Hey man, there are technological fixes for these bugs. He is pointing out the weaknesses. On-line games aren't exactly critical infrastructure, but it is still important to know what the problems are so that they can be fixed or avoided in the future. In any case, the DMCA is ridiculous here; he's not circumventing any technological measures, and there are no copyrighted works being accessed. The DMCA does not outlaw hacking.
Fuck, man, nothing is "necessary" for a progamming task. I can code up any feature in C, just as I can code up any feature in ML. But partial application and closures are a extremely painful in C++. (Yes, I have seen the Boost library.)
Here's what this article says, distilled by the cynic:
If you define a binary "greater" function template, defined in terms of the operator, then it will not do the "right thing" for all types.
Wow, surprise!! But that's precisely because there is more than one "right thing" for comparison to do. What is the specification for the greater function? How could you even write one? What if we *want* greater to do pointer inequality or case-insensitive or lexicographic ordering? Distinguishing the types in your program and the operations on them is the key to an understandable program. That's one of the many reasons why overloading and template specialization are a problem.
As a technical tool to teach template specialization, okay, fine. But this is not good programming practice.
You know what's pretty funny? All of the chess commentators are saying that he's putting himself at a disadvantage, or "torturing himself" by doing it with the stupid 3D virtual reality glasses. That's probably not what X3D wanted to hear...
Damn, really? It was much harder for me to recompile sshd several times in the span of a few days than it was to go to windowsupdate and get the patches installed automatically. That one was a pain in the ass. (I suppose if RPM still worked on my system I could have used that, but it's not my fault that it's broken...)
"All copy-protections can be hacked, but if (we) give people what they are asking for in terms of value, they won't go out and steal it. It's called trusting the consumer."
Other than the inappropriate use of "steal," what's wrong with this? It's exactly the kind of sentiment that I want to see from a record company. Provide better quality sound (no compression artifacts), packaging, make it easier to get than on a P2P service, convince me that the artist gets a significant portion of that money, and if it's cheap enough, I'll buy it. (For many of the CDs I buy, that's already true! Just not BMG...)
However, something tells me that the DRM files on the CD aren't going to be what people are asking for in terms of value.
I'm no zealot, in fact, I really like a lot of the Apple products, but stating that the G5 is the world's fastest personal computer is really just a lie.
> But in the process, the truth is already out there. You can't censor the public record.
Actually, the record or parts of it can be sealed if it contains, say, trade secrets. If they win the case that might happen. Either way, it will damage their reputation to be sued over the memos.
I probably agree with this guy, but he flies off the handle a bit much. In several of his responses he is swearing at people, demanding that they never reply to him, etc. He came across, to me, as a total prima donna. He issued his ultimatum on Oct 22, and demanded CVS access in two months, or else he'd take his toys and go home... but then apparently shortened the deadline to five days. (?)
That said, the X development model seems pretty fucked up...
Re:Does adding every ingredient make it better?
on
C# 2.0 Spec Released
·
· Score: 1
Nobody is saying real-world experience isn't important, just that University is not the place to get it. (And nor should it be!)
Re:Does adding every ingredient make it better?
on
C# 2.0 Spec Released
·
· Score: 1
So don't use things that can corrupt memory like that.
You can't just not use it, because it is easy to screw up, even when you program the way you suggest. Or, other programmers in your project can screw up. Or, the libraries you use can screw up. C++ does not provide a way to isolate faults to a specific module. (Unit testing can help, but that is not part of the language.)
Also, garbage collection in C++ is typically worse than in languages with built-in support, because the GC has to be conservative. So you do indeed get worse performance than in a language built for it.
C++ can do anything in a clean way, you just have to know how to use it.
For example, C++ has terrible support for functional programming, because it lacks nested anonymous functions. Yes, you can code this up, like you can code up any feature in any language, but it is not clean. Simulating algebraic datatypes with class hierarchies is equally ridiculous.
Re:Does adding every ingredient make it better?
on
C# 2.0 Spec Released
·
· Score: 1
From what I understand, such a limit has nothing to do with compliance...
(from the C++ ANSI/ISO std.)"The bracketed number following each quantity is recommended as the minimum for that quantity. However, these quantities are only guidelines and do not determine compliance.
