Of course, if being honor students, gifted muscicians, eagle scouts, and a 4 of the damn nicest people I've ever had the joy of knowing is "messed up" I'm also damn glad I don't take this view.
Actually, that is kind of messed up. I've never met a kid who aimed to be such a "perfect" child without being pushed or shamed by the parents.
When you teach a child to ride a bike, you eventually have to let go of the handlebars and let them go it alone. And they may crash, many times even. But that is life. Similarly, if you keep an omnipresent eye over everything your child does until they leave the house, how prepared do you think they will be to go it alone? Not at all.
The people I've met with overprotective parents inevitably went "wild" when they finally got to college. Those that didn't were somewhat "off", and had trouble interacting with the regular populace (and by "regular", I don't mean "party-goer").
As for me, my parents were practically negligent of my upbringing. I wouldn't suggest others treat their kids the same way, but I've turned out OK. When facing the challenges laid before me by life, such as whether to use drugs, whether to party rather than study, etc, I always knew I was on my own two feet. No one was there to catch me if I fell.
I never considered, "If I do this, what will my parents think?" They didn't really care one way or the other. Instead, I always thought, "What impact will this have on MY future?" Someone who's set about controlling another's life, whether with good intentions or bad, inevitably stands in an adversarial position to that person.
That person becomes your "enemy", whether it is spoken or not. Things become wanted and sought after not because of their intrinsic worth, but because in a struggle for freedom the child invariably "rebels" and strives to escape the shackles of the controlling environment. The child doesn't want drugs, or promiscious behavior, but really wants freedom.
Of course, this doesn't accurately portray every kid. There are plenty that pursue life's vices simply because they are bored or don't care, and get away with it because the parents are negligent. But you should keep in mind that your attempts to control your children are very likely to backfire later in life.
Kids want privacy, and freedom over their selves. Take those away, and they will hate you, whether they admit it or not.
I've used Linux for eight years and I can definitely tell you that all those nice GUIs provided by Red Hat, Mandrake, and SuSE just hide the stuff you're going to have to learn anyway. It all comes down to monkeying with flat files off the/etc tree.
I'll admit, I use the GUI tools when I can. But that's because becoming an expert is not what I'm trying to get out of Linux now. I just want to use it. If I had no philosophical objections to proprietary software, I'd probably be using Windows.
I guess what I'm trying to say is that you'll become better able to use Red Hat by figuring out Gentoo or LFS. It's like lifting 100lb barbells to build up for the shot-put.
An anecdote to put this in perspective is a conversation I had with my hiring manager. She said that people who only know C and C++ can pick up Java and VB without any trouble, but those who only know Java or VB have great difficulty with C++. So when looking for candidates, she didn't even care whether someone knew VB or Java, she only cared about C/C++.
Similarly, if I had to hire a Red Hat admin or developer, then I'd choose someone who had only used Gentoo over someone who had only used Red Hat. Why? They are better able to do the job.
Windows and related software isn't discounted for academics? What about all those other people saying they get XP for $5, or free? And I can't think of many applications that are free for Linux but have no free (for personal use, at least) equivalents for Windows.
That $25 bucks doesn't get you much. System updates? You get that with Debian or Mandrake for free.
The only thing Red Hat has going for it over Windows is that it is transparent where Windows is opaque. So you can at least look at the code even though you're not allowed to copy the software. But anyone who can't take the time to use Gentoo or Debian isn't going to have much use for their kernel or application source code.
FYI, students actually have studying to do, so our time is not necessarily cheap.
Then use Windows. The whole point of Linux and GNU was to get away from proprietary systems and restrictive licensing. So you really don't gain anything by installing either Red Hat or SuSE over Windows, unless you really like the Linux interface, proprietary or no.
On the other hand, many install less popular operating systems for the experience of running them. I know I did. But as these commercial distros become easier to install and use, the knowledge and experience you gain from setting them up yourself is negligible.
Red Hat does an end run around the GPL forcing customers to buy a support contract for every installation, and key components of SuSE have been proprietary since the beginning.
