Eh? There are lots of bindings for Qt. Binding to C++ isn't actually that hard once you get certain tricky things worked out. KDE has a library (libSMOKE) that abstracts out those tricky things, to make writing bindings much easer.
The question is: do you believe it is legitimate to protect people from themselves? I don't believe that it is. Could you imagine if the government passed legislation forcing people to install anti-virus software?
There are situations where the government should intervene to protect the population. The vast majority of these cases are:
- Where one person's failure affects everyone. For example, automotive regulations exist because if you crash your car, you could hurt other people. Laws against smoking in certain areas exist because of the dangers of second-hand smoke. Etc, etc.
- Where it is too onerous for a an individual to conduct sufficient research to make an informed decision. The FDA, for example, exists because it would be out of the capacity for an individual to do their own drug-safety testing.
I would argue that neither of these cases are true for privacy protection. A breech of your violation does not affect me. You choosing to do business with a firm that cannot properly manage your privacy does not affect my choosing to do business with a firm you can properly manage mine. Certainly, I would argue that it is not out of the capacity of individuals to research and see which companies can be trusted with privacy and which cannot.
Are 'reasonable' US laws any different from 'unreasonable' ones? When the world wants to create laws that they think everyone should follow, they'll do it through a UN treaty. That's the whole reason the UN exists --- it is the embodiement of a social contract between nations. The US cannot just become the de-facto ruler of the world and decree that certain US laws shall be laws in all countries. Technically, the US has the prerogative to do so, because a company could choose to just choose not do business in the US, rather than follow US laws everywhere, but it would be entirely against the spirit of international law.
The British Empire fell for the same reason all empires fall. Eventually, the colonies get sick and kick you out, and they can do so because of the logistical challenges of ruling so many disparate countries scattered over such a wide area.
It must be noted, however, that one of the prime reasons Britain was able to establish such an empire in the first place was that it embraced elements of a modern, free-market economy sooner than the other countries in Europe.
So basically, what you are saying, is that if you want to do business in the US, you have to follow US laws all over the world? That smacks of cultural imperialism if you ask me! The US can keep its laws in its own damn country. Certainly, I'd hate to see anything like PATRIOT or DMCA get spread any further than it already has!
India is doing great? In what alternate reality? Try to see beyond the tiny little area of IT work and see that India is really not doing very well.
China is, on the other hand, doing very well. However, they've also been moving strongly towards a free market economy. For example, they recently joined the WTO.
Historically, the countries that have embraced free trade (France, Great Britain, Germany, United States) have done better than those who have not. Indeed, if you study economic patterns in western Europe over the past six or seven centuries, you'll see a strong correlation between increases in free trade and increased economic power.
I would hate to see anybody overlegislate this. Remember, the Republican's in power aren't exactly free-marketer's themselves. They'll do stupid stuff, anti-conservative, stuff like support a steel tarrif just to pander to voters.
Now, there is a very good free-market solution to this: Do nothing. If people care about their privacy, they will make sure to deal only with companies that put measures in place to protect it. If people do not care about their privacy (and it appears to be the case that they do not) then they won't. However, just because other people would rather save $5 than keep their privacy does not mean that you have to.
Normally, I think the "buy American!" mentality is stupid, but in this case, there is a legitimate reason why "buying American" is in your best interest.
You don't have to be a neo-con to believe in the free market. You just have to have a brain. I'm a yellow-dog democrat, and I still realize that protectionism is bad.
"When a programming language is created that allows programmers to program in simple English, it will be discovered that programmers cannot speak English.
So what? At this point, I've given up ever hoping that the "programmers out ther ein the trenches" will ever wise up to the superior tools that are out there. There is no point in bringing those people into discussions about languages, because they'll never accept something unless its almost exactly the same as what they are already using. Witness C#, which is ever so slowly absorbing Lisp techniques. So far, it has GC, and in 2.0 will have lambda. In another decade, it'll get multimethods, another decade later it'll get macros. By about 2030, they will have matched Lisp circa 1980.
If you've ever used C# or Java, you can shut up promptly. Because those languages are slowly introducing advanced features in bite-sized pieces that you "real programmers" can handle.
Closures are everything! All the crap that modern languages do with special-purpose features, like generators, iterators, etc, can be done with closures. Hell, OOP can be done on top of closures.
I'm familiar with optimizing compilers for Lisp, but I still maintain its hard to write one:) It doesn't help that the best freely-available CLOS (CMUCL's, based on PCL) isn't exactly fast itself.
