Because you'll spend a week trying to work around problems, and then give up, because it still doesn't work?
The server for my card didn't work until I patched it - for some reason, locking the console locked up the machine, and the API never returned. After working around this, xinit wouldn't work unless I was root. You need a beta (?) version of Watcom C to compile XFree86, and I wasn't willing to put that on a machine where I compile production code. I ended up putting off getting it to work for the forseeable future.
Mind you, the X11R5 package we got from QSSL was just as bad, and never worked either.
Ok, I looked at it, and that's not what I meant. They're trying to authenticate the client/server binaries. I'm proposing to authenticate the players. To play, you need to generate your online ID, and a key pair. When you first log in to a server, you give it your ID. The server checks a database for your ID. If the server doesn't know you, it asks you for your public key, and stores it in the database. Your key is marked "untrusted", because the server has no reason to trust you.
Other players see you on the server, and can play against you, but they know not to trust you, because the server marks your ID "untrusted" in the UI.
This is where things get complicated. How do you gain players' trust? There are several possible ways. You can gain trust with time. Play long enough (a week? a month? The period of time should be long enough that people will be discouraged from playing straight for a while, and then going on a cheating spree, until servers and players recognize them for the jerks they are). A trusted player can recommend you to the server. This can be done automatically when you log in (I think this will cause problems for players with low-bandwidth connections, when the server starts sending ID's of newly connected players to everyone who's already playing, basically asking "is it ok to trust this player?"). This can be done between games, through e-mail, or through a menu option on other player's menus.
When connecting to the server the next time, you submit your ID, as before, except that now the server knows you, and challenges you to prove your identity, by sending you a challenge packet, containing a random number. You sign the random number, with your private key, and the server authenticates the signature with your public key, which it stored during your first log in. Now, when players see you logged in, they know that no one can pretend to be you.
The end result is that players can build up reputation, based on other players' recommendations. No one can pose as another, honest player, and cheat, lowering the true player's reputation. New players can be introduced to the game by friends who are already trusted, and the Quake community will become just that - a community.
Another advantage(?) for anyone who'll try to implement such a system is that the complexity of the protocols (not so much the authentication protocol, which is quite standard, but the automatic maintenance of the web-of-trust, possibly with servers updating each other periodically, causing reputation to propegate through the Net much like Usenet news, players' votes to lower or raise reputation values being adjusted based on the voters' reputations, etc. etc. etc.) guarantees some interesting design and coding experiences.
I don't think you can draw those parallels between constitutional rights and copyrights. In the former case, You are given a right by the Constitution, and are asked to Give up your right by a Third Party. In the latter case, a Third Party is given a right to distribute copyrighted material, under certain conditions (spelled out in the GPL). Said Third Party is not allowed to distribute this copyrighted material under any other conditions. By asking You to give up a right, said Third Party is Talking Through His Ass, because he doesn't have the legal right to distribute the copyright holder's materials under these conditions.
Interesting point... The reason why I don't play RPG's is that none of them have a character that's a 30 year old geek computer programmer with a +5 Pentium III.
Actually, I prefer the current situation. Since the system is known to be flawed, I'm only liable for $50 if someone steals my CC number, and in many cases, not even that. If a secret, proprietary and "secure" system will be put in place, shortly afterwards customers will be liable for all transactions carried out in their name. Once someone figures out how to hack into this newfangled system, we're all in trouble. Good luck explaining to the courts exactly why a complex security system is insecure, and how someone could have presented themselves as you online, and bought those 3 Ferrari's. Either that, or I'm just paranoid.
How many peple actually watch the commercials? Would it be safe to say that the overwhelming majority fast-forwards through the commercials?
I heard quotes of studies that show 90% of the viewers sit through the commercials. Informal studies I've done (sitting with friends, watching TV) suggest the same.
I won't ask you to disclose anything you shouldn't, but does your employer actually turn a profit? Does Amazon, for that matter?
Obviously, if enough people walk away, it makes a difference. I just think that it's important that businesses know why people walk away. Otherwise, walking away will make even less of a difference.
All I wrote them was "I won't be buying from you in the future, and here's why:". No caps, and no threats.
