Interview Update With Bjarne Stroustrup On C++0x

An anonymous reader writes "DevX interviewed Bjarne Stroustrup about C++0x, the new C++ standard that is due in 2009. Bjarne Stroustrup has classified the new features into three categories: Concurrency, Libraries and Language. The changes introduced in Concurrency makes C++ more standardized and easy to use on multi-core processors. It is good to see that some of the commonly used libraries are becoming standard (eg: unordered_maps and regex)."

Comments

Re:It hurts you to learn C++ is still being used.

[Free the Cowards]

Been there, done that.

Most of the time, the potentially reduced running time of the C++ implementation never comes close to the months saved in development.

And when it does, it's trivial to go in and write the speed-sensitive portions of the program in a faster language.

Re:And Then COBOL 2009

[Tenrosei]

hours finding memory leaks? How bad are you at debugging? Plus there are tons of tools to assist with memory leak detection. However, protection is no substitute for abstinence. Learn to write code better. BTW go do some embedded software with C# or Java.

Re:C#++?

[Noodles]

Because performance is important to some people.

I just don't get it....

[AmazingRuss]

...what do people find so difficult about C++? Use the standard libraries, exception handling, and make sure your news all have deletes, and it's no more difficult than any scripting language. I actually prefer it over scripting languages, which have their place, but feel all sloppy and unspecific. It's like the difference between building a house out of 2x4s and building one out of sticks you found laying on the ground.

Re:I just don't get it....

[SupplyMission]

Flexibility. In C++ it is essentially impossible to make, say, a dictionary where each key can refer to an object of a completely different type. This is what you refer to as "sloppy", but I actually find this flexibility to be essential in designing good software. The fact that C++ does not allow it forces me to either twist my program's design in unnatural ways to fit the language, or do a lot of extra work to twist C++ to fit my program's design.

If you're not offended by the idea of using the Boost libraries, the boost::any class will let you be sloppy like that.

Manual memory management. In any complex program, balancing your news with deletes is not as simple as you make it out to be. Object ownership is a tough problem. Lots of C++ code solves this problem by making a lot of defensive copies, which in turn hurts performance greatly.

The boost::shared_ptr class has changed the way I write C++ code. It's a header-only class, so it's possible to only include it, and nothing else from Boost, in your project.

Readability and writability. With all the type information being declared all over the place, big template declarations, and the like, I find that C++ takes considerably more effort to both read and write.

I have found that wise use of typedefs can hugely improve readability and writability. On the other side of the coin, some people go overboard with typedefs, essentially making worse the problem they originally intended to solve. Like code formatting or any other convention, it has to be used where it makes the most sense. (Avoid developers with hammers who think everything is a nail, or needs to be turned into a nail, so they can hit it.)

For example, I find this to be quite readable and writable:

typedef std::vector<SomeHairyClass> SomeHairyList;
typedef std::map<CuddlyKeyClass, SomeHairyList> HairyListMap;
HairyListMap foo;

Saying all that without some judicious use of typedefs would of course be nearly unreadable, especially if later you have to make use of iterators and nested types in the vector or map.

The really big issues for me are the flexibility and the lack of libraries. The rest is less important. But with C++ it's like building a house out of 2x4s that you're not allowed to cut to length, whereas with moer modern languages it's more like building a house out of prefabricated rooms, with a ready supply of 2x4s and tools to shape them as you need if the prefabbed rooms don't fit your needs.

On the contrary, I find that C++ gives you all the 2x4's, wood panelling, nails, screws, and all the tools you need. You just have to become a half-decent carpenter before you can start building your house. It's just a little bit harder to quickly make working prototypes (build a home from prefab parts), but as you indirectly pointed out, there are plenty of other languages that are better suited to that building style. The right tools for the right job.

Please note that this is just my opinion, and you asked for it. Feel free to disagree, but please don't flame.

Duly noted, just adding mine to the discussion. :)

Re:And Then COBOL 2009

[ardor]

I'll consider Java and C# as C++ replacements once they get:

1. REAL templates, not this generics joke
2. Proper RAII, which many programmers mistakenly believe to be useful for memory management ONLY (scoped locks come to mind) (Java/C# finalizers are no replacement)
3. Java: operator overloading

These points are serious, especially the first, without real templates, generic programming/metaprogramming at compile-time is not possible. These two are one of C++'s biggest strenghts, though.

To be fair, C# 3.0 is somewhat nice, especially its functional core. Java is a totally uninteresting language with very small expressiveness. Of course, if the job requires it, there is no discussion, but in my spare time, I prefer C++.