In any case, you gotta admit, this is the worst kind of "fix" for such a problem.
The problem with perl extends far beyond its syntax!
Re:Does adding every ingredient make it better?
on
C# 2.0 Spec Released
·
· Score: 1
As far as I know there are no legal programs that don't compile
???
that still doesn't make it impossible to compile
No, you don't understand what I'm saying. I am saying that some legal C++ programs (ie, those that use template metaprogramming) cannot be compiled (they instruct the compiler to loop forever) and that it is not possible for a compiler to distinguish those programs from the programs that can be compiled successfully.
If you think perl is nice, well, perhaps there is no hope...
Re:Does adding every ingredient make it better?
on
C# 2.0 Spec Released
·
· Score: 1
You've been out of school for a while, haven't you?
No, I finished in CS undergrad in 2001 and have been a grad student since, and I have TA'd several classes since then. Perhaps it's not true at all schools, but it *should* be.
Re:Does adding every ingredient make it better?
on
C# 2.0 Spec Released
·
· Score: 1
You are kidding right? My college teaches every programming class from CS101 to CS412-GUI_design with Java. You can get your BSCS without ever using C, VB, scheme, etc. Just Java.
I said they don't teach languages except as tools to teach more important concepts. The fact that they never force you to use any language seems to support that claim--the language is just a tool, so they might as well standardize on a tool.
However, it is hard to do system tasks like OS programming in Java, so it's a poor tool. If your program doesn't require any system level programming, that's kind of a bummer.
Re:Does adding every ingredient make it better?
on
C# 2.0 Spec Released
·
· Score: 1
Which university do you go to? I'll transfer.
CMU. It's in my web link right there, man.
Re:Does adding every ingredient make it better?
on
C# 2.0 Spec Released
·
· Score: 1
Power also includes the ability to isolate faults, something that C++ absolutely does not provide. Any module in your program can screw up the memory space of any other one. To me, this is more important than being able to "access the machine", something which is available anyway in any real implementation of a general purpose language, and something which is hardly ever necessary in an application.
Re:Does adding every ingredient make it better?
on
C# 2.0 Spec Released
·
· Score: 1
What do you mean by broken? Broken implies to me "not working," and last I checked, both of those languages work predictably.
Some examples of why C++ is broken:
- Compilation is an impossible task. By that I mean that there are some legal programs that cannot be compiled, and some legal programs that can be compiled, and no compiler can tell them apart. (This is due to the Turing-complete template system.)
- There is exactly 1 standard-compliant implementation of C++ that exists, and you probably don't use it. The standard is so convoluted that most implementors just make up their own C++-like language to implement instead.
Why Perl is broken:
- Compilation is impossible, as above.
- There exists one implementation, which defines the language.
- Seriously, you don't think perl is broken?? (I don't program in it, so I don't have any specific examples, but they always come up with my programming language friends who dare...)
Complicated is mostly opinion-based, so I'd love you know what "not complicated" languages you're comparing them to.
I think Java is a less complicated language. It's kind of tedious to program in, though, because it lacks certain useful features. (Two of which are apparently being added in C#) I think SML is a fairly simple language, but one that is quite powerful despite its simplicity. In fact, I like to program in it much more than in C++. (However, it remains difficult to learn for many programmers because there is not much that can be learned by analogy with, say, C.)
Re:Does adding every ingredient make it better?
on
C# 2.0 Spec Released
·
· Score: 1
The point of a computer science education is no more to make you ready for a programming job than a degree in mathematics is to make you ready to be an accountant.
Anyway, I didn't say you only need a superficial understanding of the language. A good program teaches you what you need to know to effectively make use of the concepts of the class (if the class is a programming class at all!) but does not waste time going through each feature in the language/toolkit like some tour of the zoo. With a solid foundation it's easy to pick that other stuff up.
Re:Does C# have continuations?
on
C# 2.0 Spec Released
·
· Score: 4, Informative
You should get out more. There's a world of programming paradigms most people have never heard of, because they're still stuck using C-alike block structured OO languages.
