Bjarne Stroustrup Quotations 1

1.
Please talk about C + + books.
No, there's no way a book is the best for everyone. But for the real programmers, if he likes to learn some new concepts and techniques from the "Classic style" book, I recommend my
C + + programming Language,
The third and special editions of the 1998 edition. The book is about pure and pure C + +, completely independent of the Platform and library (of course, the standard library). The book is intended for programmers with some experience to help them master C + +, but not for inexperienced beginners or for casual programmers to taste C + + fast food. So the focus of the book is on concepts and techniques, and it's a lot of work on completeness and accuracy. If you want to know why C + + will become today, my other book The
Design and Evolution of C + +
can give you a satisfactory answer. Understanding the principles and limitations of design can help you write better programs. Www.accu.com is one of the best book review sites, and a lot of experienced programmers are speak up here, so take a look.
2.
How long does it take to learn C + +?
It depends on what you mean by "learning". If you are a Pascal programmer, you should be able to quickly make your C + + level close to Pascal, and if you are a C programmer, you can learn to use C + + for better C-style programming in a day. On the other hand, if you want to fully master the main mechanisms of C + +, such as data abstraction, object-oriented programming, universal programming, object-oriented design, and so on, which have not been familiar to them before, it is not surprising to spend the last year or two. So is that the time it takes to learn C + +? Maybe double it again, I think it's going to be at least as long as it takes to be a better designer and programmer. If learning a new language does not make a profound change in the way we work and think, why bother? Learning a new, different language and programming style is simple compared to being a pianist or mastering a foreign language.
3.
Understand C is a prerequisite for learning C + +?
Whether! C + + in a subset of the same is actually more than the C language itself to learn, the type of errors will be less, and not like the C circle, there is a better support library.   So you should start learning C + + from this subset. 4.
If I want to be a real OO programmer, do I have to learn Smalltalk first?
Whether. If you want to learn Smalltalk, go ahead and learn. This language is interesting, and learning new things is always a good idea. But Smalltalk is not C + +, and using Smalltalk's programming style in C + + does not have any good results. If you want to be a good C + + programmer and don't have a few months of boredom, concentrate on learning C + + and the ideas behind it.
5.
How do I start learning C + +?
It depends on your foundation and your motivation to learn. If you are a beginner, I think you'd better find an experienced programmer to help you, otherwise the mistakes you inevitably make in your study and practice will greatly discourage your motivation.   In addition, even if your compiler is equipped with sufficient documentation, a C + + book is always essential, after all, the documentation is not a good textbook for learning programming ideas. When choosing a book, it is important to note that the book teaches standard C + + from the start and uses the standard library mechanism with constancy. For example, reading a string from the input should look like this: string
S   Standard C + + style cin >> s; Instead of this: char S[max]; /* Standard
C style
*/scanf ("%s", &s); Take a look at the books recommended by the solid C + + programmers.   Remember, no book is the best for everyone.   In addition, to write a typical C + + program, and avoid using C + + syntax to write the traditional style of the program, the new bottle of old wine does not make much sense. (Unfortunately, the current Chinese C + + textbook in the market is in line with B.)
S of this standard can say no, we have to find some English information on the Internet to learn. --translator)
6.
How to improve my C + + program?
It's hard to say. It depends on how you use the language. Most people underestimate the value of abstract classes and templates and, in turn, use modelling mechanisms (cast) and macros with impunity. In this respect you can read my articles and books. The role of abstract classes and templates, of course, is to provide a convenient means to construct a single class hierarchy or reuse functions, but more importantly, they provide a concise, logical service representation mechanism as an interface.
7.
is the choice of language important?
Yes, but don't expect miracles either. Many people seem to believe that a language can solve almost all of the problems they encounter in system development, and they constantly look for the perfect programming language, and then fail again and again and again in frustration. Others have relegated programming languages to unimportant details, putting a lot of money into the development process and design methods, and they are always using COBOL,
C and some proprietary languages. A good language, such as C + +, can help designers and programmers do a lot of things, and their abilities and flaws can be clearly understood and treated.
8.
Did the Ansi/iso Standard Committee spoil C + +?