Re:Why not just call it C++#?

[Ed Avis]

Trust your uncle Bjarne. If you don't use it, you don't pay for it. You need not worry that the language is turning into C# or Python. It's still just as efficient for bare-metal programming as C ever was (and more so in some cases, with template specialization at compile time).

As for 'automatic memory management', that was one of C's big features. Remember the 'auto' keyword?

Re:It hurts you to learn C++ is still being used.

[iplayfast]

I've found that the biggest advantage for C++ is the portability. I have written an application backend for PC's (back in the days of DOS) and since then ported it through various versions of windows, Linux (for web use), Palms, and Pocket PC's.

Using C++ allowed me to very easily make the different processor needs, compatible, by writing little compatibility layers, which would swap bigend values, unpack data structures from disk into memory (so is on even boundary). and so on.

Yes the fast speed was why I originally went with the C/C++ route, but the big benefit has been the portability.

Re:It hurts you to learn C++ is still being used.

[johannesg]

...and roll on the C++-hatred! Second C++ article in a short time, and again lots of venom and anger. "Months saved in development"? Really? What are you doing, implementing your own OS before you start application development? Here's a newsflash: C++ also has support libraries, just like Java, Perl, Python and Ruby. They may not be part of the language specification (and I still think that's a weird idea to begin with, but I'm old-fashioned that way), but that doesn't mean they don't exist.

Anything you could want for in a modern language is there. And nobody is holding a gun to your head and making you write those scary templates if you don't want to.

I'm just positively amazed that Slashdot, in theory home of programmer geeks anywhere, should have such a violent dislike of C++. Not that there is nothing to criticize about it, but it is still an amazingly powerful, versatile tool that programmers anywhere would do well to learn.

Re:It hurts you to learn C++ is still being used.

[ardor]

And I can say with a straight face that you are wrong.

If you base your experiences on pre-2000s C++, you know very little of modern C++. I have been developing in it for more than 10 years, and a few years ago I would have agreed with you, but things have changed. Really.

Time for the C++ haters to post...

[serviscope_minor]

We will see the usual litany of C++ hating here in this thread. The hating will be generally based around misconceptions or problems that are 5 years old.

So to get them out of the way:

If you're leaking memory or spending time managing memory in C++, then you're using C++ wrong. Get a book written in the last 5 years.

If you're worried about compiler compatibility (with the exception of export which isn't much use anyway), get a compiler written in the last 5 years.

If you think that C does some subset of your task better, then write it in the common subset of C and C++ and quit whining. Or, write it in C and link it against your C++ code and quit whining.

If you think that templates simply provide code bloat, then get a compiler newer than 5 years old.

If you think C++ is slower than C, then get a good optimizing compiler (you know one written in the last 5 years) and do a benchmark. You will generally find that templates make C++ faster.

If you think "modern" languages are more expressive, then give "modern" C++ a try (insert comment about recent compilers here).

Sure there are valid complaints about C++, but the majority of them I hear on slashdot are complete bull. The majority of the remaining complaints will be fixed by C++0x.

One remaining problem is the lack of a vast array of standard, business oriented libraries. I don't write business oriented code, and I find the C++ STL one of the best libraries out there since it provides really good support for writing efficient algorithms.

Another problem is the difficulty in parsing C++. Sadly that's never going away.

But if you're going to complain about C++ compared to recent languages here, make sure that you're talking about recent C++ too, and try to make sure the complaints are accurate.

Re:It hurts you to learn C++ is still being used.

[FishWithAHammer]

You do know that you don't have to screw around with any of that in a managed language, right? "Very easily make the different processor needs compatible" my ass--Java/C# do it on their own.

C++ has one major problem

[Barnett]

C++ is an extremely powerful programming language and that is why I use it every day. But it has one major problem: It is too complicated. As long as you do programming full time you are OK but if too much of your time is spent on the application side of things you quickly get in trouble. This is what people like BS don't seem to get - not everyone can spend 100% of their time studying the language.

Re:C#++?

[neokushan]

If it was as good as it stands, then newer languages such as C# wouldn't take off.
Don't get me wrong, I love C++ and it's my primary programming language, but to say it's perfect as it is, is just silly.

Re:It hurts you to learn C++ is still being used.

[Free the Cowards]

No, the "premature optimization" thing applies to all areas. Especially areas where it's never fast enough.

Why? It's simple: resource management.

