And apparently dont have anyone familiar with the concept of security either. From a security standpoint physical access equals control, period. There are no security measures that can protect you if the user has physical access to the hardware. They might as well have a big red switch on the front of the box saying 'press here for fast cable, but you're not allowed to press this button'.
Never, _ever_ trust the client side to be secure or in your control.
Not needing a gui to run != not having a gui for admin tasks.
I get sortof annoyed with Linux people trying to 'show off their cli stuff' when teaching Windows admins how to run Linux. It's counterproductive and unnecessary. Linux is eminently administratable via gui, and the cli stuff will come in its own time. While it's faster and more efficient and scriptable, it's also not necessary for most tasks. Any admin worth his wage will learn it when he becomes familiar with the system.
Of course, "X1" bought the "V" commercial application, then decided they wanted to have the same feature in "V" as "X" had paid to get in "Y". Company "V Inc" charges five times the original price for "V" to add that feature. So in the end, "Y" was cheaper anyway.
The GPL only controls the distribution of the GPL code. You can license your copyrighted derivative code any way you want, but if that way isnt compatible with the GPL license you cannot distribute the GPL portions of the code. You can still do whatever you want with the code you hold copyright to.
Parodies and fair use are quite different beasts. You can probably get away with 'fair use' of GPL code if you take and distribute a function to write a review about it. You're not allowed to take large chunks of code and add it to your own however. Nor are you allowed to take 8 chapters out of a book, change the names and rewrite the ending and call it parody either. You would probably be able to get away with looking at the code to a GPL ls as you rewrite it as a parody however. Maybe that's what they did with 'dir'...
"Not true. It takes away the right of the author: - to keep his source code secret."
No it doesnt. He can keep his source code secret all he wants. He just isnt allowed to distribute the GPL code if he does. If he feels like distributing his own binaries without the GPL code he can go right ahead.
"- to limit the distribution of his work."
He can do whatever he wants with his own work. The GPL only affects the GPL code.
- to pursue a traditional software business model."
Again, he's free to do whatever he wants. The GPL affects the GPL part of the code (and only in case he does not own the copyright to it), not the code the author has written himself.
The GPL only gives rights. It grants the right to distribute the GPL code if you fulfill certain conditions. Copyright is what restricts rights; if you take code out of Solaris or Windows you hardly get rights to distribute those just because your code depends on the code you took.
You cant place any combined copyrighted software into the public domain unless you own the copyright. You cant do it with GPL code, you cant do it with BSD code, you cant do it with MIT-X11 code.
The GPL isnt any more viral than any other copyrighted code. It's just that the GPL places some restrictions on what you can do with the GPL code.
Like you say, you can place your part of the combined code in the public domain tho, so it's possible for someone else to lift your code out of the combined code and include it and distribute it in whatever way they want. The GPL code remains under the GPL tho, just as BSD code would, or Microsofts or anyone elses copyrighted code.
That's why you have a build environment and strangle anyone who plays around with it. As long as the build environment is identical with the target platform you dont get that sort of trouble.
Autoconf and automake are a bit troublesome to set up, and they wont give you an immediate benefit until you migrate your checks over to use them, but there's a reason that so many portable opensource products use them. As long as you use them the way they're meant to be used they solve a whole range of portability problems, and will cut down the time spent on portability by a lot.
However, if you read the documentation on threading and forking under linux you would realize that it's actually not forking a new process. Read up on clone(2).
The point is, tanks, stealth fighters, cruise missiles, etc are rather harder to obtain than armaments that you can carry around. They also require a fair bit more infrastructure to support.
Portable lasers that wont be much harder to obtain than stingers or similar arms _and_ are unlikely to be one-shot weapons ensure that pretty much any little dictator to be will have access to unlimited charge weapons that can easily slice dice and chop the stealth fighters and cruise missiles into small pieces.
It's a great way to level the playing field so that anyone and their little dog will be able to shoot down US planes en masse, but I somehow doubt that's the idea.
Nuclear arms are, however, not generally of the kind that are really easy to smuggle, nor are they generally carried by ground and/or air forces, nor are the raw materials very easy to obtain or handle.