Just use something else, don't reward these companies. If you're thinking of taking Red Hat or SuSE up on these offers, look elsewhere. Mandrake, Gentoo, Debian, Slackware, from scratch, whatever. You're a student--you're time is cheap. And if you actually want to learn something from using Linux, none of the commercial distros are the way to go.
Remember when WineX said they would totally open-source all of their work, but only when they got enough "donations"? Well they started getting money and decided, "Well, gee, we really like getting paid for what we do. Forget about the whole open-souncing thing."
To be fair, the issue is slightly more complex than that, but this looks to me like another WineX. Don't ever expect Linuxant's product to be totally open-source. Furthermore, just as WineX only works for the most simple or popular of games, don't expect Linuxant's product to work but for the simplest or most popular drivers.
If it was a library that had everything, i.e., something similar to the Library of Congress, then stocking Playboys would be appropriate. Though, of course, it still would not be appropriate to place them right beside the children's books.
But given that libraries cannot generally carry every book, they must make decisions as to what would carry more social value. Thus the New York Times, Wall Street Journal, Washington Post, etc. ar all going to be purchased before Playboys or Penthouses. It's not that Playboys don't belong, it's that they have less value than other alternatives.
Providing access to the Internet is providing access to a wealth of information. Librarians don't have to make a conscious choice for every site. Including all sites is just as easy as including any. It actually takes more effort to filter than simply allow everything, and so the reasons why Playboys are not in libraries is disanalogous to Internet filtering.
What would be more analagous is if a library bought an extremely large encyclopedia, and actively ripped out pages containing offensive content. Most encyclopedias, of course, don't have anything as objectionable as what you can find on the Internet, but the principle is still the same.
Um I am not sure how that is the case? The Republicans (IE Conservatives) in America are in general Anti-Gun Control. This will only strengthen their idea that filtering is a liberal issue.
I'm not following your logic. Typically, overzealous filtering has enraged liberals because it blocks sites that discuss breast cancer, safe sex, or "heretical" opinions. And in general, many liberals take issue with the fact that filtering library connections interferes with academic freedom.
This is the first time I am aware that filtering has struck a nerve with bedrock conservatives. It seems, unfortunately, that the response of many conservatives here is that:
1) Ownership of guns is morally defensible.
I, and many liberals, on the other hand, would argue that this filtering is wrong because:
2) Political advocacy should never be filtered.
It seems that many conservatives hold that (2) is false, and rather than giving it up, take up (1) instead. But (1) shouldn't matter. This is a free speech issue, not a gun-ownership issue. I hoped that conservatives would agree with this, but it appears they don't.
Well, it's already shaping up into a liberal vs. conservative flamewar. But it shouldn't be. I hope Symantec's actions help Republicans realize that overzealous filtering isn't just a liberal issue, it's an everyone issue.
Completely agree. I've used MS products at work for the past few years because I had to. Despite all the ramblings on slashdot otherwise, I can't say that they are technically inferior. I would have done things differently in many places, and I have run into some obscure bugs. But all-in-all, they do allow the job to get done.
If the choice of vendors were up to me, however, I would never choose MS. And it has nothing to do with price or technical merit. It has to do with openness. I can't believe that people are still keeping all of their data in proprietary formats when so many open alternatives are available.
As soon as you start depending on proprietary products and data from software company X, you have ceded control over to them. Rather than them working for you, you work for them. Back in the days when it was proprietary or nothing, it's more understandable. But today there is no excuse. The few wiz-bang features that MS provides simply is not worth the price of handing over so much control to them.
Yes, you can always migrate off of a proprietary solution. But the longer you wait, and the more you let the proprietary solution seap into the bowels of your business, the more costly it will be. And do you think software vendor X is going to help you migrate? Fat chance. Company X will do everything it can to make staying with their product cheaper than migrating. But the cost of migration is usually so high that this hardly reduces the price of product at all.
It's like forking an opponents queen and rook with a knight in chess. Whatever the opponent does, they are screwed.
