It already appears like Mr. Katz has been corrected, but I just wanted to pipe in as well so that my opinion can be republished in his next book.
To address a couple of points.
Metallica is not sicking lawyers at 335,000 people. They are instead asking Napster to ban these 335,000 people from using the Napster service. This is what Napster has agreed to do. This was their "claim" to legitamize their service, i.e. "We don't endorse pirating, if you see piracy going on let us know."
As far as Napster providing more to the world than Metallica. How can that possibly be true? Napster provides nothing of any value, all they provide is a mechanism for exchanging music. The only content of any value is the pirated works of groups like Metallica which are exchange over Napster.
So in essense without Metallica, Napster would be worthless.
Metallica, Dr. Dre and other bands have a right to protect their work, after all it's their work and not yours.
I think the basic point that Metallica is making is that Napster is in fact a service whose sole purpose is to encourage the distribution of pirated content.
Supposedly Napster is this legitimate service and there are only a few rogue individuals using it for pirating. Or so claims Napster, and so claimed Mr. Katz until today when he began to defend the pirates, who as it turn out are the majority users of Napster.
One of the things which puzzles me is what kind of business plan does Napster have? It's a company, they are providing a service. But how do they anticipate making money off this service? Or even how do they expect to pay for their support costs to provide the service?
I have not used Napster, but I suspect at some point here, they're going to be pushing advertising at you, or something like that. That's where the RIAA get's pissed off, because here's a company whose business plan is to make money off the redistribution of copyrighted works without reimbursing those artists for their work.
MP3 has it's uses in the recording industry. It's an awesome format for what it was intended for... providing samples of music. I love being able to listen to a 30 second segment of a track to realize that it is the kind of music I like, or it's the song I thought of. I've used this a lot on cdnow.com before buying CD's.
I see an unfortunate trend amongst Open Source advocates such as Mr. Katz that they feel it is their right to take advantage of the work of other people.
Witness this latest book by Mr. Katz. What does it contain? Well it contains all of the responses that we have posted to slashdot.org. I don't know if my respones are in there because I'm not stupid enough to buy such a book, but if they were... did I give permission to be republished? Am I getting royalties off the sale of the book?
When Windows 95 and NT 4.0 were both late for release, the media turned to hyping up OS/2.
It's something the media does... they have to report on things, but when the current product is stagnant and you're sitting around waiting for this promised new version... you get bored.
So Windows 2000 has been late by about 18 months or so, the media got bored and started looking at Linux.
Next year Windows 2000 will dominate all the media, and in another year or two they'll start looking for some other product to look at. (Well now that linux hasn't lived up to it's hype, remember) Maybe it'll be BeOS, or MacOS X, or whatever.
I've said this before. But quite frankly the scripting functionality of modern windows, especially Windows 2000 is actually better than that offered by Linux.
Microsoft realized this was a weakness, and that's why they release Windows Scripting Host.
And part of what makes Microsoft's scripting so much more powerful is that they've presented within their scripting engines easy configuration of the entire system.
This appears to primarily be about violations of the Digital Media Copyright Act, which makes it illegal to distribute, sell, or create software which breaks copy protection.
The arguments I see here don't seem to take that into consideration.
I think a large aspect of this, which does actually go back to Mr. Raymond's comment on trading performance for security, is that Netrek is designed in such a way that all the game logic is stored on the server. The client only displays what the server offers it, no additional information is available for it's use.
Essentially, even though the Netrek clients have been blessed and trusted in some manner, in reality they aren't trusted at all. The server does not provide sufficient ability to the client to even allow it to cheat.
What we regard as cheating in the Netrek community is using computer based crutches to supplement your lack of skill. That could be auto-aim, or displaying cloaked ships, etc.
This game architecture obviously came about due to the open source nature of the game.
It's rare that I actually agree with Eric Raymond on anything, so I think this must be an amazing realignment of the planets and stars.:)
But as someone else pointed out in another article, the computerized players are not much better than the best human players, and in many aspects of the game are worse.
And this my friends is an example of why Netrek is a superior game to Quake.:)
Seriously, Netrek has depth to the game, it's not just about who has the fastest trigger finger. In some cases not firing at an opponent is a smart move, and a large part of learning the game is understanding such strategies.
