But, I believe that in 10 years C and C++ will not be languages what will be an asset in the toolbox of developers….why>? Because I believe that these languages MAY be viewed in the future as having too much flexibility that thwarts simplicity in design.

The languages I believe will be prevalent for developing with in the future are C#, F#, Matlab and VB.Net……and also oddly enough, VBA…..which I believe will be the new Cobol in 25 years……

I develop small-scale applications using VB.Net and VBA and the last thing I want to have to deal with is deeply profound memory issues……in my mind, I am glad to have an automated memory process for working with objects and variables…..which is the Garbage Collector……I don’t want to have to deal with these issues……when the PARALLEL methods come out in the next version of VB/C# multithreading will be much easier to deal with as well……

If the CLR can be constructed to assist my programming efforts in these areas, why on earth should I be bothered with learning languages where I have to add another level of abstraction to the programming process by understanding memory constructs and multi-threading?

I think it comes with the evolution of a language…..as programming languages evolve, so too will the need to be less involved with the intimate processes that occur “in the background” of computer processing which are handled in an automated way…or at least in a way that requires MUCH LESS intervention by the developer.

Mostly I was defending the assertion that nobody really understands C++ — which I think is arguably true of the whole language, for most practical versions of “nobody”. (I think there are a good number more people who can understand any given part of it with a minute or two of memory-refresher reading, though.) The idea of what constitutes “basic competency” in a language with that many dark corners is a quite fuzzy one, and we could have an entertaining and entirely meaningless argument over a beer about it!

(Meanwhile, it occurs to me that FizzBuzz would be a really good practice problem for template metaprogramming. Thank you for the implicit suggestion there!)

Of course you’d need to be better versed in the more advanced features of C++ beyond OOP if you are going to work in it on a regular basis. My message here is essentially “Even if you plan to be a VB, C#, Java Programmer it will help you to be a better programmer if you aren’t completely C++ illiterate.”

That said, your points are very well taken, thanks for your response.

I would think that a “competency in C++” (as opposed to a competency in a subset of C++) would require at least having a reasonable understanding of the concepts behind such meta-programming, such as how one constructs “empty” types and then uses them to pass information for making (arbitrarily complicated) compile-time decisions. But this is not even mentioned in Stroustrop’s big book on C++; one needs to get to specialized books on C++ template programming to get any explanation of it.

In any event, I think that’s a moot point. I am advocating only a basic competency in the language, not a memorization of the entire STL. I am just raising the bar a little by specifying C++ because I think that understanding OOP syntax is a critical skill for today’s programmers.

PS: I appreciate your input, even if we might disagree. A key reasons I started this blog was to create some meaningful discussion on some core aspects of programming.

C is a small, low-level language. There is a certain elegance to its simplicity.

C++ is a monstrous language that no one really understands.

It’s important to know the difference, especially if you’re advocating learning one of them.

Learn C before graduating.
Part two: C. Notice I didn’t say C++.