I'm not completely disagreeing with you, but I think it would take a major change of focus for Microsoft to fork an open-source JVM implementation and do with it as they like. Microsoft seems to be very committed to making.NET as the development platform for Windows. Encouraging Java development in any way would seem to hinder that goal.
Why not let Sun provide an official seal of approval?
I implement a jvm (which I can do right now, since the specs are available - well, theoretically, it's a bit of an undertaking). But you don't want to use just some random jvm. So I then submit my jvm to Sun so they can run it through some strenous testing of its execution and source code (assuming it's open source), and eventually I get a 'yay' or 'nay.'
Of course, this would not be a trivial service, and would probably have to cost a considerable hunk of money. But could some sort of certification scheme alleviate the forking fears? For example, let's say someone produces a jvm optimized for some particular model of execution. Sun could eventually okay it, but make sure its known that by using this jvm, there will be tradeoffs between this thing and the other.
I think you're still missing the point. My parents don't want to write shell scripts. They don't want to monitor system usage. They don't even care enough to reorganize the icons on their desktop. They want stuff to Just Plain Work. They don't care how, they don't care why; it doesn't matter. When my mother wants to do online banking, she is not conerned at all with the computer itself. Nor should she have to. Our job is to make sure that she doesn't have to. Users like my parents view the computer as an appliance like a tv, a dishwasher, an oven or a fridge. I don't care in the least how my fridge works, and I don't "respect the greatness" of it. I put food in, the food stays relatively fresh and cold, and I stay happy. That's how my parents want to be able to view their computer.
I think it's important that MAC OS X users learn to use the UNIX command line. Even if they don't like it, they need to respect it. If it wasn't for OS X being powered by UNIX, it would not be anywhere near as stable as it is right now. I'm not "dissing" OS X, because I use it and love it, but any user shouldn't be without the knowledge required to run the UNIX command line.
Why? Why would my parents, who only do application level stuff (web browsing, word processing, email, games), need to learn the "power of Unix"? They're non-technical end-users. They aren't concerned with harnessing the power of their machine, and nor should they have to be.
Someone else on slashdot, during the last hitchiker thread, addressed the issue of changes quite well: the series has appeared in a variety of mediums, and each time it was different than it was in the other mediums. It's going to be no different with the movies. So try to just enjoy the ride instead of saying, prematurely, "This person isn't right for this party for reasons x, y and z."
Isn't it interesting that after a few days of access to the source code, exploits are appearing for obvious bugs; yet MS have had the source code available to themselves for years but still managed to neither find nor fix these same obvious problems.
The leaked source code is old. We have no way of knowing if problems in the code have already been taken care of or not - we can't see the successors.
First, I thought Buffy ended because Joss Whedon wanted it to end, not because the networks cancelled it. Second, those 42-year-old soccer moms have a hell of a lot more money than their teeny-bopper kids, and are much better advertising targets, so that doesn't sound like the reason.
While providing freedom of choice might be reason enough to justify a project, practical programmers could be asking: What's the point?
Linux, maybe? Mac OS X? Free BSD? I see getting C# programs to run on other platforms as a practical purpose. Later on in the article he acknowledges that you'd be able to run these programs on Linux, but that's more like a throw away concession he makes. He plays dumb in the beginning, and makes himself look silly.
How is making C# a standard on Windows and Linux going to hurt Microsoft?
I think that the people behind the project have better goals than that - namely, getting a particular tool to work on Linux. People use Linux for a variety of things. It would be nice if C# - just another tool - worked under it. What's the big deal?
There is an obvious practical purpose to getting C# programs to run on Linux. The real question, however, is will the.NET framework on Linux be good enough so that people will actually want to use it? That's a very real question, but it's not the main one he's asking.
That's what popped into my head when I was reading the list - well, skimmed near the end, it started to feel like preaching and I lost my patience.
Surprise, surprise, that paper is what they reference for "the internet is stupid."
But I think that TCP/IP, the protocol the internet is built on, is a great counter example to end-to-end arguments. TCP/IP provides the abstraction of a virtual stream of bytes. It does this by ensuring packet integrity not at the application level (the "ends"), but along the way, inside the ends. This is in contrast to the FTP example used in the paper, which says that you'll need to do data integrity checking at the application level, so doing it along the way is redundant.
