"Do we really want the gov (at any level) to start getting their hands in this?"
Um... The Internet was _created_ by the military (DARPA); which is part of our government. It used to be called ARPA Net.
DARPA decided it didn't want to be the owner of the Internet and stuck the Department of Commerce with the problem. DoC didn't want the headache either and set up ICANN; which it has been trying to hand the Internet to for years.
It seems to me that the gov already has its hands into this rather deeply, and now is not the time to let them wash their hands.
I bet any other TLDs started up with an agreement that they could do it. VeriSign has no such agreement, and it would be unthinkable to allow such an agreement in the case of the.com and.net domains.
Basically, VeriSign is making a power grab. The only appropriate response to such a power grab is to kill the company making the grab. It's time to bankrupt VeriSign.
So what if it hurts camera operators, key grips, makeup artists, and costumers? They're not going to get any sympathy from someone doing merchandizing after having been a programmer for 20 years.
To the extent that he is the "ideological" leader of anyone I know it's always for his basic folksy refusal to be anyone's ideological leader. We like him. We don't "follow" him.
Did I miss a meeting or a memo or something?"
If "Show me the code" isn't an ideology then what is an ideology? Sure, Stallman created the fundament of Free Software -- just like John the Baptist -- but it was Linus who shoved the 'path' up their noses.
I follow Linus in the same sense I'd follow someone up a mountain path. Linus is ahead of me in line.
Bill's running a boom-and-bust operation. His is a socially unacceptable short-sighted viewpoint -- like most of Big Business these days. Only in it for the moment, and without regard to the long-term implications.
There are companies out there which take the long view, don't try for explosive growth, and do very well over time. Why should we take shit from the sharks?
"The Verizon case found that the RIAA has the right to get the identities of users who they allege are violating copyright laws by sharing copyrighted music."
Well, actually, it only found that the RIAA has the right in one circuit (2nd?). The precident doesn't have legal force in the others -- although the others can use it in making up their minds.
I was 36 when I got a driver's licence. When the office moved and I had to take 4 busses to get to work I was getting a lot of reading done, but everyone else was getting pissed -- so shove it.
Thanks. Between this case and the abortion clinic cases it has been pretty well codified that free speech is the right to speak and be heard by willing listeners. The word 'willing' is very important.
To some extent the 'willing' comes from the right to assemble. If speech was not limited by the willingness of the recipient then you could use free speech rights to disrupt an attempt to assemble.
" (b) "California electronic mail address" or "California e-mail address" means any of the following:
(1) An e-mail address furnished by an electronic mail service provider that sends bills for furnishing and maintaining that e-mail address to a mailing address in this state.
(2) An e-mail address ordinarily accessed from a computer located in this state.
(3) An e-mail address furnished to a resident of this state."
Note especially (2). How hard would it be for me to ordinarily access my email from a computer located in CA? In fact, how hard would it be for comcast (my provider) to set things up so that my POP is in CA? (Yeah, I know that an obvious reading of the law doesn't make it clear....)
If other states don't follow with similar laws this loss will be offset by all the people moving their email boxes to CA. I got as far as having read the definitions in the law, and they're real doozies. Whoever wrote this knows exactly what is going on.
I haven't kept up to date on ICANN, but the Dept of Commerce was trying to get out of the loop. There's been a continuing low-level war over the root servers for years.
Either Commerce still holds contracting authority, or it has been moved over to ICANN.
"I think it's a bit dishonest to try to lump C in with PHP and Perl, which I would think was the implication of "weakly typed" in our context (or such was my assumption)."
I learned to think in Lisp and then went on to program in PDP-11 Assembler (sort of the same as Lisp). I didn't lump them in with PHP and Perl -- although Perl is kind of neat. Rather than thinking in terms of weakly typed languages, think in terms of unbound types. You get better code by overlaying the type you need at the moment over the bits than by modifying the bits to fit the type.
C++ does have some nice features, but strong typing isn't one of them. In fact, strong typing is the major reason I refused to program in Pascal and Ada.
Objects are stupid too. They are almost the exact reverse of what they should be. An object should have no basic type and should have types added to it as necessary.