To my knowledge, the only time any RSA keys used in Netrek were actually cracked was when I stumbled across a bug in the key generation program back in 1994 or so.
The key generation program was not using a sufficiently random seed, and as a result it really only generated any one of perhaps a couple of thousand key combinations.
It was not difficult for me to setup my computer at the time to create several thousand keys, and simply compare them against the known public keys to see if I had a match.
In one night I had cracked some 40 client keys.
But I made an attempt to report it to all the client maintainers at the time, and after the key generation program was fixed we generated new keys and recompiled the clients.
If you can solve the man in the middle attack I think it is somewhat workable. There is likely some method of doing this using information specific to the client computer as part of the encrypted key. Such as the IP address, etc.
The difficulty there lies in trying to play from behind a firewall.
Another comment... just to clarify, the exact source is not released because it is actually generated during the build based on the input parameter of the RSA private key. Really the full source is easily available, and anybody can compile a client, it's just that you need a RSA key that has been accepted by the servers to be useful.
The server encrypts something using the client's public key, sends it to the client who then decrypts it and responds accordingly.
Only something holding the private key can decrypt the message.
The old mechanism used to use a rather simply encryption/decryption program based off of some simple math. It worked, as long as the source was trusted. Then someone posted the source to the newsgroup and it was all over.:)
It actually works quite well. It's difficult to pull the key out of the binary, because it's been obfuscated into the binary.
As Carlos pointed out in another message if the key is compromised, it's easy to yank the key, create a new one, and recompile the binaries.
And since the code is obfuscated per compilation, to find the key out would require the same amount of work each and every time... i.e. disassembly of the client. Sooner or later someone will give up.
Netrek isn't very popular these days, because of eye-candy games like quake. But at it's height back in the early 90's there were quite a few people who looked at the authentication mechanism and tried to break it... thus strengthening it.
Well go figure, Quake comes out and tries to do the same thing as Netrek started doing over 10 years ago, and they run into the same problems.
I don't recall if it was Andy McFadden, or Sam Shen who first implemented it, I think they were both involved to some degree. But Netrek uses RSA public key encryption to do a client verification.
Public keys of authorized clients are stored on the server. A client connects, uses it's private key to prove it's identity and away you go.
The only thing closed is the private key. This is held by the client maintainer, and is also embedded into the compiled binary in a obfuscated manner.
If a private key comes into question, it can be yanked off the server immediately, requiring people to obtain new clients for that platform, whatever.
I think the other problem is that Quake allows too much processing to be done at the client side. In Netrek everything is server based, so about the only thing a borg client can do is auto aim for you, etc. No different than a robot playing your position. You can't defy the netrek rules at any rate, by increasing power or controlling whether you got hit or not.
His comment... "Windows was built for a single computer and then the network was added on as an afterthought."
is kind of interesting.
Unix was built for a single computer and then the network was added on as an afterthought as well.
So I assume he's speaking of just Linux, right?
But then which Windows is he speaking of? Sure Windows 3.1 didn't have networking built in to begin with.
But Windows NT was designed with the network in mind.
I find it interesting how blindly Linux advocates jump onto bandwagons. The quotes from this individual show about as much thought as your typical marketing droid.
I checked broadband.com and an awful lot of the content I came across seemed to be in MS format. Especially video's.
A lot of places have been dumping Real in favor of MS Windows Media because of disagreements with Real.
ABC last month or so had a story out where they selected MS over Real because of Real's refusal to place their logo less prominently on a site carrying a web video feed of the Drew Carey show.
This is really Microsoft's secret to success. They wait until the competition starts shooting themselves in the foot, and then they move in with very favorable terms.
The DVD-Audio standard is widely desired by a great number of Audiophiles. Considering that there have been a number of relatively successful products sold over the years just to try to improve the CD format...I don't think it'll be a failure.
You have to understand that there are actually people in this world who appreciate music quality. Hint: They don't buy MP3 players.
I'm not talking about the idiots who go off and send emails.
I'm talking about publishing a questionable news story without verification... especially since the posting was in the form of flamebait and included a link to the ISP's contact information.
I realize that the editors at slashdot think thought they were providing a community service by informing the public. But this is just one of numerous examples of poor research being done prior to publishing the article.
Slashdot, like most other news media outlets these days appears to throw ethics to the wind in the rush to get the story out first.