This isn't about money, however. I've been buying from Amazon because it was convenient, and I didn't have to spend time looking. My time is valuable, too.
I'm letting them know, politely, that their practices will cost them more than just attorney fees. To make the point clear, I'm attaching the total sum I spent in their on-line store during this year.
While Amazon's service is generally good, and shopping there is easy and convenient, I can't in good faith endorse anyone who abuses the patent system the way they have. I'll be more than happy to shop there again, if they make it clear that they will make it their policy to not patent obvious software "inventions".
If I'm not mistaken, the GUID referenced is a GUID assigned to a Windows installation, and embedded in Office documents, which, in turn, can be used, and has been used, to track down the documents' authors.
The machine GUID is one of the reasons MS won't endorse the practice of cloning Windows NT disks to do mass installs.
I should mention that my knowledge of this issue is very superficial, and relies mostly on/. postings.
It is possible, with a component based API. Define the interface to the layout engine, using something like COM (or even a C++ class, although there are disatvantages to using C++) and write the UI to use whichever layout engine is available. Then, as long as layout engines support the API, you can mix and match user interfaces and layout engines.
This kind of design is what makes it so easy to drop IE5 into, say, a Visual Basic application, or any other Windows app that supprts OLE controls (or ActiveX controls, or whatever MS calls them this afternoon).
I've seen a coworker put together a voice activated web browser, in VB, with basic functionality, in about 10 minutes of work. Windows is built around components nowadays in pretty much the same way that Unix is built around text and pipes. As long as you have the right components, everything is scriptable to a surprising degree.
Perhaps you can tell me why Micro Channel is a historical footnote. It certainly isn't because of Linux.
I don't know. I don't know what IBM's licensing terms were for MCA. I don't think this is relevant to what's happening now.
I agree that the industry benefits from open standards, but a standard that's open enough for the industry may not be open enough for Free Software. The MPEG standard is open, but encumbered. That's not really a problem if you're a company that makes MPEG analyzers for the cable TV industry, or if you're selling a digital TV settop box. It is a problem if you want to write a free MPEG encoder, and release the source under the GPL.
Granted, a large userbase is required to insure that hardware manufactures provide Linux support from the moment they release new hardware, but there is nothing that will keep us from finding compatible hardware and running Linux on it.
If compatible hardware is either unavailable, or outdated, is that really a choise? How much functionality are you willing to sacrifice just to run a Free operating system? A lot of people don't care whether they get DVD support or not. That's reasonable. Sound support may not be an issue for most Linux users. Where do you get to the point where Linux is only useful as a platform on which you can develop the Linux kernel, and as an embedded system, but that's it?
Linux has excellent hardware support today because it has a large enough user base to make a difference in hardware manufacturers' bottom line. This is more evident where Linux has a really large user base - can you show me a SCSI controller without Linux support? Do you think this is a coincidence, or could it have something to do with Linux's substantial presence in the server market?
The war, if you choose to think of it that way, is about making sure that even a clueless Linux newbie can easily installed brand new, high end, hardware and expect it to work.
To put it differently, the war is on open standards, for hardware and communications protocols. That includes HTML, and browser plugins. Linux has to support most web content for your clueless newbie to have a reason to install Linux in the first place.
I'm not advocating Linux dominance on the desktop - just a sufficient presence to make a difference in the marketplace.
... That seem to think that one aspect of Linux can cause us to lose the "war" with Windows. To most of us linux users, its not a war to begin with. I could care less if others use Linux, I know that I can use it, it takes care of my needs, and I never have to reboot my machine. If there is a Desktop war going on, I wonder who's fighting? MS certainly sees us as a threat, but we couldn't care less about them.
You should care whether others are using Linux. Without a large user base, there will be nothing to stop the industry from making all hardware proprietary. Today, if you use Linux, you know that you can't use Winmodems, some high end sound cards, some really expansive, professional 3D cards, but that's pretty much it. None of this is something you can't live without. However, if Linux won't have a large user base, there will be no market forces that can force hardware manufacturers to open their hardware specs. If, five years from now, the average PC will have 90% proprietary hardware, with drivers protected by patents, you won't have a machine to run your free operating system on.