Weapons like portable lasers fall into the same category as rifles, stingers etc. You dont need to raid an ICBM silo to obtain one. Shoot down one plane or helicopter using conventional means or take some ground troops and if you're lucky you've got yourself a laser. Which is probably not a one-shot per unit weapon.
These weapons wont help the US. They'll equalize the playing field even more as dubious regimes obtain them by making the US air force so much flying target practice.
"Lasers are in line with Rumsfeld's idea of transforming the military, which is to come up with wonder-weapons that other countries can't emulate."
Uh, yeah. Guerillas, the Taliban, etc all have these huge stinger factories and AK factories, because they're really actually making them themselves.
Riiiight.
Just like everything else it'll take a year or two and then it's out on the weapons market and in five years more people like Saddam and the Taliban have laser batteries slicing and dicing US bombers into teeny weeny metal squares.
Other countries dont emulate weapons. They buy them, steal them, smuggle them or are given them. One thinks that a defense secretary would know this, but apparently learning requires actually having a brain. Maybe he can buy one on the black market.
Frankly, I doubt it's the admins. They've probably configured uploading privilidges for marketing or the CFO or CEO or someone, believing someone responsible for something like this in management has perhaps skimmed through company regulations on confidential data. Admins are rarely involved in the actual launch of things like financial statements unless it's a really really small company.
If it's the CEO or CFO who did such a mindnumbingly stupid thing then the 'sue Reuters' rather than 'fire the responsible person' strategy makes much more sense (well, not really to a sane person, but we all know that sanity isnt a requirement for advancing to executive positions in a corporation).
"The incident has severely damaged confidence in us as individuals and in Intentia as a company," says Björn Algkvist, CEO of Intentia International AB."
Um, yeah. If you cant tell the difference between 'storing confidential data in an access controlled place on your internal network' and 'storing confidential data on an open-for-all external site' it sure will damage my confidence in Intentia as a company. Incompetent is a fairly fitting description.
Superhuman aim and reflexes is possible because the client tells the server 'player shoots at positition X'. Again, you trust the client. If the client instead tells the server 'player requests character turn left', 'player requests aim move down' and 'player presses shoot', you dont get the same problem. Or you can use a combination depending on the game setup.
Sure, it isnt perfect. Above all, you'll get latency problems for players since they are going to be limited by both server-set maximum turning and aiming speed as well as reality disconnect due to lag between client and server. You can optimize it a whole lot tho, and it _will_ stop aimbots and other cheating from being even near as effective as it is today.
Palladium still wont stop cheating tho, it just makes it a bit harder. Even if the game binaries are untouchable, the network packets arent. It'll just move to more advanced proxycheats instead.
The perfect security for online gaming is already here. It's simple. It's called: Put the security on the server side. Treat the client as a dumb terminal and only send data meant for the player and only recieve data indicating the requested actions of the player. It aint rocket science, it's just that some game programmers appear to be about as dense as a black hole or have spent the last 15 years living in a box without ever reading about anyone else implementing a networked game.
Open source programs are distributed in the preferable form for modifying them. Source code, that is. The reason they are distributed with source code is that it's really really easy to change.
Binary proprietary programs are only really easy to change, as opposed to really really easy. Changing something like a MAC adress reading is trivial with a debugger.
If the last thing they want is users making changes to this program then they shouldnt be distributing it at all So it still isnt any use to them.
Hint: security for multiplayer games is done server side or you are totally, completely and utterly screwed from beginning to end and nothing you can ever do about it will change that. A player can always see any data ever sent to the client and always control and make up any data going back to the server. Anyone even dreaming otherwise is deluding themselves.
Sure it will work 90% of the time. For the 90% that dont cheat, that is.
The average Cheater Joe off the street will definitely know exactly how to change it. Which makes the whole exercise pointless.
Heck, client side security with no passwords and disks shared to the world works great 90% of the time. Unfortunately it isnt the 90% that is the problem. It's the rest. And for the rest, repeat after me, client-side security will never ever ever work. If you dont have physical control over a computer you cannot trust anything it tells you.
The present system - Even apart from the ethical issues of denying treatment to patients due to patents there are several problems with the system.
Pharmaceutical companies are, or have become, herd animals. Large amounts are invested that target redundant treatments. We dont need an unlimited number of ulcer treatments, antidepressants and painkillers just because they happen to be popular sellers. Many of them are not significant improvements, or even any improvement over generic drugs.
