Sorry, but based on your sig, I would suggest you don't have a lot of credibility in assessing the President's motivations in any impartial way.
Had Mr. Bush admitted some change of heart or mind on any policy or issue of national importance, you would be pointing him out as hypocritical for calling Kerry a flip-flopper. You would be saying he talks out of both sides of his mouth. You'd be saying how the mighty have fallen. And whatever the issue was, it would be broadcast up and down the media every day between now and election day.
It was a no-win question, and all 3 candidates chose not to answer it. Only one candidate didn't try to hide behind double-speak.
...you can't believe that partisans will populate the entire judicial system.
They don't have to. With the world speeding up as it does, taking action to stop a terrorist may be a matter of hours, not days or weeks. Thus, the Patriot act has limited the oversight of the process to streamline it. One overzealous judge could in fact slow down the process such that we have another 9-11-type calamity on our hands.
This is why we sometimes cut corners protecting civil rights when there is a clear and present danger. I believe there still is one.
The preamble to our Constitution lists the reasons for its existence:
- Form a more perfect Union.
- Establish justice.
- Ensure domestic tranquility.
- Provide for the common defense.
- Promote the general welfare.
- Secure the blessings of liberty.
Each of these addresses a different area in which our Federal Government is to provide a measure of security for the People.
Union: Secure from the weaknesses of widely disparate and disputing local governments.
Justice: Secure from crime and wrongdoing.
Tranquility: Secure from riots and anarchy.
Defense: Secure from invasion and foreign threats.
Welfare: Protected against financial ruin, economic disaster. (Note it says promote, not provide.)
Liberty: Secure from Government control:Tyrany.
Yes, I do have to eventually trust my neighbor. But there are many things the government can do to encourage me and my neighbor to be trustworthy.
The basic "rights" are summed up in the Declaration: Life, Liberty, Pursuit of Happiness. No, they can't be guaranteed, but they can be secured.
There is a fine line to be found between protecting the rights of individuals and protecting the right of the People to be secure. The Patriot Act sought to define the line, giving the Executive more power to track these financial transactions, without scrutiny of the individual being investigated, and with limited oversight.
We need some kind of oversight, because the Executive may abuse the power. Not every executive will be as trustworthy as others in regard to protecting the rights of individuals.
One thing to consider, however, is that with judicial oversight, you can have another form of tyrany, where an overzealous judge prevents an Executive from doing his job to protect the People. We only have an appeals process for this, which hopefully results in a well-reasoned balance of rights. However, as the judicial confirmation process becomes more and more politicized, you can expect more and more partisans being placed in lifetime-tenured posts.
No judge is ever going to rule less power for the Judicial branch. I, for one, do not welcome our judicial overlords. Lex Rex.
Big deal. DMCA only applies to USA. Crackers in other parts of the world will break it and publish their results. Then someone else will write a nice application using the results, and people within the USA will download it and use it. Information wants to be free, and until the corporate entities come to this realization, they will be fighting a losing battle. DMCA cannot succeed in the long run to achieve its goals. Not without some kind of 1984-style monitoring, and americans are too libertarian / civil-liberties minded to accept this.
The biggest problem I have with the STL is that many debuggers don't have nice support for it. Thus, it's very difficult to observe what's stored in that nice vector or tree, when you hit the breakpoint. A few newer ones are adding automatic display of STL structures, which will be very nice, but for now, I'm stuck with VC++6, which is horrendous WRT these kinds of data structures.
I'd certainly welcome any ideas people have on how they actually debug code using the STL.
And, if your application doesn't happen to be written in this "verifiable subset", what's your choice? Trust it?
I assume all the default CLR subsystems are defaulted to reject non-safe code?
I wonder how many other OS's MS is supporting with their CLR? If I want to run under Linux, what's the likelihood that an MS produced bytecoded program will run there unmodified and untested. (Not that Java is there yet, but at least that's their goal. MS's goal doesn't seem to be WORA. It seems to be Add More Nonstandard Stuff, so they have to use our system. This is an understandable goal-- and I support their right to do it-- but I don't have to be drawn into the trap, either.)
Pardon me for not trusting our Redmond friends, it's just that they have a track record of abuse. Maybe they have a new attitude of philanthropy I haven't noticed yet.
Did you RTFA?.NET is fundamentally insecure because it decided to support the insecure features of C/C++. I'm not willing to get bit by these weaknesses years down the road when the script kiddies start taking advantage of them.