Do you seriously think that Crative Labs or NVidia care about Free Software, or Open Source? As public companies, they care about their bottom lines, and nothing else. If Linux has a large user base, they open up their hardware specs, and you have sound support in the kernel. If nobody uses Linux (or any other free operating system, for that matter), they have no reason to open their drivers, or to publish hardware specs. On the contrary - to be competative, they will obtain patents for every last detail of their hardware and software, effectively preventing any Free code from supporting their hardware.
This is an important issue, and one that many Linux users fail to understand. A substantial user base is essential, not to us "winning" anything, but to Linux being a viable platform for commodity hardware. Without easy access to affordable Linux workstations, there won't be a Linux.
As for the "war" with Microsoft, you may not think you're that Linux is at war with Microsoft. That doesn't change the fact that Linux poses a threat to Windows's dominant position on the desktop, and therefore to Microsoft's profits. Call it whatever you want, I'll bet they're calling it War.
The only bulletproof way of running X would be to put an X terminal, complete with separate memory and CPU, in the box, and connect your keyboard and monitor to that. This way, if X crashes, you can just reboot it, and your main system (a high availability server, perhaps?) doesn't even know something happened. Take out the graphics code from the kernel, if you want stability, and take it out of the box altogether, if you want REAL stability.
There was an article in the New Yorker about it. Supposedly, Russia was developing a strain of smallpox, that was immune to the American vaccine, as a biological weapon.
Smallpox. Most of the population isn't vaccinated, and with the high rate of mobility in today's society, you can have a worldwide epidemic on your hands before the victims show the symptoms of anything worse than a flu.
Slashdot Mirror
Because you'll spend a week trying to work around problems, and then give up, because it still doesn't work?
The server for my card didn't work until I patched it - for some reason, locking the console locked up the machine, and the API never returned. After working around this, xinit wouldn't work unless I was root. You need a beta (?) version of Watcom C to compile XFree86, and I wasn't willing to put that on a machine where I compile production code. I ended up putting off getting it to work for the forseeable future.
Mind you, the X11R5 package we got from QSSL was just as bad, and never worked either.
Ok, I looked at it, and that's not what I meant. They're trying to authenticate the client/server binaries. I'm proposing to authenticate the players. To play, you need to generate your online ID, and a key pair. When you first log in to a server, you give it your ID. The server checks a database for your ID. If the server doesn't know you, it asks you for your public key, and stores it in the database. Your key is marked "untrusted", because the server has no reason to trust you.
Other players see you on the server, and can play against you, but they know not to trust you, because the server marks your ID "untrusted" in the UI.
This is where things get complicated. How do you gain players' trust? There are several possible ways. You can gain trust with time. Play long enough (a week? a month? The period of time should be long enough that people will be discouraged from playing straight for a while, and then going on a cheating spree, until servers and players recognize them for the jerks they are).
A trusted player can recommend you to the server. This can be done automatically when you log in (I think this will cause problems for players with low-bandwidth connections, when the server starts sending ID's of newly connected players to everyone who's already playing, basically asking "is it ok to trust this player?"). This can be done between games, through e-mail, or through a menu option on other player's menus.
When connecting to the server the next time, you submit your ID, as before, except that now the server knows you, and challenges you to prove your identity, by sending you a challenge packet, containing a random number. You sign the random number, with your private key, and the server authenticates the signature with your public key, which it stored during your first log in. Now, when players see you logged in, they know that no one can pretend to be you.
The end result is that players can build up reputation, based on other players' recommendations. No one can pose as another, honest player, and cheat, lowering the true player's reputation. New players can be introduced to the game by friends who are already trusted, and the Quake community will become just that - a community.
Another advantage(?) for anyone who'll try to implement such a system is that the complexity of the protocols (not so much the authentication protocol, which is quite standard, but the automatic maintenance of the web-of-trust, possibly with servers updating each other periodically, causing reputation to propegate through the Net much like Usenet news, players' votes to lower or raise reputation values being adjusted based on the voters' reputations, etc. etc. etc.) guarantees some interesting design and coding experiences.
Thanks. I'll check it out.