The first Lisp implementation, on the IBM 704, had special looping instructions with two 15-bit operands, "address" and "decrement." The machine had special instructions that would load 15 bits from a 36-bit word into the machine's index registers. Since a cons cell was represented by a single word with 2 15-bit pointer fields, the operations ended up being called "contents of address register" and "contents of decrement register." They thought of "first" and "rest" a short while later, but nobody wanted to switch.
'car' and 'cdr' are historic terms. Kinda like the 'creat' system call in UNIX, or A lot of people like them because of the nostolgic value, and the fact that they are short and can be typed with the fingers of one hand. You can always use 'first' and 'rest' if you want.
However, note that cons cells aren't always part of a list, they can (and often are) used as pairs of values. At that point, 'first' and 'rest' don't make a great deal of sense. Because of the generality of the cons cell, 'car' and 'cdr' makes as much sense as any other names.
And object[index] is a bit of a hack. What do you do when you've got a hash-table, and want to allow iterative access to keys and values seperately. You end up doing a function like:
object.key(index) or object.value(index)
That means a different syntax for accessing elements vs keys or values. With prefix syntax, you get a nice, uniform:
1) For the first 10 elements, you can just use "first," "second," "third," and so on.
2) You can also use the function (nthcdr) to get any of the other elements.
3) Since lists are also collections, you can use (elt) as well.
The parens don't bother people who have used Lisp for a decent amount of time. They find Lisp readable because they use good editors, which use syntax highlighting, parens matching, and auto-indentation to make code readable.
How does my analogy break down under those conditions? That was precisely my point! Programmers need to be well-versed with a range of tools, and know how to use them wisely.
The reason this new language is interpreted is probably because it has semantics similar to Python. Writing a good optimizing compiler for that is hard. You'd need a lot of the techniques introduced by Lisp compilers (type-inference, etc) to get a good implementation. In Python, everything is an object, and thus conceptually everything requires a dynamic dispatch.
Slashdot Mirror
Eh? There are lots of bindings for Qt. Binding to C++ isn't actually that hard once you get certain tricky things worked out. KDE has a library (libSMOKE) that abstracts out those tricky things, to make writing bindings much easer.
OUr long-arm statutes are fairly tame for good reason. We're supposed to be the freest country in the world, remember?
The question is: do you believe it is legitimate to protect people from themselves? I don't believe that it is. Could you imagine if the government passed legislation forcing people to install anti-virus software?
There are situations where the government should intervene to protect the population. The vast majority of these cases are:
- Where one person's failure affects everyone. For example, automotive regulations exist because if you crash your car, you could hurt other people. Laws against smoking in certain areas exist because of the dangers of second-hand smoke. Etc, etc.
- Where it is too onerous for a an individual to conduct sufficient research to make an informed decision. The FDA, for example, exists because it would be out of the capacity for an individual to do their own drug-safety testing.
I would argue that neither of these cases are true for privacy protection. A breech of your violation does not affect me. You choosing to do business with a firm that cannot properly manage your privacy does not affect my choosing to do business with a firm you can properly manage mine. Certainly, I would argue that it is not out of the capacity of individuals to research and see which companies can be trusted with privacy and which cannot.
Are 'reasonable' US laws any different from 'unreasonable' ones? When the world wants to create laws that they think everyone should follow, they'll do it through a UN treaty. That's the whole reason the UN exists --- it is the embodiement of a social contract between nations. The US cannot just become the de-facto ruler of the world and decree that certain US laws shall be laws in all countries. Technically, the US has the prerogative to do so, because a company could choose to just choose not do business in the US, rather than follow US laws everywhere, but it would be entirely against the spirit of international law.
The British Empire fell for the same reason all empires fall. Eventually, the colonies get sick and kick you out, and they can do so because of the logistical challenges of ruling so many disparate countries scattered over such a wide area.
It must be noted, however, that one of the prime reasons Britain was able to establish such an empire in the first place was that it embraced elements of a modern, free-market economy sooner than the other countries in Europe.
So basically, what you are saying, is that if you want to do business in the US, you have to follow US laws all over the world? That smacks of cultural imperialism if you ask me! The US can keep its laws in its own damn country. Certainly, I'd hate to see anything like PATRIOT or DMCA get spread any further than it already has!
India is doing great? In what alternate reality? Try to see beyond the tiny little area of IT work and see that India is really not doing very well.
China is, on the other hand, doing very well. However, they've also been moving strongly towards a free market economy. For example, they recently joined the WTO.