I'm not sure you get what I mean. When I said "What would that imply?" I meant it as "What does that do to our understanding of how the universe works, i.e., science?" not "Does that mean there's a god?"
The universe is wonderful enough without asking whether it means anything
Right, and that's not what I'm doing. But if the universe does in fact have a beginning, as the big bang theory suggests, then that certainly has implications.
and my atheism is of the sort that actively disapproves of theism, so any theory that removes one more of the theists planks is attractive to me.
Which I find, as I stated before, a dangerous mindset. Science is not about disproving religion, it's about observing how the universe works and trying to explain it. If the universe has a beginning, well that's really cool and it makes me wonder about how it all works. If it doesn't, and is some sort of steady state phenomenon, well, that's cool too and it makes me wonder about how it all works. Make sense?
I'm curious, what do you mean by "prefer"? Holding onto a theory because it happens to gel with what you want the universe to be like is dangerous.
Note that I am also an atheist, and I actually welcome the "Well, what would a beginning mean?" questions - mainly because I don't know, and that inspires wonder. It's what the universe is like that fascinates me, not what I want the universe to be like.
Re:Templates are strong typed in C++
on
How C# Was Made
·
· Score: 1
Knowing that you have to implement certain member (or global) functions in order for a templated class to compile is hardly arcane knowledge.
Re:Constraints on type parameters
on
How C# Was Made
·
· Score: 1
You're right, "member function" was the incorrect phrase. Sometimes they will be member functions, sometimes they will not. But, really, that wouldn't change what a constraint would look like in the language.
Re:Constraints on type parameters
on
How C# Was Made
·
· Score: 1
Requiring every class that supports == to derive from EqualityComparable is just a cumbersome extra for no benefit in a generic, C++ based world, where there are more ways to define and implement interfaces than just inheritance.
For the C++ world, I don't like that way either. If constraints were in C++, I think they should be done on a member function basis. That is, you explicity state which member functions must be implemented.
Templates are strong typed in C++
on
How C# Was Made
·
· Score: 5, Informative
C++ is the opposite. In C++, you can do anything you damn well please on a variable of a type parameter type. But then once you instantiate it, it may not work, and you'll get some cryptic error messages. For example, if you have a type parameter T, and variables x and y of type T, and you say x + y, well you had better have an operator+ defined for + of two Ts, or you'll get some cryptic error message. So in a sense, C++ templates are actually untyped, or loosely typed. Whereas C# generics are strongly typed.
I disagree with that assessment. Both C# and C++ generics/templates are strongly typed. It's just that the enforcement happens in different places.
In C++, if you try to stick a class into a templated class when that class doesn't have a particular member function defined, the compiler will yell at you, just like Hejlsberg said. But for some reason, this doesn't count as type checking? Yes, template error messages can be strange (and very long) if you're not familiar with them. But that's just a lesson in "know your tools."
To me, "strongly typed" is strict type enforcement at compile time. C++ templates certainly do this.
Constraints, however, are something that I think are a generally good idea. Stroustrup's reasoning for not including them in C++ was that "Requiring the user to provide such information decreases the flexibility of the parameterization facility without easing the implementation or increasing the safety of the facility." (The Design and Evolution of C++, Stroustrup, 343).
He does, however, show an interesting way to get around this using inheritance:
template <class T> class Comparable {
T& operator=(const T&);
int operator==(const T&, const T&);
int operator<=(const T&, constT&);
int operator<(const T&, const T&); };
template <class T : Comparable> class vector {// . .. }; (The D&E of C++, Stroustrup, 344)
This technique is similar to how C# does constraints, class List where T: IComparable. One is supported and enforced by the language, the other is a natural consequence of the languages facilities. In general, I think that constraints are probably a good thing. Having an error message like "Can not instantiate class Y<T> because T does not implement z()" is probably best, and when looking at a class' declaration, it would be nice to see up front what assumptions the templated class makes.
Okay, I'll bite. Read how the aglorithms in the STL are implemented. Stroustrup's book has excellent examples that are more digestable than cracking open the STL source if you don't feel like approaching that.
