This is only true if you don't count any functional or logical programming languages. As soon as you are using Haskell, all your knowledge about C or Java is just useless.Quote:
Dont learn "programming languages" per se. Learn programming. When you start study programming for real, you will soon realise all languages are 99% the same. All have if statements, loops, booleans etc.
First a programming language has multiple paradigms, not only one. Second you mixed up the terminology. Both, Java and C, are imperative, Java is Object oriented while C is procedural. And lastly, C is used for low level problems, because it is a compiled language which can be used to optimize very well, which is partly due to its paradigms (not so different from machine code) but mostly due to the fact that it is a compiled language. You can also use C++ to write an operating System, or Pascal, Ada, D, and so on, while in Java it's completely impossible.Quote:
What changes is the paradigm for example, java is more used for object oriented type of problems. While C is used for dealing with imperative programming, which means its easier to use it on lower level problems like operative systems.
There is no OOP in C. You can Simulate objects in C, but that due to the turing completeness of C. You can also simulate objects in SQL, because SQL is also turing complete, same for Brainfuck. But neither SQL nor Brainfuck nor C are OOP.Quote:
But that doesnt mean you cant use C to object oriented, it's just harder.
C isn't hard either. You can start with any language you like, in my opinion there is no hard or easy language, there are only different learning curves. If you are an impatient person, who want to have great results after a short period of time, you are right, Python might be better fitted than C, but if you are patient enough, you can also start of with C, it might take a little longer till you can write the great software you like (even so I don't think anyone who does not want to build hardware based software should start learning C, this language just sucks)Quote:
What you should do is first learn the concept of what programming is, then I recommend python to beginners, because it's such an easy language and yet so powerful.
What he probably wants to say is that OOP is not just possible in C in theory, but also practical enough to be used by huge real world applications (e.g. GNOME, GTK, Gimp).Quote:
There is no OOP in C. You can Simulate objects in C, but that due to the turing completeness of C. You can also simulate objects in SQL, because SQL is also turing complete, same for Brainfuck. But neither SQL nor Brainfuck nor C are OOP.
Well not really, the examples you gave are standard procedural programming like it exsisted since Algol and therefore is older than the oop idea. but the core essence of oop is the binding of functionality to the dataset, which is not in the means of C.Quote:
As long as you can't name real world applications that simulate OO in SQL or Brainfuck, I'd say that your comparison is unfair.
No, you seem to have a clear misunderstanding what OOP means. It's true, that a truing complete language can simulate any other Turing complete language, therefore you can simulate OOP in C. But the key word is to simulate. OOP is a language paradigm, which describes the composition of datatypes with functionality. A language either has this paradigm or not. Like the english language can express the same as the Japanese, this doesn't mean that the Japanese gramma can be used within an english sentence.Quote:
The alphabet of a turing machine is a single set I agree.Quote:
A turing machine doesn't support different Types, as the working alphabet is a single Set. You can simulate types, but you will never have a type system in a turing machine.
This is the difference between equivalence and equality. The means might be equivalent as they both are Turing complete, but not equal, as there is clearly no OOP in C.
Yet it is practical to write code in C that is object oriented. It is not practical to do the same in SQL or your other analogies.Quote:
I still don't know what you mean with object oriented code in C.Quote:
typedef struct { ... } Datastruct;
typedef Datastruct* PDatastruct;
int foo(const PDatastruct dataRef, otherparams) {
if (!dataRef) return -1; // invalid dataset
// Do something with dataRef as dataset
}
typedef struct { ... } ParentStruct;
typedef { ParentStruct prefix; ... } MyStruct;
typedef struct { ...; union { ... } childElements } MyStruct
Yeah but than you still have only one type, the type of the Set. The big point of types is, that many different types are just a substitution for different types. Take for example char and bytes. 'A' == 65. This is in a Turing machine not directly possible, as a Set can't hold 2 equal values (by definition). This can be simulated, using a substitution rule, but than it's still the equality <> equivalence problemQuote:
But in a set you can include other sets. {{..},{--}} See this is only one set, including two sets.
Of course you can simulate OOP (or any other paradigm) in any language. The point warfley tried to make is that some languages support OOP out of the box and have special syntax and what not for OOP (and those are the language you should call OOP languages). Other languages (like C) dont have stuff like inheritance or interfaces (but you can still simulate that). If we all would use your argument, then any language would fit all paradigms. Haskell would then be a OOP language, because Haskell eventually gets lowered to machine code and thats turing complete and thus haskell must be an OOP language (hint: haskell actually isn't a OOP language).Quote:
If I get a simple inheritance program, lets say class geometric figure includes class triangle and class circle (example) that is the OOP paradigm, and i can translate that into c and in turn into assembly, and into a turing machine. How can it not be able to do OOP paradigm in c?
A language paradigm has nothing to do with the result, it's about getting to the result. Only because you can either walk to the grocery store, or take the car, this doesn't mean walking equals driving.Quote:
If I get a simple inheritance program, lets say class geometric figure includes class triangle and class circle (example) that is the OOP paradigm, and i can translate that into c and in turn into assembly, and into a turing machine. How can it not be able to do OOP paradigm in c?
I'm writing from the following stance: OOP (or programming paradigms in general) isn't about syntax or language features, but how we organize and design code.Quote:
The typical procedural coding style since user defined types (so since Algol) is something like this:
And inheritance was realized using prefixes in structs:Code:typedef struct { ... } Datastruct; typedef Datastruct* PDatastruct; int foo(const PDatastruct dataRef, otherparams) { if (!dataRef) return -1; // invalid dataset // Do something with dataRef as dataset }
or using unions:Code:typedef struct { ... } ParentStruct; typedef { ParentStruct prefix; ... } MyStruct;
This was long before the idea of OOP even existed. Than some day, some developer thought: It's pretty stupid to always need to add an extra param for a data reference, and the need to check if it's nil. Also inheritance using prefixes or unions is pretty confusing, and increases the chances of errors. So why not letting the compiler do it for you.Code:typedef struct { ...; union { ... } childElements } MyStruct
And with this the idea of OOP was born, which got first really implemented in Smalltalk.
So calling the old procedural style OOP is like calling a caterpillar a butterfly, as the essential development, which is the least required minimum to be called that, is missing.
Otherwise, what is the point in discriminating between procedural and OOP style, as, if you can call the old style OOP, this would be the same.
[Only registered and activated users can see links. Click Here To Register...]Quote:
A programming paradigm is a style, or “way,” of programming.
Never use the phrase “programming language paradigm.”
A paradigm is a way of doing something (like programming), not a concrete thing (like a language). Now, it’s true that if a programming language L happens to make a particular programming paradigm P easy to express, then we often say “L is a P language” (e.g. “Haskell is a functional programming language”) but that does not mean there is any such thing as a “functional language paradigm”.
[Only registered and activated users can see links. Click Here To Register...]Quote:
Programming paradigm (in this course)
A pattern that serves as a school of thoughts for programming of computers
Das ist doch quasi die Definition von Syntax Zucker und hat nichts mit irgendwelchen Paradigmen zu tun. Zuerst kommt das Paradigma, dann die Sprache, mit der man das komfortabler anwenden kann.Quote:
This was long before the idea of OOP even existed. Than some day, some developer thought: It's pretty stupid to always need to add an extra param for a data reference, and the need to check if it's nil. Also inheritance using prefixes or unions is pretty confusing, and increases the chances of errors. So why not letting the compiler do it for you.