"Keeping track of types is not a computationally difficult problem. If ones' algorithm or program depends on poorly defined input or ouput types, then it is a poorly defined algorithm. Generalization of algorithms is handled by polymorphism, not dynamic typing or (worse yet) weak typing."
Keeping track of types is a computationally impossible problem. The denoted type is not the connotated type. Have you ever tried to correctly splice a polymorphism into the middle of a heirarchy? It's easier to start over.
You seem to have no concept of the fact that the problem domain tends to mutate over time. Spagetti code is bad enough. Spagetti types fail in unknowable ways.
For the record, I started in 1971 and while I used LISP, Fortran, and the like I did most of my programming in PDP-11 Assembler until 1984. I then moved over to mostly C; which is not a strongly typed language. I've done most of my programming in C++ from 1997 until now; which is a strongly typed language. My work has always involved commercial products with repeated release cycles. C is a better language than C++.
I'm rather aware of the stopping problem. There is no such thing as provable correctness of a language -- at least not if the language has the power of a turing machine. There might be provable correctness of an algorithm if it is simple enough.
I agree that strong typing can catch simple errors, but it does so at the expense of making the program more complicated -- introducing more opportunity for complex errors. There is no way of getting around the fact that programming is computationally difficult. If programming didn't have mathematically unsolvable problems computers would have become intellegent decades ago.
The assignment for today, class, is to read the private key off any consumer DRM chip. If you don't have an electron microscope you can use one of the high school's microscopes.
Extra credit will be given to anyone who can open the chip and read the key in under ten minutes.
Please remember that it is illegal to publish this number online.
I've been at this programming thing long enough to know that everything Graham said is completely correct.
In fact, strong typing is stupid. If you want to treat a byte as a char in one statement and as a bit vector in the next you shouldn't have to beg the computer to let you do exactly what it does on its own. (Study the bit structure of ASCII sometime, and you'll understand.)
Slashdot Mirror
"Do we really want the gov (at any level) to start getting their hands in this?"
Um... The Internet was _created_ by the military (DARPA); which is part of our government. It used to be called ARPA Net.
DARPA decided it didn't want to be the owner of the Internet and stuck the Department of Commerce with the problem. DoC didn't want the headache either and set up ICANN; which it has been trying to hand the Internet to for years.
It seems to me that the gov already has its hands into this rather deeply, and now is not the time to let them wash their hands.
I bet any other TLDs started up with an agreement that they could do it. VeriSign has no such agreement, and it would be unthinkable to allow such an agreement in the case of the .com and .net domains.
Basically, VeriSign is making a power grab. The only appropriate response to such a power grab is to kill the company making the grab. It's time to bankrupt VeriSign.
So what if it hurts camera operators, key grips, makeup artists, and costumers? They're not going to get any sympathy from someone doing merchandizing after having been a programmer for 20 years.
Let the whole of Big Media rot.
I don't watch Hollywood movies anyway.
""Show me the code."
To the extent that he is the "ideological" leader of anyone I know it's always for his basic folksy refusal to be anyone's ideological leader. We like him. We don't "follow" him.
Did I miss a meeting or a memo or something?"
If "Show me the code" isn't an ideology then what is an ideology? Sure, Stallman created the fundament of Free Software -- just like John the Baptist -- but it was Linus who shoved the 'path' up their noses.
I follow Linus in the same sense I'd follow someone up a mountain path. Linus is ahead of me in line.
Bill's running a boom-and-bust operation. His is a socially unacceptable short-sighted viewpoint -- like most of Big Business these days. Only in it for the moment, and without regard to the long-term implications.
There are companies out there which take the long view, don't try for explosive growth, and do very well over time. Why should we take shit from the sharks?
"The Verizon case found that the RIAA has the right to get the identities of users who they allege are violating copyright laws by sharing copyrighted music."
Well, actually, it only found that the RIAA has the right in one circuit (2nd?). The precident doesn't have legal force in the others -- although the others can use it in making up their minds.
I was 36 when I got a driver's licence. When the office moved and I had to take 4 busses to get to work I was getting a lot of reading done, but everyone else was getting pissed -- so shove it.
Driving might have been a privilege in 1900, but these days it's more of an obligation.
Thanks. Between this case and the abortion clinic cases it has been pretty well codified that free speech is the right to speak and be heard by willing listeners. The word 'willing' is very important.