I don't think you can draw those parallels between constitutional rights and copyrights. In the former case, You are given a right by the Constitution, and are asked to Give up your right by a Third Party. In the latter case, a Third Party is given a right to distribute copyrighted material, under certain conditions (spelled out in the GPL). Said Third Party is not allowed to distribute this copyrighted material under any other conditions. By asking You to give up a right, said Third Party is Talking Through His Ass, because he doesn't have the legal right to distribute the copyright holder's materials under these conditions.
Has anyone considered using a PGP like web of trust model, with client authentication, to help prevent cheating?
Interesting point... The reason why I don't play RPG's is that none of them have a character that's a 30 year old geek computer programmer with a +5 Pentium III.
Actually, I prefer the current situation. Since the system is known to be flawed, I'm only liable for $50 if someone steals my CC number, and in many cases, not even that.
If a secret, proprietary and "secure" system will be put in place, shortly afterwards customers will be liable for all transactions carried out in their name. Once someone figures out how to hack into this newfangled system, we're all in trouble. Good luck explaining to the courts exactly why a complex security system is insecure, and how someone could have presented themselves as you online, and bought those 3 Ferrari's.
Either that, or I'm just paranoid.
Not to mention that you can play in direct sunlight. Backlit LCD's become invisible in sunlight.
...He'd die of lead poisoning.
Having been to the Western Cable Show with equipment to set up, I can tell you that the video feed can be pretty bad there.
How many peple actually watch the commercials? Would it be safe to say that the overwhelming
majority fast-forwards through the commercials?
I heard quotes of studies that show 90% of the viewers sit through the commercials. Informal studies I've done (sitting with friends, watching TV) suggest the same.
Just to make it clear, I never suggested that anyone call Amazon employees "the stupidest idiots on the planets", or insult them in any other way.
There are civilized ways to express your dislike of someone's business practices, and I see no reason why people shouldn't use them.
I won't ask you to disclose anything you shouldn't, but does your employer actually turn a profit? Does Amazon, for that matter?
Obviously, if enough people walk away, it makes a difference. I just think that it's important that businesses know why people walk away. Otherwise, walking away will make even less of a difference.
All I wrote them was "I won't be buying from you in the future, and here's why:". No caps, and no threats.
I sent it to feedback@amazon.com. I don't know whether this address goes to a human or to /dev/null, because I haven't gotten a reply yet.
Thanks for the link.
This isn't about money, however. I've been buying from Amazon because it was convenient, and I didn't have to spend time looking. My time is valuable, too.
I'm letting them know, politely, that their practices will cost them more than just attorney fees. To make the point clear, I'm attaching the total sum I spent in their on-line store during this year.
While Amazon's service is generally good, and shopping there is easy and convenient, I can't in good faith endorse anyone who abuses the patent system the way they have. I'll be more than happy to shop there again, if they make it clear that they will make it their policy to not patent obvious software "inventions".
If I'm not mistaken, the GUID referenced is a GUID assigned to a Windows installation, and embedded in Office documents, which, in turn, can be used, and has been used, to track down the documents' authors.
/. postings.
The machine GUID is one of the reasons MS won't endorse the practice of cloning Windows NT disks to do mass installs.
I should mention that my knowledge of this issue is very superficial, and relies mostly on
It is possible, with a component based API. Define the interface to the layout engine, using something like COM (or even a C++ class, although there are disatvantages to using C++) and write the UI to use whichever layout engine is available. Then, as long as layout engines support the API, you can mix and match user interfaces and layout engines.
This kind of design is what makes it so easy to drop IE5 into, say, a Visual Basic application, or any other Windows app that supprts OLE controls (or ActiveX controls, or whatever MS calls them this afternoon).
I've seen a coworker put together a voice activated web browser, in VB, with basic functionality, in about 10 minutes of work. Windows is built around components nowadays in pretty much the same way that Unix is built around text and pipes. As long as you have the right components, everything is scriptable to a surprising degree.
PAL is interlaced, just like NTSC, giving you 50 fields/sec.
Perhaps you can tell me why Micro Channel is a historical footnote. It certainly isn't because of Linux.