If you can manage to "extend" in the java sense or use ":" for inheritance in a C++ sense without touching point 3 or point 4 of your list, then you do not violate the authors rights of the code from wich you inherit. Regardless of license.
If you could, sure. But you can't. Extending a class is a strong form of static linking. If you know how to extend a class by linking it dynamically, or extend a class without linking at all, then I am very interested to learn how.
Your understanding is wrong. There is a license that works the way you described, it's called the CPL.
The GPL does not allow proprietary and GPL code to occupy the same memory space, though it is somewhat vague. For example, if there are a set of GPL C++ classes, and you derive from these classes, then your derivations fall under the GPL. If you link in a static or dynamic library that is GPL, then your code is also GPL. Let's break it down:
Invoking a GPL program in a separate memory space from a proprietary program, sending it input, and reading output - OK under the GPL.
Packaging your proprietary programs with a GPL program - OK under the GPL.
Loading a GPL dynamic library and having your proprietary program invoke functions from it - Not OK under the GPL, but it is OK under the LGPL.
Statically linking a GPL library with your proprietary program - Not OK with the GPL, but some libraries have a waiver that allows precisely this, such as glibc.
Pasting GPL code into your proprietary program - Not OK under any GPL derived program.
Extending GPL coded classes in object-oriented code - Not OK in any version of the GPL that I am aware of, though a waiver clause similar to glibc may be possible.
Last Christmas everyone at our small (50 person) company got fairly nice bonuses. Right after the new year, several business deals fell through, and many had to be laid off. The lay-offs may have been avoided if the bonuses weren't so high. The smaller you are, and the fewer clients you have, the more likely you are to find yourself in a similar situation. So, yes, give bonuses, but don't go overboard--you and your employees may end up regretting it.
Fishing expeditions are not allowed. When you are the plaintiff, you have to have your evidence ready before entering court. Defendant has much more leeway in discovery. Otherwise, as PJ pointed out, you could simply file claims against random people, searching for any infringement anyone might have done. This would be abuse of the court system. It would be like allowing the police to preemptively search random people's homes without reasonable suspicion. Sure, they would find more evidence and prosecute more crimes, but the benefits do not outweigh the consequences.
SCO shouldn't need evidence--everything is already laid out, "infringing" source code is known to all. But they have nothing, so they are abusing discovery to go fishing. IBM, on the other hand, can ask for pretty much anything, since they didn't bring action.
You make some good points, but you are making some assumptions:
1) The non-profit has money to pay staffers, rather than relying on volunteers. 2) No techie experts are volunteering for the non-profit.
If you have a surfeit of human resources, and very little equipment, then open source makes a lot of sense.
My impression is that these non-profits receiving the donated computers are generally short on cash. If they have 30,000 to pay a staffer, then they likely have a few hundred to spare for a computer.
But they may not have Linux expertise, and that is a problem. On the other hand, no computer runs itself, and Windows problems can stymie all but the most ambitious self-learners.
I am not in great disagreement with you. Much will depend on how "Red Hat Enterprise Linux" is defined. If simply taking out the few proprietary packages causes it to no longer be "Red Hat Enterprise Linux", then sure, there's no problem. But I don't think most people are buying RHEL for the add-on packages. It's for third party commercial compatibility and a stable kernel. Red Hat, perhaps out of self-interest, does not make it clear that one can simply remove a few packages and redistribute at will. I should qualify that--there are clear documents outlining how to resell the personal edition, but I saw none-such for the enterprise editions.
It is easy to recompute from the initial position for each iteration if you know that no new information is going to influence the equation after the player has jumped. However, this assumption doesn't hold, as was noted in the explanation. A rocket may come zooming in and push the player off his path. Or the player may exert "air control".
The only way these "free will" elements can be accounted for with a recomputation from the initial position is if every acceleration vector is stored for the entirety of the player's airborne trip. That is, a movement LEFT will have to be stored, and then if the player moves RIGHT that also has to be stored. Then you have to recompute from the beginning:
Jump vector - acceleration + LEFT movement + RIGHT movement + influence of striking rocket + air vent influence, etc.
