It was (and remains) an explicit goal of XPCOM to be compatible with COM on Windows. Of course, there are no open source COM implementations on other platforms, so Mozilla.org had no choice but either to roll their own COM clone, or else go with some CORBA solution. Given that XPCOM is used for very frequently accessed components (e.g. file access APIs, graphics calls), I think it was the right decision to avoid CORBA.
> As I understand it, the biggest difference > between the MPL and the GPL is that with the MPL > you have to assign the changes back to Netscape.
The *NPL* gives Netscape special rights to redistribute your changes in closed source products (not the same thing as assigning changes back to Netscape). Substantial new code (i.e. new files) being added to Mozilla by third parties is generally being added under the *MPL*, which is similar to the NPL but gives Netscape no special rights.
> As I further understand it, this is because > Netscape wants to make money from your work (if > it was to "benefit the community" they would let > you distribute your own copy of Moz)
They do let you distribute your own copy of Moz (both NPL and MPL licensed code), so your comment is basically wrong.
It's true that Netscape wants to be able to release closed-source products based on their own source code, even after it has been modified by others, thus the NPL. This does not seem to be a big deal.
Even RMS agrees that the NPL and MPL are free software licenses.
Almost all the source code has been, or soon will be, released. Only the parts specific to RSA await the expiry of the patent. Until then, you can substitute your own RSA implementation (taken from, say, OpenSSL) and build your own binary from these sources. OK, it would be illegal if you're in the US, but you can do it.
If this is a rountine evolutionary step, then where are all these life forms? Hiding themselves because they're shy? Interstellar space isn't that big if you have a few million years to make the trip.
Same problem exists for everyone who thinks that advanced life forms are "easy".
You're making the common mistake of assuming that a "neural network" as stuided in computer science bears more than a passing resemblance to an actual system of biological neurons.
"Bad news, Mr Jones. I, script kiddie, just exploited a buffer overrun to get root on your brain emulator. As of now, I AM YOUR GOD. I control every aspect of your reality. Worship me."
This works for any VR setup that you can't remove or take off, as well. Count me out.
The only ones that sound doable in less than 30 years are the ones that would vaporize your brain instantly in a burst of high-energy radiation. Good luck.
Penrose's arguments about computability are fallacious. The problem is that we don't actually know that human mathematical insight is noncomputable. We may well discover systems with Godel-style statements that are true, but whose truth we cannot decide.
However, there is still a (perhaps small) possibility that he is right in claiming our mental processes rely partly on quantum processes. If so, then our mental state probably could not be duplicated accurately, because it would be corrupted in the process of reading it.
OK, the anthropic problems can be gotten around using exotic probabilistic techniques, but that tech would make Kurzweil's brain scanning look completely trivial.
I haven't heard of any SSL patches for Mozilla to date. Do you have any references? I don't think so. The Cryptozilla hack done by Eric Young and others early in '98 was more a proof of concept than anything else, and their code is completely useless in the rewritten codebase.
It's true that someone could have, and could still, produce an SSL plugin for Mozilla based on OpenSSL or something like that. They haven't though. But the big breakthrough is that we can now talk about the issues, standardize the APIs, and leverage a lot of the code Netscape has already written. Also Netscape will be releasing binaries which will give US users access to an RSA-licensed implementation.
RSA isn't that much code. It should be almost no effort to produce an "international" patch with the required functions.
Also, note that Netscape will be releasing a precompiled binary containing a licensed implementation of the RSA code. So non-US developers will be able to build their own binaries, and US users can download and use Netscape's.
You won't find an electable candidate whom you agree with on every issue. That is simply human nature. It is necessary to prioritize the issues and weight candidates' positions accordingly.
I personally think campaign finance reform is of paramount importance; it's hard to make progress on any issue when the political system is fundamentally corrupt. Thus, although I can't vote, if I could, I would vote for McCain --- even though I detest his positions on net censorship and other issues. (McCain has the advantage over Bradley that he doesn't just talk the talk, he's walked the walk for reform... politicians' promises should be discounted.)
It helps that it seems pretty clear that *his* number #1 issue is also C.F.R., and according to this/. article he doesn't spend a lot of time pushing net censorship.
TCP/IP has been standard nomenclature since long before Windows had it. UDP is a thin veneer on IP, but TCP is a very significant component of the protocol suite.