If the discovery is truly worth something in the real world, then it is worth licensing the patent-holder's property to achieve your goal. If not, then it is not.
That's awfully broad criteria for a lowly patent clerk to be using. I think they need a bit more guidance in how to achieve that.
I disagree with your statement about whether you would have gone for it anyway. If you would have gone for it and would have shared it once you had it, then maybe. Problem is, in the real world, that's an awful lot of hypotheticals to figure out, and is very much a matter of opinion. The job of the patent clerk has to be made easier. It's hard enough to decide whether something is novel. Seeing whether someone put in investment is easy. Perhaps we need people to detail the steps they went about in the creation of their idea, so the patent clerk could determine both the amount of effort and the likelihood of this being invented by another person without the same level of investment.
Perhaps I wasn't clear. Here's an example. Say you're a small-time inventor. You work out of your garage. You spend years, mortgage your inheritance, and you come up with an algorithm that helps producers deliver smaller/faster/cheaper. You want to see this innovation used over the world, but you also need to be reimbursed for your years of investment. You could create a software implementation that perhaps gets you some of that back. Problem is, today you have an extremely competitive market. You're not making your own market. You're competing against Microsoft, Peoplesoft, NA, and Pirhanasoft. Copyright and trade secret doesn't get you there, because with 10 minutes in a decompiler, they can determine your method and use it in their existing products, to their existing customer base. You get squashed like a bug. The only solution is patent protection for your algorithm. Without this protection, you are hesitant to mortgage your inheritance and spend years of perspiration to create your algorithm. So, since it is novel, it never gets created, and the world does not get the benefit. This is a chilling effect on the market. We want and need people to make those kinds of investments.
Now, the current system clearly is flawed. But the cure cannot be to eliminate the system for all software. Better would be to make sure that people only get patents who have invested, that it is truly novel -- nobody else would have come up with it without making the substantial investment. The important thing is that we reward the making of that investment, when it produces something marketable. Because, many times people will invest and fail. This is already a negative feedback. There must be a big positive payoff if you actually succeed.
Now, when the market was immature so many years ago, it may be seen that patent protection was not necessary. However, today, with such heavy competition and monopolization, it is much more needed. We can't just return to the good old days, when you could create something innovative, and expect copyright to make you rich. Copyright wasn't designed for this kind of protection, and ultimately, it fails to deliver what is needed. Trade secret coupled with anti-circumvention law may help, but what we really need is to allow maximum openness in promoting our ideas, so that those who are good at marketing can do the marketing, and those that are good at research and theory can work in their lab and be paid to keep innovating. This is what's ultimately best for everyone, and is the intent of the patent system.
How do you know? They haven't heard a case since 1981. Half the justices we have now weren't present back then. They may have different opinions than you expect.
The software industry was very immature, when it was growing by leaps and bounds without patent protection. It's much more of a competitive and global market, and it may be that people need more protection for their research money than they needed when there were fewer dangers from competitors.
You say there's no reason to think it will hurt the industry. How do you know? I can think of a reason immediately. If I were investing in research, I might not choose to hire computer scientists for research in an area for which I couldn't have patent protection. I might redirect that money to other areas (e.g. pharmaceuticals) where I would stand a better chance of profiting. This could slow progress in our industry.
You seem to place a great emphasis on the difference between discovery and invention, assuming that the "maths already existed" even before you discovered it. I would argue this is simply wrong. If nobody has ever "discovered" it, then it is new, and potentially patentable, provided it meets the novelty requirements. The question to be decided by the patent examiner is could this method have been "discovered" by someone else without making the investment that the patent applicant made? If the answer is yes, then you don't get a patent, sorry. If no, then we should reward you for making your investment, so that all the world will benefit from your work. You get a temporary monetary incentive from anybody who wants to use your concept. If it's truly novel and solves an important problem, then there may be a way to make money off it, which could benefit everybody.
How can this be true? What about the RSA patents that recently expired? Or, the patent on LZ compression that led to the GIF format issues? Algorithms have been and are patentable if they meet the requirements of novelty and prior art. And I would argue they should be. It takes a lot of research effort to come up with these methods, and after investing in so much research, why shouldn't you be able to license it as broadly as possible in the market, so that everyone in the world can benefit? If your only protection were copyright/trade secrets, then reverse engineering would kill your ability to profit, and would kill many other people's ability to use it, because you wouldn't publish your research. Instead, it would foster an environment of secrecy, the unintended consequences of which would be that new and unimagined uses for your technique would not materialize, and we couldn't benefit from collaboration the way we do.