It's not impossible, but it is not a simple coding change. Even the best throw in a stupid hack every now and then, but I don't think this is one of them.
What you're doing is expecting Red Hat to give its support away for free, or not support at all. That's not mandated by anyone's reading of the GPL.
What I'd like to see ideally is a support contract that doesn't interfere with the rights granted under GPL software. The way to do this is, in my view, not by placing restrictions on the customer's use or distribution of the product. Rather, they should tie the support contract to a machine.
I understand Red Hat's motivation. They don't want someone buying a support contract, and then installing the software on 10 machines. If this was allowed, the first machine to have a problem would suddenly be the one that the support contract is tied to. This allows the customer to get support for ten machines for the price of one.
Red Hat's response has been that only one copy of their product may be running at the customer's site for every support contract. Unfortunately, this interferes with the GPL. What they should say, rather, is that machines must be tied to the support contract before any support is delivered. I guess they could use dongles, but it given the remote nature of support, perhaps a better option would be assigning each support contract to a MAC address.
Of course, every means to enforce compliance can be circumvented. A router could make all machines seem to come from one MAC address. But their current means are not any better, it relies on the word of the customer.
So you return the box and download the SRPMs (with no restrictions) from greenshoe.com. I'm out $1500 and you can make copies for all your friends. If a nasty bug in my code causes your computer to burst into flames, I won't help, but that's your choice. What's the problem here?
The GPL only forces source distribution to those that the binaries were distributed to. Under your scenario, a company could only provide source to those that buy the software, but restrict those that receive the source from redistributing it via EULA or some other rider contract. However, this contradicts the terms of the GPL. The GPL expressly forbids placing such terms on redistribution of GPL software.
If such restrictive rider agreements are allowed, then compliance with the GPL is entirely optional. But this essentially nullifies the GPL. The whole point of the GPL is that restrictions on redistribution of source and binaries are not an option. If the libre requirement of the GPL cannot be enforced, then the GPL essentially reduces to being the BSD license.
What I'm saying is that if the GPL is to maintain any force, it must maintain all force. As soon as there is one exception that allows a vendor to circumvent the libre portions of the GPL, then every vendor and every product can use this exception, leaving the GPL without force.
Just because you aren't allowed to back out of a contract you agreed to doesn't mean your freedoms guaranteed by the GPL are being taken away.
Sure it does! Otherwise, everyone could attach a simple rider contract to the sale of GPL software that simply said, "By purchasing this software, you waive all rights granted to you under the GPL." If such riders were allowed, the GPL would be meaningless.
Sheesh. Yes I have read the GPL, many times. As I've already said, if there are non-GPL components, they can attach whatever license they want to those. I was restricting my discussion to the GPL components. If replacing the logos, icons, and sundry other copyright pieces suddenly makes the product no longer "Red Hat Advanced Server" in terms of the support contract, then you may have a point.
I guess it depends on how the contract defines the software product, and precisely how much non-GPL software is included. I was under the (possibly mistaken) impression that Red Hat generally keeps all their distros clean of proprietary stuff, except for the notable exception of logos and icons.
This is from fallible memory, but my impression is that the biggest difference between the personal and enterprise editions was not in software add-ons, but in features compiled into the kernel. I was describing the problem from the view of a typical person who just wants a basic Red Hat distro layout with the more advanced kernel.
To take another example: A warranty for your router will only be valid if you do not modify it yourself, and that is true even when the software in the router is GPL. They do not restrict your right to change the code, but the warranty will still be affected (nullified) by it.
I think this example works in my favor. If you can extract the GPL software from the product without affecting it, or if Linksys or whoever provides this software to you separate of the product, then you can do whatever you want with it, redistributing it to your heart's content. RHAS puts restrictions on use and redistribution, not just for the original package (as would be the case for a router or some such), but also for any copies that are made of the original package. This is where, I believe, it is in a grey area of compliance with the GPL.