There is one really really big and important difference between TCP and UDP: TCP does flow control and congestion control, but UDP does not, unless you layer it on in your application, which most people don't.
Without TCP congestion control, the Internet would collapse in a flaming heap tomorrow. This would also happen if everyone decided to switch to UDP for everything.
For whoever doesn't know what congestion control is --- it's a scheme that slows down your transmission rate when you detect that your data's being lost in the network, presumably because of congestion. The idea is to avoid overloading the network so everyone can get a useful share. It works very well, having been studied and tuned for decades. Anyone who tries to reimplement it in their own UDP-based protocol is very unlikely to get it right first time.
When you're considering where to live, don't forget to think about the possibility of a terrorist nuke. That's why downtown DC and NYC are right out for me.
The Republicans are not consistently anti-big-government. They LOVE to give the Pentagon more money than the generals themselves know what to do with.
Anyway, restricting government power is not a cure-all for corporatism. If not for the government, we'd all be resigned to a long night of Microsoftian hegemony. Or looking further back, we'd all be subject to the result of the merger between Standard Oil, Ma Bell and Ford.
I know that most Libero-Objectivists think that corporate monopolies can do no wrong, but I've never bought that argument.
Complete disclosure would be nice. However, I cannot understand why it would be a good idea to throw open the floodgates and let the bribes flow in every direction. Making sure that third parties can get in on the take isn't going to help much.
One of my friends once suggested that the sale of legislative influence should be vigorously prosecuted as treason... which I believe is still a capital crime in this country.
To a large extent, DARPA is (and has been for a long time) little more than a really big government subsidy providing quality R&D to the US computer industry. The Internet is not, and never has been, about defending the USA from foreign enemies. (Unless you count the fact that Echelon supposedly uses TCP/IP.)
Some argue that maintaining the preeminence of US high-tech industries is a requirement for defending the USA from foreign enemies. They might be right. However, if that's the way it is, let's be honest about it... and I don't see this argument coming from Libertarians, anyway.
I say all this as someone who has been living for years off DARPA money... I assure you that my research is no more useful for defence than for any other application of computers.
> the Internet as it exists today is almost > entirely the product of freedom and capitalism
Freedom, yes. Capitalism, no. Capitalism has brought us NSI and spam email.
Disclaimer: I'm not a US citizen so I can't vote myself, but I do live in the US, and what happens in the US has a big effect on the rest of the world, so I do care about this.
To me, the big issue is simply campaign finance reform. Behind most of the evil legislation that's passed these days, you can see the shadowy hand of corporate lobbying and influence. It's a sickness that is destroying democracy in this country (and elsewhere).
Basing your vote on other issues is futile: it doesn't matter who's the President or what they try to do, as long as corporations have a lock on the legislative process.
So therefore I favour John McCain. Sure, he may have some odious positions on Net censorship and other things, but the fact is he has a track record of trying to deal with the corruption at the core of the political system. It's pointless to hack away at the branches of the tree of evil and ignore the root (apologies to someone-or-another for the quote).
As the campaign unfolds many people may make all kinds of promises for reform, but we all know those promises are worthless. Vote on the track record.
This is another bit of disinformation that the media has bought into, and now Slashdotters as well.
Go back and reread Bruce Schneier's Crypto-Gram discussing this issue. Trying to control the end-user's use of content *with software* is fundamentally impossible. No matter how smart the DVD people were, they were doomed when the first player was released that did decryption in software.
It's just like software copy protection in the 80s: a patient, skilled hacker/cracker will be able to break whatever you come up with. Then they can package the crack in a program that anyone can download and run. Empirical evidence shows that this process is usually much easier than you'd think. Xing may have screwed up by not putting in as much obfuscation as other people, but they can't really be blamed.
Anyone (e.g. Intertrust) who tells you that they have a secure solution, that does not involve tamper-resistant hardware, is simply lying.
I don't know about copyrights, but regarding patents your statement is completely false. A patent in the US is not valid in the UK. A UK patent must be obtained separately. For example, RSA is patented in the US but not elsewhere.
Slashdot Mirror
It was (and remains) an explicit goal of XPCOM to be compatible with COM on Windows. Of course, there are no open source COM implementations on other platforms, so Mozilla.org had no choice but either to roll their own COM clone, or else go with some CORBA solution. Given that XPCOM is used for very frequently accessed components (e.g. file access APIs, graphics calls), I think it was the right decision to avoid CORBA.