Historically, the countries that have embraced free trade (France, Great Britain, Germany, United States) have done better than those who have not. Indeed, if you study economic patterns in western Europe over the past six or seven centuries, you'll see a strong correlation between increases in free trade and increased economic power.
I would hate to see anybody overlegislate this. Remember, the Republican's in power aren't exactly free-marketer's themselves. They'll do stupid stuff, anti-conservative, stuff like support a steel tarrif just to pander to voters.
Now, there is a very good free-market solution to this: Do nothing. If people care about their privacy, they will make sure to deal only with companies that put measures in place to protect it. If people do not care about their privacy (and it appears to be the case that they do not) then they won't. However, just because other people would rather save $5 than keep their privacy does not mean that you have to.
Normally, I think the "buy American!" mentality is stupid, but in this case, there is a legitimate reason why "buying American" is in your best interest.
You don't have to be a neo-con to believe in the free market. You just have to have a brain. I'm a yellow-dog democrat, and I still realize that protectionism is bad.
A quote I really like:
"When a programming language is created that allows programmers to program in simple English, it will be discovered that programmers cannot speak English.
- Anonymous"
So what? At this point, I've given up ever hoping that the "programmers out ther ein the trenches" will ever wise up to the superior tools that are out there. There is no point in bringing those people into discussions about languages, because they'll never accept something unless its almost exactly the same as what they are already using. Witness C#, which is ever so slowly absorbing Lisp techniques. So far, it has GC, and in 2.0 will have lambda. In another decade, it'll get multimethods, another decade later it'll get macros. By about 2030, they will have matched Lisp circa 1980.
Dylan is very not prototyped-based. Its class-based. Extremely cool language, though.
Acedemic morons in their ivory towers.
I bet the learn to spell in those ivory towers.
Check this presentation. Page 2.
If you've ever used C# or Java, you can shut up promptly. Because those languages are slowly introducing advanced features in bite-sized pieces that you "real programmers" can handle.
Closures are everything! All the crap that modern languages do with special-purpose features, like generators, iterators, etc, can be done with closures. Hell, OOP can be done on top of closures.
I'm familiar with optimizing compilers for Lisp, but I still maintain its hard to write one :) It doesn't help that the best freely-available CLOS (CMUCL's, based on PCL) isn't exactly fast itself.
The first Lisp implementation, on the IBM 704, had special looping instructions with two 15-bit operands, "address" and "decrement." The machine had special instructions that would load 15 bits from a 36-bit word into the machine's index registers. Since a cons cell was represented by a single word with 2 15-bit pointer fields, the operations ended up being called "contents of address register" and "contents of decrement register." They thought of "first" and "rest" a short while later, but nobody wanted to switch.
Um, in a hash-table, [] is usually already taken for the lookup function. So you can't use index, which should be an integer for iterative access.
'car' and 'cdr' are historic terms. Kinda like the 'creat' system call in UNIX, or A lot of people like them because of the nostolgic value, and the fact that they are short and can be typed with the fingers of one hand. You can always use 'first' and 'rest' if you want.
However, note that cons cells aren't always part of a list, they can (and often are) used as pairs of values. At that point, 'first' and 'rest' don't make a great deal of sense. Because of the generality of the cons cell, 'car' and 'cdr' makes as much sense as any other names.
What's wrong with (elt object index)?
And object[index] is a bit of a hack. What do you do when you've got a hash-table, and want to allow iterative access to keys and values seperately. You end up doing a function like:
object.key(index) or object.value(index)
That means a different syntax for accessing elements vs keys or values. With prefix syntax, you get a nice, uniform:
(key object index) or (value object index)
Blargh. The function isn't (nthcdr) its (nthcar). (nthcdr) gets the cdr rather than the car.
Okay, how much Lisp do you know? I ask because:
1) For the first 10 elements, you can just use "first," "second," "third," and so on.
2) You can also use the function (nthcdr) to get any of the other elements.
3) Since lists are also collections, you can use (elt) as well.
The parens don't bother people who have used Lisp for a decent amount of time. They find Lisp readable because they use good editors, which use syntax highlighting, parens matching, and auto-indentation to make code readable.
How does my analogy break down under those conditions? That was precisely my point! Programmers need to be well-versed with a range of tools, and know how to use them wisely.
The reason this new language is interpreted is probably because it has semantics similar to Python. Writing a good optimizing compiler for that is hard. You'd need a lot of the techniques introduced by Lisp compilers (type-inference, etc) to get a good implementation. In Python, everything is an object, and thus conceptually everything requires a dynamic dispatch.