do they do?

7
Newell, Perlis, Simon 1967Newell Simon
1976Simon 1969/1996

• CS the science of computers
• not a natural science, but
• a science of the artificial
• computers includes
• hardware, algorithms, etc.
• So
• CS the artificial science engineering of
  computers surrounding phenomena

8
Knuth 1974

• CS the study of algorithms
• Algorithms what you can teach a computer
• Which functions (I-O) can be efficiently
  computed?
• Need a computer to find out!
• So
• CS study of algorithms
  surrounding phenomena

9
Hartmanis 1992

• CS study of information
• how to represent info
• how to process info
• the machines that do this

10
Shapiro 2001

• CS natural science of procedures
• including algorithms,
• recipes (specifications vague)
• non-halting ( interactive) procedures
• heuristics (incorrect O/P)

11
Brooks 1996

• CS ? science
• not concerned with discovery
• CS engineering
• concerned with making
• physical computers
• S/W systems

12
Denning 2010

• Computing is a 4th great domain of science
  alongside the physical, life, and social
  sciences.
• CS
• discovery (science)
• implementation (engineering)
• of information processes

13
What Is Computation?

• function
• set of I/O pairs (relation)
• same I/P ? same O/P
• function machine

14
What Is Computation? Functions

• But function machine ? function!
• functions dont do anything
• function machine needs gears
• i.e.) algorithm!
• but not all functions are algorithmic
  (computable)!
• function machine
• computer!

15
Computable Functions

• A function is computable
• there is an algorithm that computes it
• i.e.) that specifies how
• I/P can be transformed into O/P

16
Algorithm

• Algorithm for problem P
• finite procedure for solving P
• finite instructions
• completable in finite time (?)
• completable in finite steps (?)
• instructions unambiguous for executor
• know how to do each instruction
• know what to do next
• must halt (?)
• O/P must be correct (?)

17
4 Great Insights of CS

• Bacon/Leibniz/Boole/Turing/Shannon/Morse
• All information about any computable problem
  can be represented using only 2 nouns 0, 1

18
4 Great Insights of CS

• Turings Insight
• Every algorithm can be expressed in a language
  for a Turing machine
• arbitrarily long tape divided into squares
• read / write head
• only 2 verbs ( basic instructions)
• move(dir) (where dir L, R)
• print(sym) (where sym 0, 1, nil)

19
4 Great Insights of CS

• Boehm Jacopinis insight
• structured programming
• only need 3 grammar rules
• sequence (first do this then do that)
• selection (if P, then do this else do that)
• repetition (while P do this)
• recursively, this that can be
• any basic instruction
• any complex instruction created by grammar rules
• optional
• exit
• named procedures
• recursion

20
4 Great Insights of CS

• Church-Turing Thesis
• An algorithm isdef anything equivalent to a
  Turing-machine program
• TM Post production system lambda calculus
  Markov algorithm recursive functions
  register machines, etc.
• There are non-computable functions
• e.g.) Halting Problem

21
So, what is CS?

• CS tries to answer these questions
• What is computation?
• More generally, what is a procedure?
• Which functions are (or are not)
• computable?
• efficiently computable?
• theoretical CS
• How can they be computed?
• what specific algorithms do the job?
• what machines are needed?
• other branches of CS