I don't know. I don't know what IBM's licensing terms were for MCA. I don't think this is relevant to what's happening now.
I agree that the industry benefits from open standards, but a standard that's open enough for the industry may not be open enough for Free Software. The MPEG standard is open, but encumbered. That's not really a problem if you're a company that makes MPEG analyzers for the cable TV industry, or if you're selling a digital TV settop box. It is a problem if you want to write a free MPEG encoder, and release the source under the GPL.
Granted, a large userbase is required to insure that hardware manufactures provide Linux support from the moment they release new hardware, but there is nothing that will keep us from finding compatible hardware and running Linux on it.
If compatible hardware is either unavailable, or outdated, is that really a choise? How much functionality are you willing to sacrifice just to run a Free operating system? A lot of people don't care whether they get DVD support or not. That's reasonable. Sound support may not be an issue for most Linux users. Where do you get to the point where Linux is only useful as a platform on which you can develop the Linux kernel, and as an embedded system, but that's it?
Linux has excellent hardware support today because it has a large enough user base to make a difference in hardware manufacturers' bottom line. This is more evident where Linux has a really large user base - can you show me a SCSI controller without Linux support? Do you think this is a coincidence, or could it have something to do with Linux's substantial presence in the server market?
The war, if you choose to think of it that way, is about making sure that even a clueless Linux newbie can easily installed brand new, high end, hardware and expect it to work.
To put it differently, the war is on open standards, for hardware and communications protocols. That includes HTML, and browser plugins. Linux has to support most web content for your clueless newbie to have a reason to install Linux in the first place.
I'm not advocating Linux dominance on the desktop - just a sufficient presence to make a difference in the marketplace.
... That seem to think that one aspect of Linux can cause us to lose the "war" with Windows. To most of us linux users, its not a war to begin with. I could care less if
others use Linux, I know that I can use it, it takes care of my needs, and I never have to reboot my machine. If there is a Desktop war going on, I wonder who's fighting?
MS certainly sees us as a threat, but we couldn't care less about them.
You should care whether others are using Linux. Without a large user base, there will be nothing to stop the industry from making all hardware proprietary. Today, if you use Linux, you know that you can't use Winmodems, some high end sound cards, some really expansive, professional 3D cards, but that's pretty much it. None of this is something you can't live without. However, if Linux won't have a large user base, there will be no market forces that can force hardware manufacturers to open their hardware specs. If, five years from now, the average PC will have 90% proprietary hardware, with drivers protected by patents, you won't have a machine to run your free operating system on.
Do you seriously think that Crative Labs or NVidia care about Free Software, or Open Source? As public companies, they care about their bottom lines, and nothing else. If Linux has a large user base, they open up their hardware specs, and you have sound support in the kernel. If nobody uses Linux (or any other free operating system, for that matter), they have no reason to open their drivers, or to publish hardware specs. On the contrary - to be competative, they will obtain patents for every last detail of their hardware and software, effectively preventing any Free code from supporting their hardware.
This is an important issue, and one that many Linux users fail to understand. A substantial user base is essential, not to us "winning" anything, but to Linux being a viable platform for commodity hardware. Without easy access to affordable Linux workstations, there won't be a Linux.
As for the "war" with Microsoft, you may not think you're that Linux is at war with Microsoft. That doesn't change the fact that Linux poses a threat to Windows's dominant position on the desktop, and therefore to Microsoft's profits. Call it whatever you want, I'll bet they're calling it War.
The only bulletproof way of running X would be to put an X terminal, complete with separate memory and CPU, in the box, and connect your keyboard and monitor to that. This way, if X crashes, you can just reboot it, and your main system (a high availability server, perhaps?) doesn't even know something happened. Take out the graphics code from the kernel, if you want stability, and take it out of the box altogether, if you want REAL stability.
Hoy, hoy, hoy!
There was an article in the New Yorker about it. Supposedly, Russia was developing a strain of smallpox, that was immune to the American vaccine, as a biological weapon.
Smallpox. Most of the population isn't vaccinated, and with the high rate of mobility in today's society, you can have a worldwide epidemic on your hands before the victims show the symptoms of anything worse than a flu.