Pharmaceutical company have no incentive to genereate cures. Returning customers are more profitable.
The efficiency of research and knowledge sharing within the industry may not be at the highest possible level; there is profit in not letting your competitors know about successful or even unsuccessful developments.
Base research is being patented. Patents are not designed to be awarded for discoveries, only inventions. Unfortunately, that concept appears to have fallen out of the patent offices guidelines.
Yes, I'm prepared to vote for disbanding the research part of the pharmaceutical industry. The kind of people running the industry arent trustworthy and they've abused the patent system long enough.
True, the R&D costs for drugs are high and it's risky.
Which is why private enterprise should get out of it and leave it all to global collaborating state funded research instead. We dont need 25 patented cures for ulcers or 150 patented SSRI anti-depressants. We dont need 'sponsored' doctors switching patients from old and well tested generic medicines to new untried patented ones when the old ones worked just as well or better.
The pharmaceutical industry has shown it cant handle either the risks nor the ethics of the issues. It's time to relieve them of that responsibility.
Lets see, for those who cannot use BitKeeper (which with the license restriction in place mean pretty much anyone working at any major IT company since most large IT companies dabble in one place or the other with revision handling systems) there is no revision handling system at all.
So, an alternative that is as good as or better than nothing? Um... anything?
Except, of course, the GPL doesnt say you cant work on proprietary code at all if you work with GPL code. It doesnt say you have to give up your daytime job to use Linux at home.
The BK license says you cant use free bitkeeper for _anything_, not to check out linux kernel sources, not to write a program for the Red Cross, _if_ you develop any competing product in any other sort of function in your life.
The GPL dictates under what terms you can distribute the GPL software.
The free BK EULA dictates what you are allowed to spend your time doing.
Slashdot Mirror
And apparently dont have anyone familiar with the concept of security either. From a security standpoint physical access equals control, period. There are no security measures that can protect you if the user has physical access to the hardware. They might as well have a big red switch on the front of the box saying 'press here for fast cable, but you're not allowed to press this button'.
Never, _ever_ trust the client side to be secure or in your control.
I dont use Windows at home, and I dont use it at work. It's just not worth the hassle.
As far as games go there are enough, and when I get bored of those it's more fun to work on developing my own games instead.
So, exactly what of that does not Evolution support?
Not needing a gui to run != not having a gui for admin tasks.
I get sortof annoyed with Linux people trying to 'show off their cli stuff' when teaching Windows admins how to run Linux. It's counterproductive and unnecessary. Linux is eminently administratable via gui, and the cli stuff will come in its own time. While it's faster and more efficient and scriptable, it's also not necessary for most tasks. Any admin worth his wage will learn it when he becomes familiar with the system.
Of course, "X1" bought the "V" commercial application, then decided they wanted to have the same feature in "V" as "X" had paid to get in "Y". Company "V Inc" charges five times the original price for "V" to add that feature. So in the end, "Y" was cheaper anyway.
The GPL only controls the distribution of the GPL code. You can license your copyrighted derivative code any way you want, but if that way isnt compatible with the GPL license you cannot distribute the GPL portions of the code. You can still do whatever you want with the code you hold copyright to.
Parodies and fair use are quite different beasts. You can probably get away with 'fair use' of GPL code if you take and distribute a function to write a review about it. You're not allowed to take large chunks of code and add it to your own however. Nor are you allowed to take 8 chapters out of a book, change the names and rewrite the ending and call it parody either. You would probably be able to get away with looking at the code to a GPL ls as you rewrite it as a parody however. Maybe that's what they did with 'dir'...
"Not true. It takes away the right of the author:
- to keep his source code secret."
No it doesnt. He can keep his source code secret all he wants. He just isnt allowed to distribute the GPL code if he does. If he feels like distributing his own binaries without the GPL code he can go right ahead.
"- to limit the distribution of his work."
He can do whatever he wants with his own work. The GPL only affects the GPL code.
- to pursue a traditional software business model."
Again, he's free to do whatever he wants. The GPL affects the GPL part of the code (and only in case he does not own the copyright to it), not the code the author has written himself.