The only thing Microsoft ever did to damage the Amiga was writing the AmigaBASIC interpreter. It was essentially a very early version of what would later become QuickBASIC.
That is, it no longer required line numbers, and supported functions and subroutines and such.
However it was slow as hell.
Actually one of the things which hurt the Amiga was the lack of good development tools and information.
I think it's funny that slashdot always uses 'Microsoft' to get people interested in an article.
I remember when AmigaDOS v2.04 came out, and how it was trumpeted that the Guru Meditation error had finally been eliminated.
Guru meditation errors were obnoxious because it locked up your whole machine with this pile of useless numbers staring you in the face. Much like the BSOD on NT machines, or kernel panics on Unix machines except Guru's happened much much more often.
Anyway so it was funny when the marketing claimed that AmigaDOS v2.04 was much more reliable because they'd eliminated the Guru meditation problem.
Well, they hadn't eliminated it, just renamed it. Not quite the same thing.:)
Actually my Amiga 500 started guruing far more often with v2.04, it would even guru when I turned off my external 2400 baud modem. That's how unstable the platform was...:(
#1. No, someone cannot relicense your code without your permission since you hold the copyright. If your claim were true I could release GPL'ed code under a BSD license.
#2. No, my example allows for commercial use, but requires the commercial entity to ask for permission. And possibly give a financial reward in exchange for allowing it to use your code.
I see no reason for the GPL's existence other than Stallman's crazy idea that programmers should all work for the government(like he does).
Slashdot Mirror
He said average.
Very few albums ever sell 1 million copies. Perhaps only a couple of dozen artists per year.
It already appears like Mr. Katz has been corrected, but I just wanted to pipe in as well so that my opinion can be republished in his next book.
To address a couple of points.
Metallica is not sicking lawyers at 335,000 people. They are instead asking Napster to ban these 335,000 people from using the Napster service. This is what Napster has agreed to do. This was their "claim" to legitamize their service, i.e. "We don't endorse pirating, if you see piracy going on let us know."
As far as Napster providing more to the world than Metallica. How can that possibly be true? Napster provides nothing of any value, all they provide is a mechanism for exchanging music. The only content of any value is the pirated works of groups like Metallica which are exchange over Napster.
So in essense without Metallica, Napster would be worthless.
Metallica, Dr. Dre and other bands have a right to protect their work, after all it's their work and not yours.
I think the basic point that Metallica is making is that Napster is in fact a service whose sole purpose is to encourage the distribution of pirated content.
Supposedly Napster is this legitimate service and there are only a few rogue individuals using it for pirating. Or so claims Napster, and so claimed Mr. Katz until today when he began to defend the pirates, who as it turn out are the majority users of Napster.
One of the things which puzzles me is what kind of business plan does Napster have? It's a company, they are providing a service. But how do they anticipate making money off this service? Or even how do they expect to pay for their support costs to provide the service?
I have not used Napster, but I suspect at some point here, they're going to be pushing advertising at you, or something like that. That's where the RIAA get's pissed off, because here's a company whose business plan is to make money off the redistribution of copyrighted works without reimbursing those artists for their work.
MP3 has it's uses in the recording industry. It's an awesome format for what it was intended for... providing samples of music. I love being able to listen to a 30 second segment of a track to realize that it is the kind of music I like, or it's the song I thought of. I've used this a lot on cdnow.com before buying CD's.
I see an unfortunate trend amongst Open Source advocates such as Mr. Katz that they feel it is their right to take advantage of the work of other people.
Witness this latest book by Mr. Katz. What does it contain? Well it contains all of the responses that we have posted to slashdot.org. I don't know if my respones are in there because I'm not stupid enough to buy such a book, but if they were... did I give permission to be republished? Am I getting royalties off the sale of the book?
You do realize that this message is copyrighted?
When Windows 95 and NT 4.0 were both late for release, the media turned to hyping up OS/2.
It's something the media does... they have to report on things, but when the current product is stagnant and you're sitting around waiting for this promised new version... you get bored.
So Windows 2000 has been late by about 18 months or so, the media got bored and started looking at Linux.
Next year Windows 2000 will dominate all the media, and in another year or two they'll start looking for some other product to look at. (Well now that linux hasn't lived up to it's hype, remember) Maybe it'll be BeOS, or MacOS X, or whatever.