Just because there is GPL software in it , doesn't mean GPL can "override" that license redhat stamps on the distro. (Please read the GPL)
I've read the GPL, and yes, I know that non-GPL software does not become GPL just because it is packaged together. Most notably, Red Hat includes easily removeable "signature" packages that identify the distro as Red Hat. This includes icons, logos, etc. They can put any terms they want on non-GPL software. I was restricting my discussion solely to the GPL software.
You have a support contract for *n* servers. If you install this software on more than *n* servers then your support contract is invalidated and all of your servers will be unsupported.
I did not see the part in the contract where it says violation simply implies termination. They could sue you upon termination for pretty much any amount. And what if this amount was keyed to the number of redistributions of the software?
I've studied the GPL quite extensively in regards to just this issue. As far as I can see, the GPL does not allow "rider" agreements to be attached to the distribution of software that restrict redistribution or use.
What people are saying here is that, while you have to buy both at once, you can simply violate the support contract if wishing to exercise rights under the GPL. But even if violation of the support contract ends with a clean "termination", that is, without further consequences, I still see it as questionable that this is compatible with the GPL. The support contract is a contract after all, and violating it is a violation of law. So you have to literally break the law to exercise your full rights under the GPL.
The problem is not with non-GPL components, it's the fact that they will not sell you RHEL without a support contract. The support contract says, among other things, that you will not install the software on more than one machine. Total violation of the GPL. This has been brought up before and people have said that it is just the terms of the support contract that will be cancelled if the software is installed on more than one machine. However, this is not in the language of the agreement.
Slashdot Mirror
When you teach a child to ride a bike, you eventually have to let go of the handlebars and let them go it alone. And they may crash, many times even. But that is life. Similarly, if you keep an omnipresent eye over everything your child does until they leave the house, how prepared do you think they will be to go it alone? Not at all.
The people I've met with overprotective parents inevitably went "wild" when they finally got to college. Those that didn't were somewhat "off", and had trouble interacting with the regular populace (and by "regular", I don't mean "party-goer").
As for me, my parents were practically negligent of my upbringing. I wouldn't suggest others treat their kids the same way, but I've turned out OK. When facing the challenges laid before me by life, such as whether to use drugs, whether to party rather than study, etc, I always knew I was on my own two feet. No one was there to catch me if I fell.
I never considered, "If I do this, what will my parents think?" They didn't really care one way or the other. Instead, I always thought, "What impact will this have on MY future?" Someone who's set about controlling another's life, whether with good intentions or bad, inevitably stands in an adversarial position to that person.
That person becomes your "enemy", whether it is spoken or not. Things become wanted and sought after not because of their intrinsic worth, but because in a struggle for freedom the child invariably "rebels" and strives to escape the shackles of the controlling environment. The child doesn't want drugs, or promiscious behavior, but really wants freedom.
Of course, this doesn't accurately portray every kid. There are plenty that pursue life's vices simply because they are bored or don't care, and get away with it because the parents are negligent. But you should keep in mind that your attempts to control your children are very likely to backfire later in life.
Kids want privacy, and freedom over their selves. Take those away, and they will hate you, whether they admit it or not.
I've used Linux for eight years and I can definitely tell you that all those nice GUIs provided by Red Hat, Mandrake, and SuSE just hide the stuff you're going to have to learn anyway. It all comes down to monkeying with flat files off the /etc tree.
I'll admit, I use the GUI tools when I can. But that's because becoming an expert is not what I'm trying to get out of Linux now. I just want to use it. If I had no philosophical objections to proprietary software, I'd probably be using Windows.
I guess what I'm trying to say is that you'll become better able to use Red Hat by figuring out Gentoo or LFS. It's like lifting 100lb barbells to build up for the shot-put.
An anecdote to put this in perspective is a conversation I had with my hiring manager. She said that people who only know C and C++ can pick up Java and VB without any trouble, but those who only know Java or VB have great difficulty with C++. So when looking for candidates, she didn't even care whether someone knew VB or Java, she only cared about C/C++.