To some extent the 'willing' comes from the right to assemble. If speech was not limited by the willingness of the recipient then you could use free speech rights to disrupt an attempt to assemble.
From the bill:
" (b) "California electronic mail address" or "California e-mail
address" means any of the following:
(1) An e-mail address furnished by an electronic mail service
provider that sends bills for furnishing and maintaining that e-mail
address to a mailing address in this state.
(2) An e-mail address ordinarily accessed from a computer located
in this state.
(3) An e-mail address furnished to a resident of this state."
Note especially (2). How hard would it be for me to ordinarily access my email from a computer located in CA? In fact, how hard would it be for comcast (my provider) to set things up so that my POP is in CA? (Yeah, I know that an obvious reading of the law doesn't make it clear....)
If other states don't follow with similar laws this loss will be offset by all the people moving their email boxes to CA. I got as far as having read the definitions in the law, and they're real doozies. Whoever wrote this knows exactly what is going on.
It goes after the advertisers, not just the spammers.
My son never bothered to rearrange the keys.
I haven't kept up to date on ICANN, but the Dept of Commerce was trying to get out of the loop. There's been a continuing low-level war over the root servers for years.
Either Commerce still holds contracting authority, or it has been moved over to ICANN.
You sure made a lot of assumptions there. Why would I be using ASCII for message strings?
"I think it's a bit dishonest to try to lump C in with PHP and Perl, which I would think was the implication of "weakly typed" in our context (or such was my assumption)."
I learned to think in Lisp and then went on to program in PDP-11 Assembler (sort of the same as Lisp). I didn't lump them in with PHP and Perl -- although Perl is kind of neat. Rather than thinking in terms of weakly typed languages, think in terms of unbound types. You get better code by overlaying the type you need at the moment over the bits than by modifying the bits to fit the type.
C++ does have some nice features, but strong typing isn't one of them. In fact, strong typing is the major reason I refused to program in Pascal and Ada.
Objects are stupid too. They are almost the exact reverse of what they should be. An object should have no basic type and should have types added to it as necessary.
"Keeping track of types is not a computationally difficult problem. If ones' algorithm or program depends on poorly defined input or ouput types, then it is a poorly defined algorithm. Generalization of algorithms is handled by polymorphism, not dynamic typing or (worse yet) weak typing."
Keeping track of types is a computationally impossible problem. The denoted type is not the connotated type. Have you ever tried to correctly splice a polymorphism into the middle of a heirarchy? It's easier to start over.
You seem to have no concept of the fact that the problem domain tends to mutate over time. Spagetti code is bad enough. Spagetti types fail in unknowable ways.
First IBM. Now china. I think I've woken up in bizarro-world.
For the record, I started in 1971 and while I used LISP, Fortran, and the like I did most of my programming in PDP-11 Assembler until 1984. I then moved over to mostly C; which is not a strongly typed language. I've done most of my programming in C++ from 1997 until now; which is a strongly typed language. My work has always involved commercial products with repeated release cycles. C is a better language than C++.
I'm rather aware of the stopping problem. There is no such thing as provable correctness of a language -- at least not if the language has the power of a turing machine. There might be provable correctness of an algorithm if it is simple enough.
I agree that strong typing can catch simple errors, but it does so at the expense of making the program more complicated -- introducing more opportunity for complex errors. There is no way of getting around the fact that programming is computationally difficult. If programming didn't have mathematically unsolvable problems computers would have become intellegent decades ago.
The assignment for today, class, is to read the private key off any consumer DRM chip. If you don't have an electron microscope you can use one of the high school's microscopes.
Extra credit will be given to anyone who can open the chip and read the key in under ten minutes.
Please remember that it is illegal to publish this number online.
"...unless there are errors in the implementation."
Which there will be. Aren't the Windoze people fond of reminding us that Linux isn't completely secure too?
The difference with DRM is that they won't be able to patch.
I've been at this programming thing long enough to know that everything Graham said is completely correct.
In fact, strong typing is stupid. If you want to treat a byte as a char in one statement and as a bit vector in the next you shouldn't have to beg the computer to let you do exactly what it does on its own. (Study the bit structure of ASCII sometime, and you'll understand.)
Um, Sony.