No, not a troll. When someone says something like "my opinion of operator overloading is that it is absolutely bad," I hope they're trolling, because it's very ignorant to say.
As an undergrad, I went to Virginia Tech. Now I'm a grad student (and a TA) at William & Mary. All of my courses at Tech were taught by full instructors or professors. I had two math classes taught by advanced grad students (they were Ph.D. candidates at that point). Here at W&M, the only course I know of that has been taught by a TA is the intro to CS for non-majors course (again, Ph.D. candidate).
Tech is a big university; W&M is a small one. I would have been shocked to walk into a junior level CS course at Tech and see a grad student teaching it. At that point, in fact, instructors didn't teach the courses anymore, only professors. Here at W&M, it's almost entirely professors from beginning to end.
The problem is that computer scientists don't make good programmers and vice-versa.
Bogus. CS, like other fields, has a diverse makeup. Some CS people do almost exclusively theory. Others specialize in software engineering. Some do systems, some networks, some programming languages. I can think of some CS professors that aren't good programmers, and I can think of some that are phenomenal.
PHd professor types that staff CompSci departments I've been in tend to have stopped learning about computers as soon as they finished their doctorate and instead concentrate on internecine politics, incomprehensible papers, and teaching the occaisional class (leaving most of that to T.A.'s who actually teach the class and understand how to compile programs).
Ditto. You just characterized the entirety of a field from your limited experience. Would you characterize all physicists from talking to a few theoretical physicists? There are experimental physicists out there too. How about soemone who specialized in quantum mechanics? There's E&M people too. Your claims are just plain silly.
Oh, and I am a TA. I have never heard of a school where TAs teach actual lecture. I teach lab. That is, I lecture for about 20 minutes, then let the undergrads do their programming lab and answer questions along the way. This is a far cry from actually teaching the course.
Not only does operator overloading have functional value, it allows functional programming. [rimshot]
That is, in C++, overloading operator () allows you to define function objects. Function objects allow you to simulate functions as a first-class entity, allowing functional style programming in C++. Check out the Functional C++ Library, FC++, it's cool stuff.
Slashdot Mirror
Obviously an open source implementation will grow and respond to demand rapidly and outpace something proprietary ...
Obviously? That's a big word to throw around that casually. I don't think it's obvious, or necessarily true, at all.
I'm not completely disagreeing with you, but I think it would take a major change of focus for Microsoft to fork an open-source JVM implementation and do with it as they like. Microsoft seems to be very committed to making .NET as the development platform for Windows. Encouraging Java development in any way would seem to hinder that goal.
Why not let Sun provide an official seal of approval?
I implement a jvm (which I can do right now, since the specs are available - well, theoretically, it's a bit of an undertaking). But you don't want to use just some random jvm. So I then submit my jvm to Sun so they can run it through some strenous testing of its execution and source code (assuming it's open source), and eventually I get a 'yay' or 'nay.'
Of course, this would not be a trivial service, and would probably have to cost a considerable hunk of money. But could some sort of certification scheme alleviate the forking fears? For example, let's say someone produces a jvm optimized for some particular model of execution. Sun could eventually okay it, but make sure its known that by using this jvm, there will be tradeoffs between this thing and the other.
I think you're still missing the point. My parents don't want to write shell scripts. They don't want to monitor system usage. They don't even care enough to reorganize the icons on their desktop. They want stuff to Just Plain Work. They don't care how, they don't care why; it doesn't matter. When my mother wants to do online banking, she is not conerned at all with the computer itself. Nor should she have to. Our job is to make sure that she doesn't have to. Users like my parents view the computer as an appliance like a tv, a dishwasher, an oven or a fridge. I don't care in the least how my fridge works, and I don't "respect the greatness" of it. I put food in, the food stays relatively fresh and cold, and I stay happy. That's how my parents want to be able to view their computer.
I think it's important that MAC OS X users learn to use the UNIX command line. Even if they don't like it, they need to respect it. If it wasn't for OS X being powered by UNIX, it would not be anywhere near as stable as it is right now. I'm not "dissing" OS X, because I use it and love it, but any user shouldn't be without the knowledge required to run the UNIX command line.