Similarly, if I had to hire a Red Hat admin or developer, then I'd choose someone who had only used Gentoo over someone who had only used Red Hat. Why? They are better able to do the job.
Windows and related software isn't discounted for academics? What about all those other people saying they get XP for $5, or free? And I can't think of many applications that are free for Linux but have no free (for personal use, at least) equivalents for Windows.
That $25 bucks doesn't get you much. System updates? You get that with Debian or Mandrake for free.
The only thing Red Hat has going for it over Windows is that it is transparent where Windows is opaque. So you can at least look at the code even though you're not allowed to copy the software. But anyone who can't take the time to use Gentoo or Debian isn't going to have much use for their kernel or application source code.
On the other hand, many install less popular operating systems for the experience of running them. I know I did. But as these commercial distros become easier to install and use, the knowledge and experience you gain from setting them up yourself is negligible.
So why should you install Red Hat over Windows?:
Red Hat does an end run around the GPL forcing customers to buy a support contract for every installation, and key components of SuSE have been proprietary since the beginning.
Just use something else, don't reward these companies. If you're thinking of taking Red Hat or SuSE up on these offers, look elsewhere. Mandrake, Gentoo, Debian, Slackware, from scratch, whatever. You're a student--you're time is cheap. And if you actually want to learn something from using Linux, none of the commercial distros are the way to go.
Remember when WineX said they would totally open-source all of their work, but only when they got enough "donations"? Well they started getting money and decided, "Well, gee, we really like getting paid for what we do. Forget about the whole open-souncing thing."
To be fair, the issue is slightly more complex than that, but this looks to me like another WineX. Don't ever expect Linuxant's product to be totally open-source. Furthermore, just as WineX only works for the most simple or popular of games, don't expect Linuxant's product to work but for the simplest or most popular drivers.
If it was a library that had everything, i.e., something similar to the Library of Congress, then stocking Playboys would be appropriate. Though, of course, it still would not be appropriate to place them right beside the children's books.
But given that libraries cannot generally carry every book, they must make decisions as to what would carry more social value. Thus the New York Times, Wall Street Journal, Washington Post, etc. ar all going to be purchased before Playboys or Penthouses. It's not that Playboys don't belong, it's that they have less value than other alternatives.
Providing access to the Internet is providing access to a wealth of information. Librarians don't have to make a conscious choice for every site. Including all sites is just as easy as including any. It actually takes more effort to filter than simply allow everything, and so the reasons why Playboys are not in libraries is disanalogous to Internet filtering.
What would be more analagous is if a library bought an extremely large encyclopedia, and actively ripped out pages containing offensive content. Most encyclopedias, of course, don't have anything as objectionable as what you can find on the Internet, but the principle is still the same.
I'm not following your logic. Typically, overzealous filtering has enraged liberals because it blocks sites that discuss breast cancer, safe sex, or "heretical" opinions. And in general, many liberals take issue with the fact that filtering library connections interferes with academic freedom.
This is the first time I am aware that filtering has struck a nerve with bedrock conservatives. It seems, unfortunately, that the response of many conservatives here is that:
1) Ownership of guns is morally defensible.
I, and many liberals, on the other hand, would argue that this filtering is wrong because:
2) Political advocacy should never be filtered.
It seems that many conservatives hold that (2) is false, and rather than giving it up, take up (1) instead. But (1) shouldn't matter. This is a free speech issue, not a gun-ownership issue. I hoped that conservatives would agree with this, but it appears they don't.
Well, it's already shaping up into a liberal vs. conservative flamewar. But it shouldn't be. I hope Symantec's actions help Republicans realize that overzealous filtering isn't just a liberal issue, it's an everyone issue.
Completely agree. I've used MS products at work for the past few years because I had to. Despite all the ramblings on slashdot otherwise, I can't say that they are technically inferior. I would have done things differently in many places, and I have run into some obscure bugs. But all-in-all, they do allow the job to get done.