Academics may publish openly, but ultimately somebody has to pay for their research time. And we want more of this kind of investment than just what can be afforded by charitable contributions, educators, and government. Basic research funded by capitalistic enterprises is responsible for many many innovations that would not have been possible without a patent system.
I disagree. The biggest problem with software patents is that the USPTO cannot seem to make a fair measure of novelty and obviousness. If they could, it might be an okay system. For example, if some organization invests several years of research to produce a new algorithm that does something wonderful, say a heuristic for the bin packing problem that's an order of magnitude faster than its next competitor, then why shouldn't they get the benefit of licensing that technology to everybody in the world who could benefit from it? We need to encourage this kind of investment in creating new technologies.
The problem isn't in the law, it's in the implementation of it. The USPTO needs to alter its methods of reviewing patent applications. One solution might be to solicit feedback from the community in an open forum. Feedback could include arguments why it's obvious as well as references to prior art.
Like many posts, I recommend you find a way to alter your career path. Many many jobs out there need someone who understands programming but does not do much of it. One person mentioned management. You could consider this path: learn project management. My ideal job (because I'm such a ADHD scatter-brain) is one that has a lot of variety. My current job is close to it, too.
I work in a lab with people who are building a system. There are scientists, engineers, business people, clients, etc. Each has needs, and I try to be available to interface with many as I design and write the software. Well, the software is a little behind schedule. So? The other people have gotten help from me, so the overall project is doing better than it would. And everyone comes to me when they need to brainstorm a solution to their problem, whether it's related to science, engineering, marketing, logistics. I get a lot of breaks from the code, and have a nice bit of variety in my day.
Pick a direction of interest for you other than computers. Look for (or work to create) a position that allows you to do more of the new interest and less programming. Ultimately, some hot shot will come along with more and up-to-date skills than you have in computers. But competition will be scarce for someone with both of your specialties. Perhaps, down the road, you'll grow tired of this one. So, pick a related third. Specialize and diversify. You may find that this leads you into projects at work that are so stimulating to you that you have a hard time leaving the office! With a little focus and determination, you can get there.
Yah, I'd love to pick up and use O'Caml, but the biggest barrier is interfacing with other libraries. I'm an embedded programmer, and often, I have to get control of a device with a canned software library. Correct me if I'm wrong, but the current state of O'Caml is that interfacing to these kinds of 3rd party libraries (using standard DLL or shared library interfaces) is almost unworkable. The thing that makes Python and Perl (and Java, too) shine as languages is their rich access to huge libraries. Without that, they'd never be the successes they are.
The researchers at INRIA have their noses in the air and refuse to understand that without good native interface support, their wonderful invention is destined to languish in obscurity, while other "lesser" languages experience greater success.
I'll keep looking at the community of O'Caml to see if they wake up and build this in. Until then, I'll be destined to stick with those that have the complete package.
Actually, Green Hills market share is one of the fastest growing in the embedded software industry. They are one of the few consistently profitable embedded tools vendors.
Your example does not contradict my basic point. These Professional Inventors do actually seek to have their ideas marketed, whether they do it themselves or sell it to someone who will. The scumbags in question do not do so. They lie in wait for some unsuspecting hardworking Joe to actually succeed using their (obvious, but) patented idea. (This, IMHO, is the biggest problem with software patents, testing obviousness.) Then, they turn loose the ravenous wolves.
As a previous post mentioned, it is very difficult to overturn an issued patent, especially on the obviousness criteria. About the only way is to prove prior art, which is not always doable, and even when you do, you are not guaranteed success. A patent is a powerful weapon. Even the IBM's and Microsoft's of the world are not immune, because their only tool to fight back (cross-licensing) does not work with these kinds of companies. The BSVs do not actually do anything for which they'd need a cross-license. All they do is hold a patent portfolio and sue the pants off anybody unlucky enough to make money using one of their obvious ideas. They are basically using weaknesses in the legal system to exploit others, which is very troubling to me.
We want those who succeed to be rewarded, and we want to punish those who seek to do nothing but exploit the weaknesses of others. The law or process needs a fundamental change.
One of my former company's customers in California had a site source license from us. They were able to port our application (>500,000 lines C/Informix) from SCO to Linux in less than a week. Of course, our application ran on more Unix's than SCO. But, I only found out about it afterwards, because we had a support agreement, and they were having one problem they couldn't resolve. After a half-day, I resolved that one, and they've moved entirely to Linux. It's a network of 10-20 computers running in excess of 1000 remote devices on a WAN.
