编译器与解释器的区别，
- history :1954 ibm develops 704,software costs >hardware costs
- 这就使人思考如何更好地去写软件
- speedcoding:1953 解释器早期例子。比手写执行效率缓慢10-20倍。占用大的内存空间 300bytes=30%ofthe machine
- fortran I 1954-1957 :写下公式机器可以执行的公式。formula translated.
- the first compiler,huge impact on computer science(theory+practice)
- led to an enormous body of theoretical work
- modern compilers preserve the outline of FORTRAN I
- the outline
- Lexical Analysis |
- Parsing ---(Syntactic)
- Semantic Analysis(Types,scopr)
- Optimization
- Code Generation(Translation)
Strcuture of Compilers
Economy
Why are there so many programming languages?
Application domains have distinctive/conflicting needs. (it is hard to design one system for all)
-
scientific computing.(need) Fortran
- good fp
- good arrays
- parallelism
-
business applications SQL
- persistence 可靠不易丢失
- report generation
- data analysis
-
system programming C/C++
- control of resources
- real time constriants
- 时空资源
Why are there are new programming languages?
programmer training is the dominant cost for a programming language. new languages tend to look like old languages.(java vs c++)
- predictions
- widely used languages are slow to change
- easy to start a new language(生产力> traning cost)
- Languages adopted to fill a void.(时代技术的进步发展)
What is a good programming languages?
there is no universally accepted metric for language design.