You have X amount of resources to put into your product. X is always finite. It's kind of tough to measure X, but you can think of it as lines of code, man-years, or even just dollars. The amount of resources you have varies a lot depending on your budget, how much time you have, and the quality of the programmers you have. But the important thing is that X is always limited.

Now you have two approaches:

1. Spend X on making the code fast from the start, and keep spending X until you run out.
2. Spend X on making the code functional and with good design. When appropriate (i.e. when the design is good and the code works), start spending X on making the program go faster. Keep spending X on speed until you run out.

Paradoxically, I hold that #2 will produce a faster program. This is because the X you spend on making the program faster in #2 will be more effective, because you've already laid the groundwork for it. It's always difficult and time consuming to optimize code that doesn't even run yet. It's much more efficient to optimize code that already works. So the result, even though you spend less X on speed, is a faster program.

Think of it as transporting a lot of material into the wilderness somewhere. If you first spend some of your resources on building a road, you'll get the job done for less time and money than if you just start hauling stuff into the woods immediately.

Some counterpoints.

[warrax_666]

1. Boost.
2. Nonsense. Boost has facilities for this ("any", iirc) and also for something called "sum" types which can achieve what you want in a better way ("variant", iirc).
3. shared_ptr, weak_ptr.
4. Yup. Going to be fixed by C++0x.
5. C++ can be written to be a lot more portable than your Ruby or Python.
6. A matter of taste.

Re:Some counterpoints.

[Kevin Stevens]

Boost has in many eyes really transcended from "just an external library" to an integral part of the C++ platform. It compiles on every major platform, and it is open source.

Boost is moving C++ forward at a rate 10x that of the standards committee. I am not sure why you felt integrating it with your project was difficult- it is header only for the most part and does not require you to use any specific pieces. Shared_ptr's, which are the most useful library of all, do tend to be viral in the sense that you have to use them everywhere, but this is a GOOD thing.

If you are doing C++ without Boost these days, you are really missing the boat.

Re:Some counterpoints.

[immcintosh]

First off, if you have a problem with using external libraries, then you just won't get anywhere with a C/C++. They are VERY general purpose, and intentionally so, and the whole idea is that implementation specific things are supposed to be provided in libraries, rather than the core language. That said:

1. Boost is actually very easy to integrate for most of its features. A few (small handful) of its components require compilation, but the vast majority of them are template-based and header only. Meaning just include a header file and there you go, you're using boost. No extra compilation/installation required.

2. This kind of thing is GREAT for doing small scripts, but HORRIBLE for doing large complex applications where type safety can be VERY important for avoiding bugs. If all you ever do are small, quick, limited scripts, then you're absolutely right that you should avoid C++, that's certainly not what it was meant for, and not so much the domain of a strongly typed language. For things like the software that runs large financial institutions and whatnot, there's a reason code like that should be avoided at all costs.

3. I have trouble imagining a situation where a real garbage collector would ever be superior to an RAII model with shared smart pointers for stuff allocated on the heap, outside of plugging up a leaking legacy app. Maybe for very simple programs, but once you get non-trivial destructors (for example, with objects that lock system resources), then you start having to do manual memory management in your GC environment anyway, and end up with a horribly ugly conglomeration of "mixed metaphors" as it were. Smart pointers really give you the best of both worlds: deterministic destruction, without having to worry about manually releasing anything. It's just a matter of getting used to declaring a smart pointer wherever you would have a "type *name" instead. So yes, I'd argue they ARE a substitute for garbage collection in almost any situation.

4. Sorry to be blunt, but you should probably RTFA on this one. The problem is solved through "concepts," which is the part of the new specification which deals with this specifically. It's essentially a C++ implementation of the "design by contract" metaphor.

5. In this case "can be" equates to "do whatever you like and it'll be portable on all major general purpose computers, and who uses Ruby or Python on embedded platforms anyway?" If you are able to compile with GCC, which is the case for pretty much every computer/OS combination in existence, then you can count on it being pretty damn portable. If you are programming for something like an ATM or a set-top box, then you probably aren't going to be using a high level scripting language anyway.

6. Yeah, it can get ugly. Thankfully this will be largely fixed with the "auto" keyword in C++0x.

Re:Some counterpoints.

[Anonymous Coward]

It is the discussion. Judging c++ with boost excluded is like judging perl with CPAN excluded. Who cares whether it's "part of the language" or not? Everyone who uses the language seriously uses them, and they're critical to understanding how the language is used effectively.

Re:Some counterpoints.