The big benefit for them was to be able to use fewer servers (they had been planning 50 cpu's), and the application ran operationally the same as before, only faster.
Slashdot Mirror
Sorry, but based on your sig, I would suggest you don't have a lot of credibility in assessing the President's motivations in any impartial way.
Had Mr. Bush admitted some change of heart or mind on any policy or issue of national importance, you would be pointing him out as hypocritical for calling Kerry a flip-flopper. You would be saying he talks out of both sides of his mouth. You'd be saying how the mighty have fallen. And whatever the issue was, it would be broadcast up and down the media every day between now and election day.
It was a no-win question, and all 3 candidates chose not to answer it. Only one candidate didn't try to hide behind double-speak.
Excellent commentary. Unfortunately, blocked by /. indentation scheme...
...you can't believe that partisans will populate the entire judicial system.
They don't have to. With the world speeding up as it does, taking action to stop a terrorist may be a matter of hours, not days or weeks. Thus, the Patriot act has limited the oversight of the process to streamline it. One overzealous judge could in fact slow down the process such that we have another 9-11-type calamity on our hands.
This is why we sometimes cut corners protecting civil rights when there is a clear and present danger. I believe there still is one.
Thanks for your thoughtful comment.
The preamble to our Constitution lists the reasons for its existence:
- Form a more perfect Union.
- Establish justice.
- Ensure domestic tranquility.
- Provide for the common defense.
- Promote the general welfare.
- Secure the blessings of liberty.
Each of these addresses a different area in which our Federal Government is to provide a measure of security for the People.
Union: Secure from the weaknesses of widely disparate and disputing local governments.
Justice: Secure from crime and wrongdoing.
Tranquility: Secure from riots and anarchy.
Defense: Secure from invasion and foreign threats.
Welfare: Protected against financial ruin, economic disaster. (Note it says promote, not provide.)
Liberty: Secure from Government control:Tyrany.
Yes, I do have to eventually trust my neighbor. But there are many things the government can do to encourage me and my neighbor to be trustworthy.
The basic "rights" are summed up in the Declaration: Life, Liberty, Pursuit of Happiness. No, they can't be guaranteed, but they can be secured.
Vote!
And, even better, run for office. We could use a few rocket scientists in Congress.
There is a fine line to be found between protecting the rights of individuals and protecting the right of the People to be secure. The Patriot Act sought to define the line, giving the Executive more power to track these financial transactions, without scrutiny of the individual being investigated, and with limited oversight.
We need some kind of oversight, because the Executive may abuse the power. Not every executive will be as trustworthy as others in regard to protecting the rights of individuals.
One thing to consider, however, is that with judicial oversight, you can have another form of tyrany, where an overzealous judge prevents an Executive from doing his job to protect the People. We only have an appeals process for this, which hopefully results in a well-reasoned balance of rights. However, as the judicial confirmation process becomes more and more politicized, you can expect more and more partisans being placed in lifetime-tenured posts.
No judge is ever going to rule less power for the Judicial branch. I, for one, do not welcome our judicial overlords. Lex Rex.
Big deal. DMCA only applies to USA. Crackers in other parts of the world will break it and publish their results. Then someone else will write a nice application using the results, and people within the USA will download it and use it. Information wants to be free, and until the corporate entities come to this realization, they will be fighting a losing battle. DMCA cannot succeed in the long run to achieve its goals. Not without some kind of 1984-style monitoring, and americans are too libertarian / civil-liberties minded to accept this.
Um, No.
The biggest problem I have with the STL is that many debuggers don't have nice support for it. Thus, it's very difficult to observe what's stored in that nice vector or tree, when you hit the breakpoint. A few newer ones are adding automatic display of STL structures, which will be very nice, but for now, I'm stuck with VC++6, which is horrendous WRT these kinds of data structures.
I'd certainly welcome any ideas people have on how they actually debug code using the STL.
And, if your application doesn't happen to be written in this "verifiable subset", what's your choice? Trust it?
I assume all the default CLR subsystems are defaulted to reject non-safe code?
I wonder how many other OS's MS is supporting with their CLR? If I want to run under Linux, what's the likelihood that an MS produced bytecoded program will run there unmodified and untested. (Not that Java is there yet, but at least that's their goal. MS's goal doesn't seem to be WORA. It seems to be Add More Nonstandard Stuff, so they have to use our system. This is an understandable goal-- and I support their right to do it-- but I don't have to be drawn into the trap, either.)