> As I understand it, the biggest difference
> between the MPL and the GPL is that with the MPL
> you have to assign the changes back to Netscape.
The *NPL* gives Netscape special rights to redistribute your changes in closed source products (not the same thing as assigning changes back to Netscape). Substantial new code (i.e. new files) being added to Mozilla by third parties is generally being added under the *MPL*, which is similar to the NPL but gives Netscape no special rights.
> As I further understand it, this is because
> Netscape wants to make money from your work (if
> it was to "benefit the community" they would let
> you distribute your own copy of Moz)
They do let you distribute your own copy of Moz (both NPL and MPL licensed code), so your comment is basically wrong.
It's true that Netscape wants to be able to release closed-source products based on their own source code, even after it has been modified by others, thus the NPL. This does not seem to be a big deal.
Even RMS agrees that the NPL and MPL are free software licenses.
Almost all the source code has been, or soon will be, released. Only the parts specific to RSA await the expiry of the patent. Until then, you can substitute your own RSA implementation (taken from, say, OpenSSL) and build your own binary from these sources. OK, it would be illegal if you're in the US, but you can do it.
It's strong and available to everyone.
If this is a rountine evolutionary step, then where are all these life forms? Hiding themselves because they're shy? Interstellar space isn't that big if you have a few million years to make the trip.
Same problem exists for everyone who thinks that advanced life forms are "easy".
You're making the common mistake of assuming that a "neural network" as stuided in computer science bears more than a passing resemblance to an actual system of biological neurons.
"Bad news, Mr Jones. I, script kiddie, just exploited a buffer overrun to get root on your brain emulator. As of now, I AM YOUR GOD. I control every aspect of your reality. Worship me."
This works for any VR setup that you can't remove or take off, as well. Count me out.
The only ones that sound doable in less than 30 years are the ones that would vaporize your brain instantly in a burst of high-energy radiation. Good luck.
Penrose's arguments about computability are fallacious. The problem is that we don't actually know that human mathematical insight is noncomputable. We may well discover systems with Godel-style statements that are true, but whose truth we cannot decide.
However, there is still a (perhaps small) possibility that he is right in claiming our mental processes rely partly on quantum processes. If so, then our mental state probably could not be duplicated accurately, because it would be corrupted in the process of reading it.
OK, the anthropic problems can be gotten around using exotic probabilistic techniques, but that tech would make Kurzweil's brain scanning look completely trivial.
You are still wrong. Fortunately this will become clear when the strong-crypto binaries and source code appear on Mozilla.org.
Mozilla is still full of memory leaks. They're being plugged but it takes time.
I haven't heard of any SSL patches for Mozilla to date. Do you have any references? I don't think so. The Cryptozilla hack done by Eric Young and others early in '98 was more a proof of concept than anything else, and their code is completely useless in the rewritten codebase.
It's true that someone could have, and could still, produce an SSL plugin for Mozilla based on OpenSSL or something like that. They haven't though. But the big breakthrough is that we can now talk about the issues, standardize the APIs, and leverage a lot of the code Netscape has already written. Also Netscape will be releasing binaries which will give US users access to an RSA-licensed implementation.
You are wrong. This is full-strength crypto. It is not easily crackable; it's as good as it gets.
RSA isn't that much code. It should be almost no effort to produce an "international" patch with the required functions.
Also, note that Netscape will be releasing a precompiled binary containing a licensed implementation of the RSA code. So non-US developers will be able to build their own binaries, and US users can download and use Netscape's.
It will be strong crypto.
You won't find an electable candidate whom you agree with on every issue. That is simply human nature. It is necessary to prioritize the issues and weight candidates' positions accordingly.
... politicians' promises should be discounted.)
/. article he doesn't spend a lot of time pushing net censorship.
I personally think campaign finance reform is of paramount importance; it's hard to make progress on any issue when the political system is fundamentally corrupt. Thus, although I can't vote, if I could, I would vote for McCain --- even though I detest his positions on net censorship and other issues. (McCain has the advantage over Bradley that he doesn't just talk the talk, he's walked the walk for reform
It helps that it seems pretty clear that *his* number #1 issue is also C.F.R., and according to this
TCP/IP has been standard nomenclature since long before Windows had it. UDP is a thin veneer on IP, but TCP is a very significant component of the protocol suite.