The GPL only gives rights. It grants the right to distribute the GPL code if you fulfill certain conditions. Copyright is what restricts rights; if you take code out of Solaris or Windows you hardly get rights to distribute those just because your code depends on the code you took.
You cant place any combined copyrighted software into the public domain unless you own the copyright. You cant do it with GPL code, you cant do it with BSD code, you cant do it with MIT-X11 code.
The GPL isnt any more viral than any other copyrighted code. It's just that the GPL places some restrictions on what you can do with the GPL code.
Like you say, you can place your part of the combined code in the public domain tho, so it's possible for someone else to lift your code out of the combined code and include it and distribute it in whatever way they want. The GPL code remains under the GPL tho, just as BSD code would, or Microsofts or anyone elses copyrighted code.
That's why you have a build environment and strangle anyone who plays around with it. As long as the build environment is identical with the target platform you dont get that sort of trouble.
Autoconf and automake are a bit troublesome to set up, and they wont give you an immediate benefit until you migrate your checks over to use them, but there's a reason that so many portable opensource products use them. As long as you use them the way they're meant to be used they solve a whole range of portability problems, and will cut down the time spent on portability by a lot.
However, if you read the documentation on threading and forking under linux you would realize that it's actually not forking a new process. Read up on clone(2).
The point is, tanks, stealth fighters, cruise missiles, etc are rather harder to obtain than armaments that you can carry around. They also require a fair bit more infrastructure to support.
Portable lasers that wont be much harder to obtain than stingers or similar arms _and_ are unlikely to be one-shot weapons ensure that pretty much any little dictator to be will have access to unlimited charge weapons that can easily slice dice and chop the stealth fighters and cruise missiles into small pieces.
It's a great way to level the playing field so that anyone and their little dog will be able to shoot down US planes en masse, but I somehow doubt that's the idea.
Nuclear arms are, however, not generally of the kind that are really easy to smuggle, nor are they generally carried by ground and/or air forces, nor are the raw materials very easy to obtain or handle.
Weapons like portable lasers fall into the same category as rifles, stingers etc. You dont need to raid an ICBM silo to obtain one. Shoot down one plane or helicopter using conventional means or take some ground troops and if you're lucky you've got yourself a laser. Which is probably not a one-shot per unit weapon.
These weapons wont help the US. They'll equalize the playing field even more as dubious regimes obtain them by making the US air force so much flying target practice.
He's also a complete moron. The next sentance:
"Lasers are in line with Rumsfeld's idea of transforming the military, which is to come up with wonder-weapons that other countries can't emulate."
Uh, yeah. Guerillas, the Taliban, etc all have these huge stinger factories and AK factories, because they're really actually making them themselves.
Riiiight.
Just like everything else it'll take a year or two and then it's out on the weapons market and in five years more people like Saddam and the Taliban have laser batteries slicing and dicing US bombers into teeny weeny metal squares.
Other countries dont emulate weapons. They buy them, steal them, smuggle them or are given them. One thinks that a defense secretary would know this, but apparently learning requires actually having a brain. Maybe he can buy one on the black market.
Because licenses usually say 'You're not allowed to blah blah blah (and then in really really fine print) except when regulated by law'.
It isnt fraud, they're just saying that 'this is an in-our-dreams license but law can override it until we've bribed some changes in it'.
Frankly, I doubt it's the admins. They've probably configured uploading privilidges for marketing or the CFO or CEO or someone, believing someone responsible for something like this in management has perhaps skimmed through company regulations on confidential data. Admins are rarely involved in the actual launch of things like financial statements unless it's a really really small company.
If it's the CEO or CFO who did such a mindnumbingly stupid thing then the 'sue Reuters' rather than 'fire the responsible person' strategy makes much more sense (well, not really to a sane person, but we all know that sanity isnt a requirement for advancing to executive positions in a corporation).
"The incident has severely damaged confidence in us as individuals and in Intentia as a company," says Björn Algkvist, CEO of Intentia International AB."
Um, yeah. If you cant tell the difference between 'storing confidential data in an access controlled place on your internal network' and 'storing confidential data on an open-for-all external site' it sure will damage my confidence in Intentia as a company. Incompetent is a fairly fitting description.
True, not as much now as it used to be, we're switching to FibreChannel attached storage for servers instead... ...which is even more expensive...