[Free the Cowards]

Again, I don't think I'm being unfair, I'm just saying why I use what I use. Python provides more libraries for the stuff I use than C++ does. "Batteries included" makes my life easier. Maybe this isn't fair to C++. So what if it's not? Should I make my life more difficult by using C++ out of a sense of fairness?

As for GC goes, no, it's not "just another memory management aid". Non-GC versus GC is the difference between having to think about memory management and not having to. Automatic refcounting still forces you to manually find and break reference cycles, and garbage collection does not.

Re:It hurts you to learn C++ is still being used.

[TomorrowPlusX]

There are vanishingly few programmer geeks left on slashdot. Most of the "programmers" here, these days, are folks who've written a few scripts or set up a movable type install.

There are a few real programmers left here, but they're lost in the noise. You know, the roaring noise made by the python and ruby folks.

This post brought to you by a C++ programmer who happens to love Python and Ruby ( and javascript! it's an amazing language ), but uses the different languages where appropriate.

Re:Objective C and C++

[cduffy]

Generally speaking, I agree with your post. However...

Technically there isn't any, c is a subset of c++ so anything you can do in c you can do in c++.

That's a really, really awful way to think about C.

C has a completely different set of best practices and design principals; anyone who puts "C/C++" on a resume I'm reviewing loses points as opposed to listing them separately.

Re:C#++?

[Duhavid]

"MBA's will not need programmers anymore, so we'll be able to code OSS full time!"

You realize that is what they said when they introduced COBOL, right?

Re:Why not just call it C++#?

[Eli Gottlieb]

Oh please. Pascal does everything C++ can now.

What tool is better than C++?

[pinkfloydhomer]

I also have an extensive experience with C++, and I tend to agree with a lot of the criticism that it gets.

But the problem is that no alternative exists for the type of problems where C++ is used extensively. I guess the most important area is games.

The world really NEEDS a language (the last low-level language) with the low-level performance of C++/C and with a full, modern library, and modern language features (threading, modern module system (not based on #includes and a crude preprocessor...), optional strong typing system a la Ada with optional runtime-checking etc etc etc.

Basically, a really nice, compiled, well-performing, modern low-level language could easily exist. But it doesn't. So we'll have to settle for C++ until someone makes something better.

Re:C#++?

[Yetihehe]

If you make language even an idiot can use, idiots will be using it. Like with VB.

Re:Objective C and C++

[immcintosh]

No, not really.

Nothing else in your post either supports, or even directly addresses this assertion.

It's not that C++ isn't good, its just that its harder to do things in it then it is to do those same things in either C# or Java. Harder to do means more expensive, and businesses all over are having to tighten their purse strings.

This is the same bullshit that's trotted out there every time this topic comes up, and it's no more true now than it ever was, which is not at all. If you do a little looking around, you'll find very elegant libraries that support every single feature you'll see in ANY other imperative language, and MANY declarative language features to boot. If you're against code reuse and third party libraries on some sort of general principle, then you're kinda missing the whole point of C++.

I'll be blunt, maybe an ass, maybe a troll, but having used all three of those languages extensively, I can say with almost absolute certainty that the only reason you should be having so much more trouble doing things in C++ (ESPECIALLY as compared to those two languages) is that you're either a very poor C++ programmer or have a pathological aversion to third party libraries.

In fact C++ is barely managing to hold its own any more against C# and Java.

Another thing that's been very much in the vogue to say lately, but I just haven't seen any meaningful evidence for. I think Bjarne covered this topic fairly even-handedly in TFA, and if he's to be believed then C++ usage is not suffering like popular belief seems to indicate. The crux of it being that web scripting was never a strong domain of C++ in the first place, and in actual applications programming C++ is still the leader of the pack.

Re:Truer words have never been spoken.

[scotch]

True as in lung cancer affects and destorys lungs, c++ affects and destroys c?

True today as it was then; i.e. not true at all.

Re:C#++?

[syousef]

If you make language even an idiot can use, idiots will be using it. Like with VB.

So lets make the language as difficult as possible. That way only good programmers will even be able to write code in it. Never mind the fact that they'll have to spend all their mental effort getting the code to work instead of focusing on the problem they're trying to solve.

The fact that an easy versatile language makes it easy for idiots to program in it is no reason to artificially make a language overly complex. That's insane. It's like making a hammer that requires a PhD to use just to prevent bad handymen from doing handywork.

In other words plenty of good code was written in VB by non-idiots who didn't want to focus on the language but had a practical problem to solve. You can leave the morons to survival of the fittest.