There is one really really big and important difference between TCP and UDP: TCP does flow control and congestion control, but UDP does not, unless you layer it on in your application, which most people don't.
Without TCP congestion control, the Internet would collapse in a flaming heap tomorrow. This would also happen if everyone decided to switch to UDP for everything.
For whoever doesn't know what congestion control is --- it's a scheme that slows down your transmission rate when you detect that your data's being lost in the network, presumably because of congestion. The idea is to avoid overloading the network so everyone can get a useful share. It works very well, having been studied and tuned for decades. Anyone who tries to reimplement it in their own UDP-based protocol is very unlikely to get it right first time.
When you're considering where to live, don't forget to think about the possibility of a terrorist nuke. That's why downtown DC and NYC are right out for me.
The Republicans are not consistently anti-big-government. They LOVE to give the Pentagon more money than the generals themselves know what to do with.
Anyway, restricting government power is not a cure-all for corporatism. If not for the government, we'd all be resigned to a long night of Microsoftian hegemony. Or looking further back, we'd all be subject to the result of the merger between Standard Oil, Ma Bell and Ford.
I know that most Libero-Objectivists think that corporate monopolies can do no wrong, but I've never bought that argument.
Complete disclosure would be nice. However, I cannot understand why it would be a good idea to throw open the floodgates and let the bribes flow in every direction. Making sure that third parties can get in on the take isn't going to help much.
... which I believe is still a capital crime in this country.
One of my friends once suggested that the sale of legislative influence should be vigorously prosecuted as treason
To a large extent, DARPA is (and has been for a long time) little more than a really big government subsidy providing quality R&D to the US computer industry. The Internet is not, and never has been, about defending the USA from foreign enemies. (Unless you count the fact that Echelon supposedly uses TCP/IP.)
... and I don't see this argument coming from Libertarians, anyway.
... I assure you that my research is no more useful for defence than for any other application of computers.
Some argue that maintaining the preeminence of US high-tech industries is a requirement for defending the USA from foreign enemies. They might be right. However, if that's the way it is, let's be honest about it
I say all this as someone who has been living for years off DARPA money
> the Internet as it exists today is almost
> entirely the product of freedom and capitalism
Freedom, yes. Capitalism, no. Capitalism has brought us NSI and spam email.
That's interesting, and I didn't know about it. It doesn't change my opinion though. It looks to me like he learned his lesson from that incident.
Disclaimer: I'm not a US citizen so I can't vote myself, but I do live in the US, and what happens in the US has a big effect on the rest of the world, so I do care about this.
To me, the big issue is simply campaign finance reform. Behind most of the evil legislation that's passed these days, you can see the shadowy hand of corporate lobbying and influence. It's a sickness that is destroying democracy in this country (and elsewhere).
Basing your vote on other issues is futile: it doesn't matter who's the President or what they try to do, as long as corporations have a lock on the legislative process.
So therefore I favour John McCain. Sure, he may have some odious positions on Net censorship and other things, but the fact is he has a track record of trying to deal with the corruption at the core of the political system. It's pointless to hack away at the branches of the tree of evil and ignore the root (apologies to someone-or-another for the quote).
As the campaign unfolds many people may make all kinds of promises for reform, but we all know those promises are worthless. Vote on the track record.
> Now if the DVD forum had done their job
...
This is another bit of disinformation that the media has bought into, and now Slashdotters as well.
Go back and reread Bruce Schneier's Crypto-Gram discussing this issue. Trying to control the end-user's use of content *with software* is fundamentally impossible. No matter how smart the DVD people were, they were doomed when the first player was released that did decryption in software.
It's just like software copy protection in the 80s: a patient, skilled hacker/cracker will be able to break whatever you come up with. Then they can package the crack in a program that anyone can download and run. Empirical evidence shows that this process is usually much easier than you'd think. Xing may have screwed up by not putting in as much obfuscation as other people, but they can't really be blamed.
Anyone (e.g. Intertrust) who tells you that they have a secure solution, that does not involve tamper-resistant hardware, is simply lying.
I don't know about copyrights, but regarding patents your statement is completely false. A patent in the US is not valid in the UK. A UK patent must be obtained separately. For example, RSA is patented in the US but not elsewhere.