Of course not! They (our) work very well. You can find some sophistry in some details, but I am personally pleasantly the language and the new standard library. Iso
C + + is better and more organized than previous versions of C + +. Compared to the beginning of the standardization process, you can write more elegant and maintainable C + + programs today. The new standard library is also a true gift. Since the standard library provides strings,
Lists, vectors, maps, and the basic algorithms that work on them, have changed dramatically in the way that C + + is used.
9.
Do you want to remove some C + + features now?
No, really. People who ask these questions probably want me to answer one of the following features: Multiple inheritance, exceptions, templates, and Rtti. But without them, C + + is incomplete.   Over the past n years, I have considered their designs over and over, and have improved their details with the standards committee, but none of them can be removed without causing a major earthquake. From the point of view of language design, the part that I dislike most is the subset that is compatible with C, but it cannot be removed, because that hurts too much for the handlers who work in the real world. C + + is compatible with C, a key design decision that is definitely not a hawking gimmick. The implementation and maintenance of compatibility is very difficult, but it has really benefited programmers so far. But now that C + + has new features, programmers can get out of the hassle-free C-style. For example, using a container class in a standard library, like a vector,
Lists, maps, strings, and so on, can be avoided with the underlying pointer manipulation techniques. 10. If you do not have to be compatible with C, will you create a language that is Java?
No, it's far. If people have to compare C + + and Java, I suggest they read the Design and Evolution of
C + +, to see why C + + is today, test both languages with the principles I followed when designing C + +. These principles are clearly different from those held by Sun's Java development team. In addition to the similarity of surface syntax, C + + and Java are very different languages. In many ways, Java is more like Smalltalk (translator: I use Sun's training materials When I study Java, which clearly reads: Java is designed with a similar syntax to C + + and similar semantics to Smalltalk. So it can be said that Java and C + + is dubious, and Smalltalk is the heart of the same. The Java language is relatively simple, part of the illusion, partly because the language is incomplete. Over time, Java grows significantly in both volume and complexity. It will grow by two to three times times in volume, and there will be some implementations of extensions or libraries. This is a path of development that every successful business language must go through.   By analyzing a language that you think has been successful in a wide range, I know there must be no exceptions, and in fact it makes a lot of sense. The above paragraph is in the Java
1.1 was written before the launch. I'm sure Java needs a template-like mechanism, and it needs to enhance support for intrinsic types. Simply put, it is for the sake of basic integrity that these tasks should be done. There are also a lot of small changes that need to be made, mostly extensions. In the fall of 1998, I was from James
Gosling (the founder of the Java language) got a proposal to add intrinsic type, operator overloading, and mathematical computation support in Java. There is also a paper, a world-class master in the field of mathematical analysis, the University of Berkeley's W.
How Java's floating-point hurts Everyone written by Professor Kahan
Everywhere ("and see how the floating-point operation of Java harms the world's mortal beings"--the translator), exposing some of Java's secrets. I find that the advocacy of Java in television and in publications is inaccurate and offensive and annoying. It is an insult to the programmer that all non-Java code is rubbish, and it is not practical and irresponsible to suggest that all the reserved code be rewritten in Java. Sun and his followers seem to think that in order to deal with Microsoft's evil "Age of Empires", they must boast so much.   But insults and scams will only push programmers who prefer to use different programming languages to the Microsoft camp. Java is not platform-independent, it is itself a platform. Like Windows, it is also a proprietary business platform. That is, you can write code for Windows/intel, or you can write code for JAVA/JVM, and in either case, you are writing code for a platform that belongs to a company that is linked to the business interests of the company. Of course you can use any language that combines the operating system's mechanisms to write programs that are available to the JVM, but things like the JVM are strongly biased to the Java language.   It does not seem to be a universal, fair, language-neutral Vm/os.   Privately, I will insist on using portable C + + for most of the work, and doing the rest in different languages. ("Java
Is isn't platform-independent, it is the platform ", B.
S of this comment for C + + users have a great impact, the translator in several foreign news groups see, some C + + master even this sentence as their own signature file, to show the attitude towards Java and to defend the determination of C + +. There are actually a lot of programmers who don't just think of their favorite language as a tool, but a belief. --translator)
11.
What do you think of the C # language?
I have nothing to say about the C # language itself. It is not easy for me to believe that the world needs another proprietary language, and this language is specific to a proprietary operating system, which is even more difficult for me to accept.   To put it bluntly, I am not a dedicated language addict, but a proponent of an open formal standard. 12.