I've said this before. But quite frankly the scripting functionality of modern windows, especially Windows 2000 is actually better than that offered by Linux.
Microsoft realized this was a weakness, and that's why they release Windows Scripting Host.
And part of what makes Microsoft's scripting so much more powerful is that they've presented within their scripting engines easy configuration of the entire system.
This appears to primarily be about violations of the Digital Media Copyright Act, which makes it illegal to distribute, sell, or create software which breaks copy protection.
The arguments I see here don't seem to take that into consideration.
I think a large aspect of this, which does actually go back to Mr. Raymond's comment on trading performance for security, is that Netrek is designed in such a way that all the game logic is stored on the server. The client only displays what the server offers it, no additional information is available for it's use.
:)
:)
Essentially, even though the Netrek clients have been blessed and trusted in some manner, in reality they aren't trusted at all. The server does not provide sufficient ability to the client to even allow it to cheat.
What we regard as cheating in the Netrek community is using computer based crutches to supplement your lack of skill. That could be auto-aim, or displaying cloaked ships, etc.
This game architecture obviously came about due to the open source nature of the game.
It's rare that I actually agree with Eric Raymond on anything, so I think this must be an amazing realignment of the planets and stars.
But as someone else pointed out in another article, the computerized players are not much better than the best human players, and in many aspects of the game are worse.
And this my friends is an example of why Netrek is a superior game to Quake.
Seriously, Netrek has depth to the game, it's not just about who has the fastest trigger finger. In some cases not firing at an opponent is a smart move, and a large part of learning the game is understanding such strategies.
We call it 'clue'.
To my knowledge, the only time any RSA keys used in Netrek were actually cracked was when I stumbled across a bug in the key generation program back in 1994 or so.
The key generation program was not using a sufficiently random seed, and as a result it really only generated any one of perhaps a couple of thousand key combinations.
It was not difficult for me to setup my computer at the time to create several thousand keys, and simply compare them against the known public keys to see if I had a match.
In one night I had cracked some 40 client keys.
But I made an attempt to report it to all the client maintainers at the time, and after the key generation program was fixed we generated new keys and recompiled the clients.
It is a difficult question to ponder.
:)
If you can solve the man in the middle attack I think it is somewhat workable. There is likely some method of doing this using information specific to the client computer as part of the encrypted key. Such as the IP address, etc.
The difficulty there lies in trying to play from behind a firewall.
Another comment... just to clarify, the exact source is not released because it is actually generated during the build based on the input parameter of the RSA private key. Really the full source is easily available, and anybody can compile a client, it's just that you need a RSA key that has been accepted by the servers to be useful.
Oh, and it's correctly spelled Netrek!
The client/server protocol in Netrek already pretty much prevents cheating.
Everything is calculated at the server side, and only information which the enduser should know about is sent to the client.
This opens up ways to automate tasks for the user, but there's absolutely no way you can become God incarnate within the game with a client.
The secret is RSA public key encryption.
:)
The server encrypts something using the client's public key, sends it to the client who then decrypts it and responds accordingly.
Only something holding the private key can decrypt the message.
The old mechanism used to use a rather simply encryption/decryption program based off of some simple math. It worked, as long as the source was trusted. Then someone posted the source to the newsgroup and it was all over.
It actually works quite well. It's difficult to pull the key out of the binary, because it's been obfuscated into the binary.
As Carlos pointed out in another message if the key is compromised, it's easy to yank the key, create a new one, and recompile the binaries.
And since the code is obfuscated per compilation, to find the key out would require the same amount of work each and every time... i.e. disassembly of the client. Sooner or later someone will give up.
Netrek isn't very popular these days, because of eye-candy games like quake. But at it's height back in the early 90's there were quite a few people who looked at the authentication mechanism and tried to break it... thus strengthening it.
Well go figure, Quake comes out and tries to do the same thing as Netrek started doing over 10 years ago, and they run into the same problems.
I don't recall if it was Andy McFadden, or Sam Shen who first implemented it, I think they were both involved to some degree. But Netrek uses RSA public key encryption to do a client verification.
Public keys of authorized clients are stored on the server. A client connects, uses it's private key to prove it's identity and away you go.
The only thing closed is the private key. This is held by the client maintainer, and is also embedded into the compiled binary in a obfuscated manner.
