what is programming theory?

eh?

i don't want to offend anyone, because i don't know what everyone's programming education history is, but i can say about me.

i have been studying programming (theory) for 10 years now (and i am nowhere near the end), and i consider myself at least competent enough to discuss about it.

on the other hand, the main topic of this site (codewalkers) is php, and php is as opposite to programming theory as you can get. php is a language that was made for very practical solutions to some web problems. for quick hacks and fixes..

from many angles, it is a very "ugly" language, but it is still the best one for the web stuff.

that is why i was a bit surprised when matt created the "programming theory" forum on this site, because i wouldn't expect 1 in 100 threads to belong to this forum (or probably much less).


but don't trust me. look up the books like "introduction to algorithms" and "the art of computer programming", and check what kind of problems that books tackle. (just a hint, those books are reference books for programming courses at MIT, stanford and similar places).


and if you still wish to discuss this things with me, that go ahead, but don't just present your opinion, but try to back it up. or if not that, you should at least be competent enough to discuss this matters.

in other words, if you don't even know what O(n*log(n)) means, don't even bother..

oh, and i forgot to answer to something.

in my opinion, many other threads don't belong there, but i just didn't react before.. (until one day it just hit me that some of you don't understand what theory is).

and about your post at, http://codewalkers.com/forum/index.php?action=displaythread&forum=theory&id=22&realm=default well, your question is far from theory (how to decompile a java class), but that whole thread is maybe the best candidate for the theory forum.

on that page you are solving very unpractical problems, and learning quite a lot. i know it is not exactly what i would call "theory" but i suppose it is as close as you can get with this web stuff..

What topics would be appropriate for this forum?

Recursion vs. iterative algorihtms
Various sorting algorithms
What makes PHP an ugly language
Things PHP can learn from LISP
The smallest subset of PHP that is Turing-complete
Reasons you can't solve undecidable problems
Does P = NP?

Zombie, looking from the programming education history point of view, I'm not able to win you in any programming discussion. I started programming in the age of ten. I taught myself with the help of books from the library and bookstore, and I'm still trying to learn everything possible and impossible about programming using books and the Internet. I'm no way even near the professional level, but I'm not near the beginner level either, so I feel I'm fairly competent to talk about this.

I'd like to hear a simple definition for "Programming theory" from someone who is more advanced in programming than me. Anyone? Or is it even possible to define (simply)?

Now to the point: This argument started from the question "Do superusers know what theory means?" We need to find a solution to that problem so we don't need to continue this. Generally, everyone of us uses his/her own brain to decide where the thread belongs to. But opinions do vary, sometimes very much. According to the FAQ page there are 18 people who are allowed to move threads. That is much. We can't expect everyone to automatically agree with everyone. That's why I'm asking for a general definition for the term "Programming theory."

O(n*log(n)) == efficency of an algorithm

Mark

For me, programming theory doesn't have too much to do with what people generally regard as programming. Lots of people think of programming as when you write a program, and I supose that it is. But programming theory has not too much to do with specific implementations, or even specific languages. In fact, we can discuss programming theory whether we are programming in assembler, basic, or whatever.

Programming theory tackles that most important part of programming: deciding what you are going to do. No matter how smart you are, if you approach a problem the wrong way you will end up wasting time. My lecturers always said that before you write the first line of code you must have a picture of the how the entire program will be layed out, what screens, functionality, etc.

As much of an art as programming is, it always demands results. Programming is about making specific solutions to specific problems, although invariably almost all the time you will face the same mundane problems. In such an environment, programming theory doesn't really have much place, except as a guideline how to tackle certain problems efficiently.

I think one of the reasons that this thread does have many posts is that the PHP contests here are very 'algorithmic', and I think it attracts more people that might post in such a thread than there would be if no contests were held.

Unfortunately, discussing how to generate Fibonacci terms in O(logn) time isn't too applicable to our daily lives, so I don't expect such conversations to go on here. In light of this, I think we can afford to relax the constraints slightly, as to what posts are applicable to this thread. Unfortunately, some are just terrible, posts such as 'how would I program a chess game?'

I think having very high-level discussions will probably be beyond the scope of most participants.

tkarkkainen: i can't give you a formal definition in english. anyway, i am not that good at expressing myself about this in english, but i gave you the titles of two books that should give you a good grasp of programming theory. look them up in your local library..

btw, here is the "theory" department of MIT. maybe you find something interesting there http://theory.lcs.mit.edu/




well said. i almost forgot about the competition, since i didn't participate for more than a year (and i got here because of it ;)).

I am not a programmer, not by any means of the word. I am a network admin and web designer. However, this thread is extremelly useful to me and I'd imagine (based on the post) to many others.

I was in essence, forced to learn PHP and as noted,



and it is a very practical and beneficial solution.

So theory, be it programming or otherwise is a beneficial subject to review, especially when one is trying to grasp the philosophy of a subject i.e. PHP, and gain an understanding, no matter how basic, of its functionality.

So thank you Matt and all the other programmers who help us here and express your thoughts on Theory.

If you are looking for a book on computation theory, I highly suggest Michael Sipser's "Introduction to the Theory of Computation". You can find it at Amazon, B&N, Quantum Books, or probably other places as well - it's rather expensive though at $103. The book is used in MIT's 6.045 class in the link above. It's not really a book for a lay person - i.e., the more formal math background you have, the better. The book covers: Automata, Languages (regular and context-free), Computability Theory (decidability and reducability), and Complexity Theory.

Really, the only acceptable qualification for a meaningful discussion of programming theory is a frim understanding of Discrete Mathematics, Computer Architecture and Algorithms. Unfortunately the foremost is the only one that can (in my opinion) be grasped fully through independent study; I was fortunate enough to have formal training in the others and it took several years... In any case, I recommend Discrete Mathematics and its Applications for anyone who wants to learn about Computer Science, Programming, and the underlying pragmatism.

yes, a solid mathematics backgraound is a must.

anyway, here is a good (free, pdf) intro book about programming theory (didn't read it, but came with good recommendations) http://www.cis.upenn.edu/%7Ewilf/AlgComp2.html

- Matt Wade

I think above definition is what matters more than anything else.. :laugh:

Now, isn't this thread, which I moved to that forum, touch above definition???

nope.


btw honcho, are you by any chance at MIT? ;)

Not any more. I graduated a couple years ago with a B.S. in Math, a B.S. in Computer Science, and a Master's in Computer Science.