Do you really recommend Ada, rather than C + +, when you're doing a bigger project? Of course not. I don't know who this rumor is, it must be an ADA, either overly fanatical or hostile.
13.
Would you like to compare C + + with other languages?
Sorry, I don't want to. You can Evolution in the Design and
The reason is found in the introductory text of C + +. Many reviewers have invited me to compare C + + with other languages, and I have decided not to do such things. Here I would like to reiterate a point I have been stressing for a long time: comparisons between languages are meaningless and unfair. It takes a lot of effort to make a reasonable comparison between mainstream languages, and most people don't want to pay that much. It is also necessary to have sufficient experience in a wide range of applications, to maintain an impartial, objective and independent position, with an impartial belief.   I have no time, and as the creator of C + +, I will never gain complete trust in the sense of impartiality. People try to compare the different languages, some of which I have noticed over and over again, and frankly I am worried. The authors try to behave impartially, but ultimately are hopelessly biased towards a particular application, a particular programming style, or a particular programmer culture. To make things worse, when one language is clearly more famous than another, some imperceptible rescue begin: The flaws in the more famous languages are deliberately watered down and disguised in a roundabout way, while the same flaws are described as fatal in less-well-known languages. Similarly, the technical information about the more well-known languages is often updated, and the technical information of the less famous languages is often a few years ago, so what is the fairness and significance of this comparison?   So my comments on languages other than C + + are strictly confined to a particular category of generality.   In other words, I think C + + is the best choice for most people when developing most applications. 14.
Does it make you feel uncomfortable that others often compare their language with C + +?
When these comparisons are incomplete or for commercial purposes, I do feel uncomfortable. The most widely disseminated comparative commentary is mostly published by a language, say, a supporter of the Z language, to prove that Z is better than other languages. Because C + + is widely used, C + + is usually the first name on the blacklist. In general, such articles are sandwiched between the products provided by suppliers in the Z language and become a means of competition in their markets. The shocking thing is that quite a few of these comments cite articles from employees working in companies that develop the z language, and these can't be tested to prove that Z is the best. In particular, there are some fragmented facts in these comparisons (so more deceptive-translators), after all, no language is the best in any case.   C + + is certainly not perfect, but please note that the fact that it is deliberately chosen may seem correct, but sometimes it is completely misleading. In the future, when you see the article on language comparison, please pay attention to who wrote it, his statement is not based on facts, with justice as the yardstick, especially the criteria for judging is not for each language quoted is fair and reasonable. It's not easy to do.
15.
is C better than C + + in small projects?
I do not think so.   In addition to the problems caused by the lack of a good C + + compiler, I have never seen a project that uses C more appropriately than the one that was used in + +. (However, the problem with the C + + compiler is still not negligible, and when you see a C + + program that has the same functionality, the executable code is one times larger than C and has a much slower speed, it will feel that way.) Translator) The following content is from the visual
C++developer ' s journal editor Elden Nelson March 2000 to B. S's interview 16.
If you now have the opportunity to design the C + + language from scratch, what are the different things you will do?
Of course, you can never redesign a language, it doesn't make sense, and any language is the product of its time. If I were to design another language today, I would still consider the beauty of logic, efficiency, versatility, the complexity of implementation and people's preferences.   Be aware that people's habits have a huge impact on their preferences. Now, I'm going to look for a much simpler syntax, and I'll limit the conflict of type systems to a few situations, and you can easily find these problems.   This makes it easy to prohibit unsafe operations. B.
The principle of S is: For bad code, even if it is not completely forbidden, at least let it be in the world, rather than hide in the dark corner of the stabbing. C + + has actually provided such a mechanism, for example, if you use an expression like reinterpret_cast<int> (pointer) that is obviously very bad, it's easy for others to find out where the problem is. Just C + + still allows you to use the traditional, C-style styling mechanism, and many people have been using this old style, so it is a lot of trouble. B.
s means how good it is to ban old-fashioned styles now! As a language designer, I'm afraid I don't have the chance, but as a language user we still have a lot of hope to improve our code. It should be time for us to ponder. --translator) I will also make the core language of the volume as small as possible, including the class and template of the key abstract features, and put a lot of other language features in the library to solve. Of course I will also make sure that the core language is strong enough that the libraries themselves are enough to be produced in this core language. I don't want the creation of a standard library to use a mystical mechanism that does not belong to the language itself. In addition, I will make the definition of this core language more precise. (There are a lot of new languages in the library that use "mysterious mechanisms that do not belong to the language itself", such as VB and Java.) Theoretically, this is almost a rogue act, so b.