If the choice of vendors were up to me, however, I would never choose MS. And it has nothing to do with price or technical merit. It has to do with openness. I can't believe that people are still keeping all of their data in proprietary formats when so many open alternatives are available.
As soon as you start depending on proprietary products and data from software company X, you have ceded control over to them. Rather than them working for you, you work for them. Back in the days when it was proprietary or nothing, it's more understandable. But today there is no excuse. The few wiz-bang features that MS provides simply is not worth the price of handing over so much control to them.
Yes, you can always migrate off of a proprietary solution. But the longer you wait, and the more you let the proprietary solution seap into the bowels of your business, the more costly it will be. And do you think software vendor X is going to help you migrate? Fat chance. Company X will do everything it can to make staying with their product cheaper than migrating. But the cost of migration is usually so high that this hardly reduces the price of product at all.
It's like forking an opponents queen and rook with a knight in chess. Whatever the opponent does, they are screwed.
If you could, sure. But you can't. Extending a class is a strong form of static linking. If you know how to extend a class by linking it dynamically, or extend a class without linking at all, then I am very interested to learn how.
The GPL does not allow proprietary and GPL code to occupy the same memory space, though it is somewhat vague. For example, if there are a set of GPL C++ classes, and you derive from these classes, then your derivations fall under the GPL. If you link in a static or dynamic library that is GPL, then your code is also GPL. Let's break it down:
Last Christmas everyone at our small (50 person) company got fairly nice bonuses. Right after the new year, several business deals fell through, and many had to be laid off. The lay-offs may have been avoided if the bonuses weren't so high. The smaller you are, and the fewer clients you have, the more likely you are to find yourself in a similar situation. So, yes, give bonuses, but don't go overboard--you and your employees may end up regretting it.
Fishing expeditions are not allowed. When you are the plaintiff, you have to have your evidence ready before entering court. Defendant has much more leeway in discovery. Otherwise, as PJ pointed out, you could simply file claims against random people, searching for any infringement anyone might have done. This would be abuse of the court system. It would be like allowing the police to preemptively search random people's homes without reasonable suspicion. Sure, they would find more evidence and prosecute more crimes, but the benefits do not outweigh the consequences.
SCO shouldn't need evidence--everything is already laid out, "infringing" source code is known to all. But they have nothing, so they are abusing discovery to go fishing. IBM, on the other hand, can ask for pretty much anything, since they didn't bring action.
You make some good points, but you are making some assumptions:
1) The non-profit has money to pay staffers, rather than relying on volunteers.
2) No techie experts are volunteering for the non-profit.
If you have a surfeit of human resources, and very little equipment, then open source makes a lot of sense.
My impression is that these non-profits receiving the donated computers are generally short on cash. If they have 30,000 to pay a staffer, then they likely have a few hundred to spare for a computer.
But they may not have Linux expertise, and that is a problem. On the other hand, no computer runs itself, and Windows problems can stymie all but the most ambitious self-learners.
I am not in great disagreement with you. Much will depend on how "Red Hat Enterprise Linux" is defined. If simply taking out the few proprietary packages causes it to no longer be "Red Hat Enterprise Linux", then sure, there's no problem. But I don't think most people are buying RHEL for the add-on packages. It's for third party commercial compatibility and a stable kernel. Red Hat, perhaps out of self-interest, does not make it clear that one can simply remove a few packages and redistribute at will. I should qualify that--there are clear documents outlining how to resell the personal edition, but I saw none-such for the enterprise editions.
It is easy to recompute from the initial position for each iteration if you know that no new information is going to influence the equation after the player has jumped. However, this assumption doesn't hold, as was noted in the explanation. A rocket may come zooming in and push the player off his path. Or the player may exert "air control".
The only way these "free will" elements can be accounted for with a recomputation from the initial position is if every acceleration vector is stored for the entirety of the player's airborne trip. That is, a movement LEFT will have to be stored, and then if the player moves RIGHT that also has to be stored. Then you have to recompute from the beginning:
Jump vector - acceleration + LEFT movement + RIGHT movement + influence of striking rocket + air vent influence, etc.
