Skip to first unread message
Dilawar Singh
Jun 30, 2014, 11:43:32 AM6/30/14
to pn...@googlegroups.com
There are many ways one can think about programming languages. Its easy to get lost in details.
For a newcomer, it is useful to keep in mind that languages are designed in (more or less) for two purposes: first, to honor an mathematical abstraction; and second, to be efficient in certain situations. The first category have only few languages: Haskell, Lisp/Scheme and Prolog come to the mind. Rest of them can be divided further depending on where they shine most. It is not enough these days to ask which Programming language to use; one also have to say what one wants to do with it.
C remains close to hardware, it can control your computer to great extent. This is the fastest programming language (almost all the time). C/C++ rules on embedded systems, simulators and kernels of your operating system. If you want speed, you want C or C++. Cost: too much typing and taking care of low level details.
Python, Perl and Java are higher-level. A great many details which C/C++ programmers must know are taken care by these languages behind the scene. The cost is significant reduction in speed. Slower still are Matlab/Scilab and Octave.
The rule of thumb: if it is easier to program in a language, it is slower. Or in other words, if you don't sweat is out, your computer has to sweat it out for you. But who cares if computer sweat?
WHY PYTHON?
Python is approximately 10-50 times slower than C++ and somewhat faster than Matlab [table]. On recursive algorithms, it performs much better than Matlab.
A recent addition in Python is cython which turns Python code to equivalent C code. This often does magic to a slow programs. You get speed of C and convenience of Python. We will conduct a workshop on Cython for those who are dealing with programs which runs for days to give results.
The attraction of Python is not speed but the ease of programming. Its syntax is cleaner and takes less effort to master. It syntax is cleaner than its closest rival perl. One has to type less to do same amount of work. And what matter most, if a language got it, Python got it. There are millions of programs written in python waiting to be used. Python can be used for almost any purpose easily.
Python makes you get started quickly and there are so many people willing to help you. All it takes few hours of work to get started with basics of language. One can then start working at the problem at hand and help others getting started with the language.
--
For a newcomer, it is useful to keep in mind that languages are designed in (more or less) for two purposes: first, to honor an mathematical abstraction; and second, to be efficient in certain situations. The first category have only few languages: Haskell, Lisp/Scheme and Prolog come to the mind. Rest of them can be divided further depending on where they shine most. It is not enough these days to ask which Programming language to use; one also have to say what one wants to do with it.
C remains close to hardware, it can control your computer to great extent. This is the fastest programming language (almost all the time). C/C++ rules on embedded systems, simulators and kernels of your operating system. If you want speed, you want C or C++. Cost: too much typing and taking care of low level details.
Python, Perl and Java are higher-level. A great many details which C/C++ programmers must know are taken care by these languages behind the scene. The cost is significant reduction in speed. Slower still are Matlab/Scilab and Octave.
The rule of thumb: if it is easier to program in a language, it is slower. Or in other words, if you don't sweat is out, your computer has to sweat it out for you. But who cares if computer sweat?
WHY PYTHON?
Python is approximately 10-50 times slower than C++ and somewhat faster than Matlab [table]. On recursive algorithms, it performs much better than Matlab.
A recent addition in Python is cython which turns Python code to equivalent C code. This often does magic to a slow programs. You get speed of C and convenience of Python. We will conduct a workshop on Cython for those who are dealing with programs which runs for days to give results.
The attraction of Python is not speed but the ease of programming. Its syntax is cleaner and takes less effort to master. It syntax is cleaner than its closest rival perl. One has to type less to do same amount of work. And what matter most, if a language got it, Python got it. There are millions of programs written in python waiting to be used. Python can be used for almost any purpose easily.
Python makes you get started quickly and there are so many people willing to help you. All it takes few hours of work to get started with basics of language. One can then start working at the problem at hand and help others getting started with the language.
--
Dilawar
Reply all
Reply to author
Forward
0 new messages