S disagree. But it doesn't hurt to start with the utility. (translator) Most importantly, I will maintain a long gestation period before the language is widely used so that I can improve on the basis of feedback from others.   This may be the most difficult, because once something is clearly outstanding and promising, everyone will be flocking to use it, and making any incompatible corrections thereafter will be very difficult. I believe that these ideas are very similar to what I originally designed for C + +, and that these ideas guide the evolution of C + + over the past ten or twenty years.   Of course, I don't think there's anything that can make me feel like a perfect language. 17.
What enhancements do you expect C + + to remove something?
Unfortunately, there are some things that should be thrown away, but I'm afraid it's hard to really erase anything. The first thing that should be discarded is the C-style styling mechanism and the type truncation transformation. Even if not banned, compiler authors should at least give a strong warning about this behavior. I want to be able to completely replace the array with something like a vector, but it's obviously not going to work. But if programmers can actively use vectors instead of arrays, they will benefit immediately.   The point is that you don't have to use the most complex and tricky tricks in C + +, and now there's a much better alternative. As for the main features, I didn't want to get rid of anything. In particular, the main features that distinguish C + + from C are not likely to be thrown away in a calm way. The people who usually ask these questions want me to pick out mechanisms such as multiple inheritance, anomalies, and templates to accept criticism. So here I want to make it loud and clear, I think the multi-inheritance mechanism is necessary for the inheritance of statically typed languages, and the anomaly mechanism is the correct way to deal with errors in large systems, and the template mechanism is a panacea for type-safe, sophisticated and efficient program design.   We can pick on these mechanisms in small details, but on the big side, these basic concepts must be adhered to. We are still learning standard C + + and are developing newer and more interesting programming techniques based on the features provided by the standard.   In particular, people have just begun to use STL and anomaly mechanisms, there are a lot of efficient and powerful technology is little known, so big can not rush to add new mechanisms. I think the current focus is on providing many new, more refined, more useful libraries than ever before, and this is a huge potential. For example, if there is a widely used, more sophisticated support for concurrent programming of the library, it would be a great Boon--c style line libraries (such as pthread--translator) is not good enough. We can also better fit in with a variety of other systems, such as SQL and different component models. People in the field of numerical computing seem to have gone ahead in this respect, and the development of efficient and sophisticated libraries like blitz++, Pooma and MTL has achieved remarkable results. (The translator visited the homepage of blitz++ and Pooma on the Internet, the former is a high-performance math library, allegedly its performance and Fortran
77, while supporting a large number of C + + features. I think anyone who knows something about the field of numerical computing knows how great it is. Pooma is a project that specializes in C + + parallel math algorithms, and its prospects are even more limitless. The translator is very agreeable to B.
S of this concept. When we have enough experience, we will be better able to decide what adjustments we should make to the standard.
18.
How should C + + and its standard libraries be tuned?
Clearly, the world has changed over the years and is moving toward a web-centric, distributed computing era. So do you think C + + can still maintain its position? It is impossible for programmers to combine several specialized languages (such as Perl, Javascript) to completely replace a common language?   (c + + is such a common language-translator) in order to match the new computing model, C + + and its standard library should be how to adjust? No language has ever been suitable for all the work, and I'm afraid I won't have it later. The actual system is often constructed in multiple languages and tools. C + + just wants to be one of several languages and tools that can complement C + + when certain specialized languages are particularly prominent in their domain. In other words, I think that if most of today's specialized languages can work together with a particular domain of C + + libraries, they will perform better.   Scripting languages often lead to hard-to-maintain code, and there is no room for optimization of the program's structure, extensibility, and maintainability. I'm not sure if the future code will really be web-centric.   Even a system that directly handles the web is primarily composed of program modules that handle local resources, such as IP connectivity. Geographic distribution and the high reliance of server Software on concurrency mechanisms are a challenge for the builders of the system. Some libraries for the above problems have emerged, and perhaps we will see that they are finally standardized.   Of course, some of the original operations and assurance rules should be added to the core language to provide better support for these libraries. In general, we need a real system/network-level security model for the web and the network. Expecting a scripting language like JavaScript to implement this model is a daydream. Note that I'm not saying that C + + provides a solution to this problem. C + + 's focus is on efficient access to system resources, not anti-fraud.