It's not impossible, but it is not a simple coding change. Even the best throw in a stupid hack every now and then, but I don't think this is one of them.
What I'd like to see ideally is a support contract that doesn't interfere with the rights granted under GPL software. The way to do this is, in my view, not by placing restrictions on the customer's use or distribution of the product. Rather, they should tie the support contract to a machine.
I understand Red Hat's motivation. They don't want someone buying a support contract, and then installing the software on 10 machines. If this was allowed, the first machine to have a problem would suddenly be the one that the support contract is tied to. This allows the customer to get support for ten machines for the price of one.
Red Hat's response has been that only one copy of their product may be running at the customer's site for every support contract. Unfortunately, this interferes with the GPL. What they should say, rather, is that machines must be tied to the support contract before any support is delivered. I guess they could use dongles, but it given the remote nature of support, perhaps a better option would be assigning each support contract to a MAC address.
Of course, every means to enforce compliance can be circumvented. A router could make all machines seem to come from one MAC address. But their current means are not any better, it relies on the word of the customer.
The GPL only forces source distribution to those that the binaries were distributed to. Under your scenario, a company could only provide source to those that buy the software, but restrict those that receive the source from redistributing it via EULA or some other rider contract. However, this contradicts the terms of the GPL. The GPL expressly forbids placing such terms on redistribution of GPL software.
If such restrictive rider agreements are allowed, then compliance with the GPL is entirely optional. But this essentially nullifies the GPL. The whole point of the GPL is that restrictions on redistribution of source and binaries are not an option. If the libre requirement of the GPL cannot be enforced, then the GPL essentially reduces to being the BSD license.
What I'm saying is that if the GPL is to maintain any force, it must maintain all force. As soon as there is one exception that allows a vendor to circumvent the libre portions of the GPL, then every vendor and every product can use this exception, leaving the GPL without force.
Sure it does! Otherwise, everyone could attach a simple rider contract to the sale of GPL software that simply said, "By purchasing this software, you waive all rights granted to you under the GPL." If such riders were allowed, the GPL would be meaningless.
Sheesh. Yes I have read the GPL, many times. As I've already said, if there are non-GPL components, they can attach whatever license they want to those. I was restricting my discussion to the GPL components. If replacing the logos, icons, and sundry other copyright pieces suddenly makes the product no longer "Red Hat Advanced Server" in terms of the support contract, then you may have a point.
I guess it depends on how the contract defines the software product, and precisely how much non-GPL software is included. I was under the (possibly mistaken) impression that Red Hat generally keeps all their distros clean of proprietary stuff, except for the notable exception of logos and icons.
This is from fallible memory, but my impression is that the biggest difference between the personal and enterprise editions was not in software add-ons, but in features compiled into the kernel. I was describing the problem from the view of a typical person who just wants a basic Red Hat distro layout with the more advanced kernel.
I did not see the part in the contract where it says violation simply implies termination. They could sue you upon termination for pretty much any amount. And what if this amount was keyed to the number of redistributions of the software?
I've studied the GPL quite extensively in regards to just this issue. As far as I can see, the GPL does not allow "rider" agreements to be attached to the distribution of software that restrict redistribution or use.
What people are saying here is that, while you have to buy both at once, you can simply violate the support contract if wishing to exercise rights under the GPL. But even if violation of the support contract ends with a clean "termination", that is, without further consequences, I still see it as questionable that this is compatible with the GPL. The support contract is a contract after all, and violating it is a violation of law. So you have to literally break the law to exercise your full rights under the GPL.
The problem is not with non-GPL components, it's the fact that they will not sell you RHEL without a support contract. The support contract says, among other things, that you will not install the software on more than one machine. Total violation of the GPL. This has been brought up before and people have said that it is just the terms of the support contract that will be cancelled if the software is installed on more than one machine. However, this is not in the language of the agreement.