Continutions are, roughly speaking, a generalization of setjmp and longjmp in C. However, to have true "first-class" continuations they need to be objects that you can pass around, store in data structures, etc. In C this isn't true, because if you return from the stack frame that did the setjmp, the continuation is invalidated. Lisp has "call/cc", some implementations of ML have "calcc" (typed), and many scripting languages have it, because it's pretty easy to implement in an interpreted language.
Continuations can be used to implement exceptions, user-level thread packages, "early exits" from recursive code, and other cool stuff.
Slashdot Mirror
Hey man, there are technological fixes for these bugs. He is pointing out the weaknesses. On-line games aren't exactly critical infrastructure, but it is still important to know what the problems are so that they can be fixed or avoided in the future.
In any case, the DMCA is ridiculous here; he's not circumventing any technological measures, and there are no copyrighted works being accessed. The DMCA does not outlaw hacking.
Fuck, man, nothing is "necessary" for a progamming task. I can code up any feature in C, just as I can code up any feature in ML. But partial application and closures are a extremely painful in C++. (Yes, I have seen the Boost library.)
I don't mean object in the "object oriented" sense, sorry for the confusion.
Amen! Not to mention that you then have a first class object, rather than some bizarre "meta" object...
Here's what this article says, distilled by the cynic:
If you define a binary "greater" function template, defined in terms of the operator, then it will not do the "right thing" for all types.
Wow, surprise!! But that's precisely because there is more than one "right thing" for comparison to do. What is the specification for the greater function? How could you even write one? What if we *want* greater to do pointer inequality or case-insensitive or lexicographic ordering? Distinguishing the types in your program and the operations on them is the key to an understandable program. That's one of the many reasons why overloading and template specialization are a problem.
As a technical tool to teach template specialization, okay, fine. But this is not good programming practice.
You know what's pretty funny? All of the chess commentators are saying that he's putting himself at a disadvantage, or "torturing himself" by doing it with the stupid 3D virtual reality glasses. That's probably not what X3D wanted to hear...
Damn, really? It was much harder for me to recompile sshd several times in the span of a few days than it was to go to windowsupdate and get the patches installed automatically. That one was a pain in the ass. (I suppose if RPM still worked on my system I could have used that, but it's not my fault that it's broken...)
"All copy-protections can be hacked, but if (we) give people what they are asking for in terms of value, they won't go out and steal it. It's called trusting the consumer."
Other than the inappropriate use of "steal," what's wrong with this? It's exactly the kind of sentiment that I want to see from a record company. Provide better quality sound (no compression artifacts), packaging, make it easier to get than on a P2P service, convince me that the artist gets a significant portion of that money, and if it's cheap enough, I'll buy it. (For many of the CDs I buy, that's already true! Just not BMG...)
However, something tells me that the DRM files on the CD aren't going to be what people are asking for in terms of value.
I'm no zealot, in fact, I really like a lot of the Apple products, but stating that the G5 is the world's fastest personal computer is really just a lie.
What the fuck? This article is from 1998. Is that a typo? Is this the Slashdot wayback machine?
> But in the process, the truth is already out there. You can't censor the public record.
Actually, the record or parts of it can be sealed if it contains, say, trade secrets. If they win the case that might happen. Either way, it will damage their reputation to be sued over the memos.
Of course there they were also *hacking* a linux box (or maybe it was OpenBSD?) in that movie, so...
Buyer's remorse... and a pretty case. ;)
I probably agree with this guy, but he flies off the handle a bit much. In several of his responses he is swearing at people, demanding that they never reply to him, etc. He came across, to me, as a total prima donna. He issued his ultimatum on Oct 22, and demanded CVS access in two months, or else he'd take his toys and go home... but then apparently shortened the deadline to five days. (?)
That said, the X development model seems pretty fucked up...
Nobody is saying real-world experience isn't important, just that University is not the place to get it. (And nor should it be!)
So don't use things that can corrupt memory like that.
You can't just not use it, because it is easy to screw up, even when you program the way you suggest. Or, other programmers in your project can screw up. Or, the libraries you use can screw up. C++ does not provide a way to isolate faults to a specific module. (Unit testing can help, but that is not part of the language.)
Also, garbage collection in C++ is typically worse than in languages with built-in support, because the GC has to be conservative. So you do indeed get worse performance than in a language built for it.
C++ can do anything in a clean way, you just have to know how to use it.