19.
What do you think of the future trend of C + +?
Will it decline in the next 10 years? Or is it basically a form of keeping it?   Or are there different forms of development and change? C + + has the most beautiful future. With it you can write great code. In addition to intentional malicious fraud, C + + will still be the best language for developing high-performance, high-complexity systems.   As far as I know, no language can be equated with C + + in terms of versatility, efficiency and refinement. I don't see a sign of fading in C + +. It will continue to grow in the future that I foresee. Of course, we'll see some changes in the next 10 years, but not as dramatically as you think. As with every language, C + + is also evolving.   "Language experts" are deafening for the noise of improvement, but the basic request of system developers is to remain stable. C + + will improve, but these improvements will be primarily to reflect lessons learned from practice, not to follow the fashion bandwagon. In order to use some new programming techniques more efficiently, such as general-purpose programming, some small features may be added. There will be a lot of libraries coming up, and I expect a new, better library support mechanism.   I hope that the new extensions will focus on general features that support abstractions rather than specific mechanisms to support certain special tasks. For example, the concept of "attribute" is useful, but I do not think there is a common programming language in it. It is easy to support this concept with a set of classes in standard C + +. If we feel that the race's support for the concept of "attribute" is not as good as it is, it will not immediately go into adding attribute mechanisms to the language, but rather consider how to improve classes and templates to help library designers approach the concept of "attributes" as closely as possible.   Perhaps a satisfactory response to the problem can be achieved by improving the mechanism of the function object. To keep C + + flexible for the next more than 10 years, it is essential to not let standard C + + drive the academic or commercial buzz. A large part of the features that people demand to add can be achieved by using existing standard C + + to develop new libraries. In fact, many of the features that people crave are already included in standard C + + and are supported by the latest compilers. For many programmers, the best way to improve code quality is not to pursue any language extensions, but to savor the latest C + + technical books with a good, slow taste (unfortunately we have not even had the opportunity to do so so far – the translator).
20.
What do you think of the growth of scripting languages?
Python, in particular, seems to provide an easier way to learn oo techniques. Some languages are good. Like Python, I like it. But I think the oo techniques you learn from different languages are not exactly the same.   Of course, every professional programmer needs to be proficient in several languages and understand the different programming and design techniques of various languages. In my opinion, a system built with scripting language is very different from a system built in a common language like C + +. The techniques learned from the two languages are clearly different.   There is no common part in OO technology that is critical to the efficient construction of various systems. 21st.
Are there plans to add new features to standard C + + to support distributed computing?
No, I don't think it's necessary. Using a better library is almost enough to solve the problem. At most, in order to support this type of library, we may add some low-level primitive operations and rule 22.
Is it possible for C + + to set a portable binary interface in the future?
If you say "portable" refers to the portability of cross-hardware and block operating systems, I guess the answer is no. We can certainly design an interpreter or a virtual machine (as in the Java approach), but in this way, the ability of C + + will be weakened by the inability to access system resources optimally. What I really hope to see in the near future is platform.
ABIs. For example, someone is trying to define C + + for Intel's new IA64 system
ABI, I think these efforts will be greatly supported by the users. Being able to compile code that is generated by different compilers can be a very meaningful thing.
Other important issues
23.
In many popular areas, C + + is losing light because it requires people to spend a lot of time dealing with basic tasks such as managing memory (because there is no garbage collection mechanism), managing dependencies between modules (because there is no package mechanism), and managing the version of the component. C + + lacks some modern languages that have been considered standard features. The coolest Java language in the rumors, for example, pays particular attention to these issues. So will solving these problems lead to the development of C + + to deviate from its fundamental purpose? How should C + + develop to ensure that our investment in this language has a reasonable return, rather than being forced to re-use another language?