Why? Why would my parents, who only do application level stuff (web browsing, word processing, email, games), need to learn the "power of Unix"? They're non-technical end-users. They aren't concerned with harnessing the power of their machine, and nor should they have to be.
Someone else on slashdot, during the last hitchiker thread, addressed the issue of changes quite well: the series has appeared in a variety of mediums, and each time it was different than it was in the other mediums. It's going to be no different with the movies. So try to just enjoy the ride instead of saying, prematurely, "This person isn't right for this party for reasons x, y and z."
Isn't it interesting that after a few days of access to the source code, exploits are appearing for obvious bugs; yet MS have had the source code available to themselves for years but still managed to neither find nor fix these same obvious problems.
The leaked source code is old. We have no way of knowing if problems in the code have already been taken care of or not - we can't see the successors.
This, ladies and gentleman, is how to do satire. (And if it's not satire, I'll be very dissapointed, not to mention embarassed.)
First, I thought Buffy ended because Joss Whedon wanted it to end, not because the networks cancelled it. Second, those 42-year-old soccer moms have a hell of a lot more money than their teeny-bopper kids, and are much better advertising targets, so that doesn't sound like the reason.
While providing freedom of choice might be reason enough to justify a project, practical programmers could be asking: What's the point?
.NET framework on Linux be good enough so that people will actually want to use it? That's a very real question, but it's not the main one he's asking.
Linux, maybe? Mac OS X? Free BSD? I see getting C# programs to run on other platforms as a practical purpose. Later on in the article he acknowledges that you'd be able to run these programs on Linux, but that's more like a throw away concession he makes. He plays dumb in the beginning, and makes himself look silly.
How is making C# a standard on Windows and Linux going to hurt Microsoft?
I think that the people behind the project have better goals than that - namely, getting a particular tool to work on Linux. People use Linux for a variety of things. It would be nice if C# - just another tool - worked under it. What's the big deal?
There is an obvious practical purpose to getting C# programs to run on Linux. The real question, however, is will the
That's what popped into my head when I was reading the list - well, skimmed near the end, it started to feel like preaching and I lost my patience.
Surprise, surprise, that paper is what they reference for "the internet is stupid."
But I think that TCP/IP, the protocol the internet is built on, is a great counter example to end-to-end arguments. TCP/IP provides the abstraction of a virtual stream of bytes. It does this by ensuring packet integrity not at the application level (the "ends"), but along the way, inside the ends. This is in contrast to the FTP example used in the paper, which says that you'll need to do data integrity checking at the application level, so doing it along the way is redundant.
I'm not sure you get what I mean. When I said "What would that imply?" I meant it as "What does that do to our understanding of how the universe works, i.e., science?" not "Does that mean there's a god?"
The universe is wonderful enough without asking whether it means anything
Right, and that's not what I'm doing. But if the universe does in fact have a beginning, as the big bang theory suggests, then that certainly has implications.
and my atheism is of the sort that actively disapproves of theism, so any theory that removes one more of the theists planks is attractive to me.
Which I find, as I stated before, a dangerous mindset. Science is not about disproving religion, it's about observing how the universe works and trying to explain it. If the universe has a beginning, well that's really cool and it makes me wonder about how it all works. If it doesn't, and is some sort of steady state phenomenon, well, that's cool too and it makes me wonder about how it all works. Make sense?
I'm curious, what do you mean by "prefer"? Holding onto a theory because it happens to gel with what you want the universe to be like is dangerous.
Note that I am also an atheist, and I actually welcome the "Well, what would a beginning mean?" questions - mainly because I don't know, and that inspires wonder. It's what the universe is like that fascinates me, not what I want the universe to be like.
Knowing that you have to implement certain member (or global) functions in order for a templated class to compile is hardly arcane knowledge.
You're right, "member function" was the incorrect phrase. Sometimes they will be member functions, sometimes they will not. But, really, that wouldn't change what a constraint would look like in the language.
Requiring every class that supports == to derive from EqualityComparable is just a cumbersome extra for no benefit in a generic, C++ based world, where there are more ways to define and implement interfaces than just inheritance.
For the C++ world, I don't like that way either. If constraints were in C++, I think they should be done on a member function basis. That is, you explicity state which member functions must be implemented.