For example, C++ has terrible support for functional programming, because it lacks nested anonymous functions. Yes, you can code this up, like you can code up any feature in any language, but it is not clean. Simulating algebraic datatypes with class hierarchies is equally ridiculous.
From what I understand, such a limit has nothing to do with compliance...
(from the C++ ANSI/ISO std.) "The bracketed number following each quantity
is recommended as the minimum for that quantity. However, these quantities are only
guidelines and do not determine compliance.
[...]
- Recursively nested template instantiations [17]."
In any case, you gotta admit, this is the worst kind of "fix" for such a problem.
The problem with perl extends far beyond its syntax!
As far as I know there are no legal programs that don't compile
???
that still doesn't make it impossible to compile
No, you don't understand what I'm saying. I am saying that some legal C++ programs (ie, those that use template metaprogramming) cannot be compiled (they instruct the compiler to loop forever) and that it is not possible for a compiler to distinguish those programs from the programs that can be compiled successfully.
If you think perl is nice, well, perhaps there is no hope...
You've been out of school for a while, haven't you?
No, I finished in CS undergrad in 2001 and have been a grad student since, and I have TA'd several classes since then.
Perhaps it's not true at all schools, but it *should* be.
You are kidding right? My college teaches every programming class from CS101 to CS412-GUI_design with Java. You can get
your BSCS without ever using C, VB, scheme, etc. Just Java.
I said they don't teach languages except as tools to teach more important concepts. The fact that they never force you to use any language seems to support that claim--the language is just a tool, so they might as well standardize on a tool.
However, it is hard to do system tasks like OS programming in Java, so it's a poor tool. If your program doesn't require any system level programming, that's kind of a bummer.
Which university do you go to? I'll transfer.
CMU. It's in my web link right there, man.
Power also includes the ability to isolate faults, something that C++ absolutely does not provide. Any module in your program can screw up the memory space of any other one. To me, this is more important than being able to "access the machine", something which is available anyway in any real implementation of a general purpose language, and something which is hardly ever necessary in an application.
What do you mean by broken? Broken implies to me "not working," and last I checked, both of those languages work predictably.
Some examples of why C++ is broken:
- Compilation is an impossible task. By that I mean that there are some legal programs that cannot be compiled, and some legal programs that can be compiled, and no compiler can tell them apart. (This is due to the Turing-complete template system.)
- There is exactly 1 standard-compliant implementation of C++ that exists, and you probably don't use it. The standard is so convoluted that most implementors just make up their own C++-like language to implement instead.
Why Perl is broken:
- Compilation is impossible, as above.
- There exists one implementation, which defines the language.
- Seriously, you don't think perl is broken?? (I don't program in it, so I don't have any specific examples, but they always come up with my programming language friends who dare...)
Complicated is mostly opinion-based, so I'd love you know what "not complicated" languages you're comparing them to.
I think Java is a less complicated language. It's kind of tedious to program in, though, because it lacks certain useful features. (Two of which are apparently being added in C#) I think SML is a fairly simple language, but one that is quite powerful despite its simplicity. In fact, I like to program in it much more than in C++. (However, it remains difficult to learn for many programmers because there is not much that can be learned by analogy with, say, C.)
The point of a computer science education is no more to make you ready for a programming job than a degree in mathematics is to make you ready to be an accountant.
Anyway, I didn't say you only need a superficial understanding of the language. A good program teaches you what you need to know to effectively make use of the concepts of the class (if the class is a programming class at all!) but does not waste time going through each feature in the language/toolkit like some tour of the zoo. With a solid foundation it's easy to pick that other stuff up.
You should get out more. There's a world of programming paradigms most people have never heard of, because they're still stuck using C-alike block structured OO languages.
Continutions are, roughly speaking, a generalization of setjmp and longjmp in C. However, to have true "first-class" continuations they need to be objects that you can pass around, store in data structures, etc. In C this isn't true, because if you return from the stack frame that did the setjmp, the continuation is invalidated. Lisp has "call/cc", some implementations of ML have "calcc" (typed), and many scripting languages have it, because it's pretty easy to implement in an interpreted language.
Continuations can be used to implement exceptions, user-level thread packages, "early exits" from recursive code, and other cool stuff.