Superhuman aim and reflexes is possible because the client tells the server 'player shoots at positition X'. Again, you trust the client. If the client instead tells the server 'player requests character turn left', 'player requests aim move down' and 'player presses shoot', you dont get the same problem. Or you can use a combination depending on the game setup.
Sure, it isnt perfect. Above all, you'll get latency problems for players since they are going to be limited by both server-set maximum turning and aiming speed as well as reality disconnect due to lag between client and server. You can optimize it a whole lot tho, and it _will_ stop aimbots and other cheating from being even near as effective as it is today.
Palladium still wont stop cheating tho, it just makes it a bit harder. Even if the game binaries are untouchable, the network packets arent. It'll just move to more advanced proxycheats instead.
The perfect security for online gaming is already here. It's simple. It's called: Put the security on the server side. Treat the client as a dumb terminal and only send data meant for the player and only recieve data indicating the requested actions of the player. It aint rocket science, it's just that some game programmers appear to be about as dense as a black hole or have spent the last 15 years living in a box without ever reading about anyone else implementing a networked game.
Open source programs are distributed in the preferable form for modifying them. Source code, that is. The reason they are distributed with source code is that it's really really easy to change.
Binary proprietary programs are only really easy to change, as opposed to really really easy. Changing something like a MAC adress reading is trivial with a debugger.
If the last thing they want is users making changes to this program then they shouldnt be distributing it at all So it still isnt any use to them.
Hint: security for multiplayer games is done server side or you are totally, completely and utterly screwed from beginning to end and nothing you can ever do about it will change that. A player can always see any data ever sent to the client and always control and make up any data going back to the server. Anyone even dreaming otherwise is deluding themselves.
Sure it will work 90% of the time. For the 90% that dont cheat, that is.
The average Cheater Joe off the street will definitely know exactly how to change it. Which makes the whole exercise pointless.
Heck, client side security with no passwords and disks shared to the world works great 90% of the time. Unfortunately it isnt the 90% that is the problem. It's the rest. And for the rest, repeat after me, client-side security will never ever ever work. If you dont have physical control over a computer you cannot trust anything it tells you.
The present system - Even apart from the ethical issues of denying treatment to patients due to patents there are several problems with the system.
Pharmaceutical companies are, or have become, herd animals. Large amounts are invested that target redundant treatments. We dont need an unlimited number of ulcer treatments, antidepressants and painkillers just because they happen to be popular sellers. Many of them are not significant improvements, or even any improvement over generic drugs.
Pharmaceutical company have no incentive to genereate cures. Returning customers are more profitable.
The efficiency of research and knowledge sharing within the industry may not be at the highest possible level; there is profit in not letting your competitors know about successful or even unsuccessful developments.
Base research is being patented. Patents are not designed to be awarded for discoveries, only inventions. Unfortunately, that concept appears to have fallen out of the patent offices guidelines.
Yes, I'm prepared to vote for disbanding the research part of the pharmaceutical industry. The kind of people running the industry arent trustworthy and they've abused the patent system long enough.
True, the R&D costs for drugs are high and it's risky.
Which is why private enterprise should get out of it and leave it all to global collaborating state funded research instead. We dont need 25 patented cures for ulcers or 150 patented SSRI anti-depressants. We dont need 'sponsored' doctors switching patients from old and well tested generic medicines to new untried patented ones when the old ones worked just as well or better.
The pharmaceutical industry has shown it cant handle either the risks nor the ethics of the issues. It's time to relieve them of that responsibility.
Lets see, for those who cannot use BitKeeper (which with the license restriction in place mean pretty much anyone working at any major IT company since most large IT companies dabble in one place or the other with revision handling systems) there is no revision handling system at all.
So, an alternative that is as good as or better than nothing? Um... anything?
Except, of course, the GPL doesnt say you cant work on proprietary code at all if you work with GPL code. It doesnt say you have to give up your daytime job to use Linux at home.
The BK license says you cant use free bitkeeper for _anything_, not to check out linux kernel sources, not to write a program for the Red Cross, _if_ you develop any competing product in any other sort of function in your life.
The GPL dictates under what terms you can distribute the GPL software.
The free BK EULA dictates what you are allowed to spend your time doing.
Thats quite different.