I haven't noticed that C + + is less used than it used to be. On the contrary, the indicators I see indicate that the use of c+ is steadily growing. But the steady growth of such a large base and the increasing standards, portability and libraries have not created any deceptive news effects. I think what you mean by "Losing your Light" is just a marketing and journalism phenomenon.     If you need a garbage collection mechanism, you can insert a garbage collector in your C + + application. There are a lot of free and commercial garbage collectors that have proven to be excellent in important practice.     It doesn't matter if you don't want to use the garbage collection mechanism. You can use standard container classes, which greatly reduce the need to explicitly allocate and reclaim memory. This way, with modern libraries and modern programming styles, you can avoid most of the memory management problems.     the same technology can also be used to avoid general resource management issues. It is not only memory that leaks, thread handles, files, mutexes, network Connections, and so on are important resources that must be properly managed in order to establish a reliable system. If you think you can solve all the resource management problems with a garbage collection mechanism, you'd better wake up from your dream.    c++ provides a number of mechanisms to manage general resources. The key means-"Get resources is initialize"-can use function objects to manage lifetime issues. This technique is supported by the local construction and anomaly mechanism of the object in the language.     avid supporters of some languages always describe C + + in caricature, whereas C + + is actually much better. In particular, I think a lot of other features are flooding in, and in C + +, these features can be easily simulated. Instead, the new language is constantly adding new features to the process, which is why it is often two or three times times larger to be used from a language's birth to a wider range of uses.     at present, the most personal and organizational, the best investment for C + + is to better understand the standard C + + and modern C + + design programming techniques. Most people use C + + in a way that actually stays at the mid-80 level, even older than that.     As for module dependencies, my point is that there should be a clear line between the work of the programming language and the work of the system, and the dependencies should be separated from the programming language as much as possible and supported by the system.     I do not think that the issue of build version should be solved by programming language, which is a problem in the system category, which should be solved by providing the corresponding library in the language. C + + has such a mechanism. Solve such aThe problem does not deflect C + + from orbit. However, adding a lot of special features to C + + will leave C + + off track and will be a step backwards in maintaining portability and platform independence.    24. 
Standard C + + has been introduced for some time, Java also strides forward and made significant progress, how do you compare Java and C + + now? What do you think Java needs to do to become a "good" language like C + +? What do you think C + + has learned from Java? Are there any Java features you think can be absorbed by C + +?
I don't compare words.   It is very difficult to do the job well, and it is seldom professional. I think that the progress of C + + will be based primarily on the problems that its users encounter in their use and their own logic. Of course, some ideas in other languages will also be considered, but they cannot be simply counter switch over.   You have to look at the technical and ideological background of those mechanisms and find the best solution to support these technologies in C + +. Sometimes the best option is to use a combination of several languages. After all, no language is a universal and excellent. C + + is now, and will continue to be, one of the best languages in a wide range of applications. However, we cannot be dragged into the crowd, and we cannot add all possible features to C + + to flatter the public.   I think Java and C + + will be very different languages now and in the future, with similar syntax, but the object model behind it is significantly different.   One important difference for me is that C + + has an ISO standard, while Java is a proprietary language. 25.
In the last few years of Java, there was a lot of deceptive rhetoric saying it would be the ultimate language and would replace C + +. How do you think Java has impacted C + + followers in the last two or three years?
And now the word about Java is still ubiquitous. Aside from the record-breaking developments in Java over the past 5 years, the fanatical public seems to think that Java will eventually replace not just C + +, but all other programming languages. But on the other hand, the use of C + + continues to grow. I don't think Java's impact on C + + has led people to turn to the resources that were intended to create a better library of C + + tools to develop the Java tool Library. Java does not have a lot of new things for people who learn to program, so there is no effect on the definition of C + +. In that field, Java has to catch up.   For example, I think it would be a mechanism for sun to add similar templates to Java sooner or later. People should realize how different the goals of C + + and Java are.
Using C + + as a design concept to measure Java, or to measure C + + with Java's design philosophy, is not going to be very good. At the end of the interview, perhaps I should say again: C + + is still my favorite language, and when you write code you will find that no language can be as efficient and sophisticated as it is in such a wide range of applications and platforms.

Http://www.cnblogs.com/lfsblack/p/3274678.html

Bjarne Stroustrup Quotations 1