If a private key comes into question, it can be yanked off the server immediately, requiring people to obtain new clients for that platform, whatever.
I think the other problem is that Quake allows too much processing to be done at the client side. In Netrek everything is server based, so about the only thing a borg client can do is auto aim for you, etc. No different than a robot playing your position. You can't defy the netrek rules at any rate, by increasing power or controlling whether you got hit or not.
His comment... "Windows was built for a single computer and then the network was added on as an afterthought."
is kind of interesting.
Unix was built for a single computer and then the network was added on as an afterthought as well.
So I assume he's speaking of just Linux, right?
But then which Windows is he speaking of? Sure Windows 3.1 didn't have networking built in to begin with.
But Windows NT was designed with the network in mind.
I find it interesting how blindly Linux advocates jump onto bandwagons. The quotes from this individual show about as much thought as your typical marketing droid.
HPFS doesn't support security mechanisms, i.e. no file ownership, no ACLs, etc.
I also question that Warp 4 was technically superior to NT.
I find it amazing the number of posters to slashdot who go around distorting reality in order to try to make a point.
I checked broadband.com and an awful lot of the content I came across seemed to be in MS format. Especially video's.
A lot of places have been dumping Real in favor of MS Windows Media because of disagreements with Real.
ABC last month or so had a story out where they selected MS over Real because of Real's refusal to place their logo less prominently on a site carrying a web video feed of the Drew Carey show.
This is really Microsoft's secret to success. They wait until the competition starts shooting themselves in the foot, and then they move in with very favorable terms.
The DVD-Audio standard is widely desired by a great number of Audiophiles. Considering that there have been a number of relatively successful products sold over the years just to try to improve the CD format...I don't think it'll be a failure.
You have to understand that there are actually people in this world who appreciate music quality. Hint: They don't buy MP3 players.
I'm not talking about the idiots who go off and send emails.
I'm talking about publishing a questionable news story without verification... especially since the posting was in the form of flamebait and included a link to the ISP's contact information.
I realize that the editors at slashdot think thought they were providing a community service by informing the public. But this is just one of numerous examples of poor research being done prior to publishing the article.
Slashdot, like most other news media outlets these days appears to throw ethics to the wind in the rush to get the story out first.
They'll cancel everything except for the Windows port of Linux. :)
The only thing Microsoft ever did to damage the Amiga was writing the AmigaBASIC interpreter. It was essentially a very early version of what would later become QuickBASIC.
That is, it no longer required line numbers, and supported functions and subroutines and such.
However it was slow as hell.
Actually one of the things which hurt the Amiga was the lack of good development tools and information.
I think it's funny that slashdot always uses 'Microsoft' to get people interested in an article.
I've never heard of this guy either.
Dave Haynie was one of the main engineers with the Amiga.
And some of the guys who supported the Amiga through freeware/shareware...
Fred Fish... long before there was the internet, there was the Fred Fish collection of software.
And Matt Dillon contributed a lot. Dmouse was one of my favorite tools, and he also had a C compiler, etc.
I remember when AmigaDOS v2.04 came out, and how it was trumpeted that the Guru Meditation error had finally been eliminated.
:)
:(
Guru meditation errors were obnoxious because it locked up your whole machine with this pile of useless numbers staring you in the face. Much like the BSOD on NT machines, or kernel panics on Unix machines except Guru's happened much much more often.
Anyway so it was funny when the marketing claimed that AmigaDOS v2.04 was much more reliable because they'd eliminated the Guru meditation problem.
Well, they hadn't eliminated it, just renamed it. Not quite the same thing.
Actually my Amiga 500 started guruing far more often with v2.04, it would even guru when I turned off my external 2400 baud modem. That's how unstable the platform was...
Jay Miner died back in 1994.
go to www.jms.org to find out more
#1. No, someone cannot relicense your code without your permission since you hold the copyright. If your claim were true I could release GPL'ed code under a BSD license.
#2. No, my example allows for commercial use, but requires the commercial entity to ask for permission. And possibly give a financial reward in exchange for allowing it to use your code.
I see no reason for the GPL's existence other than Stallman's crazy idea that programmers should all work for the government(like he does).
I don't think .25% of the market is all that signifigant.
And there is very little money currently being made. Few of the Linux related companies are profitable, if any.