C++ is the opposite. In C++, you can do anything you damn well please on a variable of a type parameter type. But then once you instantiate it, it may not work, and you'll get some cryptic error messages. For example, if you have a type parameter T, and variables x and y of type T, and you say x + y, well you had better have an operator+ defined for + of two Ts, or you'll get some cryptic error message. So in a sense, C++ templates are actually untyped, or loosely typed. Whereas C# generics are strongly typed.
// . . .
I disagree with that assessment. Both C# and C++ generics/templates are strongly typed. It's just that the enforcement happens in different places.
In C++, if you try to stick a class into a templated class when that class doesn't have a particular member function defined, the compiler will yell at you, just like Hejlsberg said. But for some reason, this doesn't count as type checking? Yes, template error messages can be strange (and very long) if you're not familiar with them. But that's just a lesson in "know your tools."
To me, "strongly typed" is strict type enforcement at compile time. C++ templates certainly do this.
Constraints, however, are something that I think are a generally good idea. Stroustrup's reasoning for not including them in C++ was that "Requiring the user to provide such information decreases the flexibility of the parameterization facility without easing the implementation or increasing the safety of the facility." (The Design and Evolution of C++, Stroustrup, 343).
He does, however, show an interesting way to get around this using inheritance:
template <class T>
class Comparable {
T& operator=(const T&);
int operator==(const T&, const T&);
int operator<=(const T&, constT&);
int operator<(const T&, const T&);
};
template <class T : Comparable>
class vector {
};
(The D&E of C++, Stroustrup, 344)
This technique is similar to how C# does constraints, class List where T: IComparable. One is supported and enforced by the language, the other is a natural consequence of the languages facilities. In general, I think that constraints are probably a good thing. Having an error message like "Can not instantiate class Y<T> because T does not implement z()" is probably best, and when looking at a class' declaration, it would be nice to see up front what assumptions the templated class makes.
No, that just makes me lazy.
Okay, I'll bite. Read how the aglorithms in the STL are implemented. Stroustrup's book has excellent examples that are more digestable than cracking open the STL source if you don't feel like approaching that.
No, not a troll. When someone says something like "my opinion of operator overloading is that it is absolutely bad," I hope they're trolling, because it's very ignorant to say.
As an undergrad, I went to Virginia Tech. Now I'm a grad student (and a TA) at William & Mary. All of my courses at Tech were taught by full instructors or professors. I had two math classes taught by advanced grad students (they were Ph.D. candidates at that point). Here at W&M, the only course I know of that has been taught by a TA is the intro to CS for non-majors course (again, Ph.D. candidate).
Tech is a big university; W&M is a small one. I would have been shocked to walk into a junior level CS course at Tech and see a grad student teaching it. At that point, in fact, instructors didn't teach the courses anymore, only professors. Here at W&M, it's almost entirely professors from beginning to end.
Where did you go?
The problem is that computer scientists don't make good programmers and vice-versa.
Bogus. CS, like other fields, has a diverse makeup. Some CS people do almost exclusively theory. Others specialize in software engineering. Some do systems, some networks, some programming languages. I can think of some CS professors that aren't good programmers, and I can think of some that are phenomenal.
PHd professor types that staff CompSci departments I've been in tend to have stopped learning about computers as soon as they finished their doctorate and instead concentrate on internecine politics, incomprehensible papers, and teaching the occaisional class (leaving most of that to T.A.'s who actually teach the class and understand how to compile programs).
Ditto. You just characterized the entirety of a field from your limited experience. Would you characterize all physicists from talking to a few theoretical physicists? There are experimental physicists out there too. How about soemone who specialized in quantum mechanics? There's E&M people too. Your claims are just plain silly.
Oh, and I am a TA. I have never heard of a school where TAs teach actual lecture. I teach lab. That is, I lecture for about 20 minutes, then let the undergrads do their programming lab and answer questions along the way. This is a far cry from actually teaching the course.
Tell me you're not currently in a CS program.
Um, STL.
Not only does operator overloading have functional value, it allows functional programming. [rimshot]
That is, in C++, overloading operator () allows you to define function objects. Function objects allow you to simulate functions as a first-class entity, allowing functional style programming in C++. Check out the Functional C++ Library, FC++, it's cool stuff.