Pardon me for not trusting our Redmond friends, it's just that they have a track record of abuse. Maybe they have a new attitude of philanthropy I haven't noticed yet.
Did you RTFA? .NET is fundamentally insecure because it decided to support the insecure features of C/C++. I'm not willing to get bit by these weaknesses years down the road when the script kiddies start taking advantage of them.
If the discovery is truly worth something in the real world, then it is worth licensing the patent-holder's property to achieve your goal. If not, then it is not.
Will giving you this patent promote progress?
That's awfully broad criteria for a lowly patent clerk to be using. I think they need a bit more guidance in how to achieve that.
I disagree with your statement about whether you would have gone for it anyway. If you would have gone for it and would have shared it once you had it, then maybe. Problem is, in the real world, that's an awful lot of hypotheticals to figure out, and is very much a matter of opinion. The job of the patent clerk has to be made easier. It's hard enough to decide whether something is novel. Seeing whether someone put in investment is easy. Perhaps we need people to detail the steps they went about in the creation of their idea, so the patent clerk could determine both the amount of effort and the likelihood of this being invented by another person without the same level of investment.
Perhaps I wasn't clear. Here's an example. Say you're a small-time inventor. You work out of your garage. You spend years, mortgage your inheritance, and you come up with an algorithm that helps producers deliver smaller/faster/cheaper. You want to see this innovation used over the world, but you also need to be reimbursed for your years of investment. You could create a software implementation that perhaps gets you some of that back. Problem is, today you have an extremely competitive market. You're not making your own market. You're competing against Microsoft, Peoplesoft, NA, and Pirhanasoft. Copyright and trade secret doesn't get you there, because with 10 minutes in a decompiler, they can determine your method and use it in their existing products, to their existing customer base. You get squashed like a bug. The only solution is patent protection for your algorithm. Without this protection, you are hesitant to mortgage your inheritance and spend years of perspiration to create your algorithm. So, since it is novel, it never gets created, and the world does not get the benefit. This is a chilling effect on the market. We want and need people to make those kinds of investments.
Now, the current system clearly is flawed. But the cure cannot be to eliminate the system for all software. Better would be to make sure that people only get patents who have invested, that it is truly novel -- nobody else would have come up with it without making the substantial investment. The important thing is that we reward the making of that investment, when it produces something marketable. Because, many times people will invest and fail. This is already a negative feedback. There must be a big positive payoff if you actually succeed.
Now, when the market was immature so many years ago, it may be seen that patent protection was not necessary. However, today, with such heavy competition and monopolization, it is much more needed. We can't just return to the good old days, when you could create something innovative, and expect copyright to make you rich. Copyright wasn't designed for this kind of protection, and ultimately, it fails to deliver what is needed. Trade secret coupled with anti-circumvention law may help, but what we really need is to allow maximum openness in promoting our ideas, so that those who are good at marketing can do the marketing, and those that are good at research and theory can work in their lab and be paid to keep innovating. This is what's ultimately best for everyone, and is the intent of the patent system.
How do you know? They haven't heard a case since 1981. Half the justices we have now weren't present back then. They may have different opinions than you expect.
The software industry was very immature, when it was growing by leaps and bounds without patent protection. It's much more of a competitive and global market, and it may be that people need more protection for their research money than they needed when there were fewer dangers from competitors.
You say there's no reason to think it will hurt the industry. How do you know? I can think of a reason immediately. If I were investing in research, I might not choose to hire computer scientists for research in an area for which I couldn't have patent protection. I might redirect that money to other areas (e.g. pharmaceuticals) where I would stand a better chance of profiting. This could slow progress in our industry.
You seem to place a great emphasis on the difference between discovery and invention, assuming that the "maths already existed" even before you discovered it. I would argue this is simply wrong. If nobody has ever "discovered" it, then it is new, and potentially patentable, provided it meets the novelty requirements. The question to be decided by the patent examiner is could this method have been "discovered" by someone else without making the investment that the patent applicant made? If the answer is yes, then you don't get a patent, sorry. If no, then we should reward you for making your investment, so that all the world will benefit from your work. You get a temporary monetary incentive from anybody who wants to use your concept. If it's truly novel and solves an important problem, then there may be a way to make money off it, which could benefit everybody.
How can this be true? What about the RSA patents that recently expired? Or, the patent on LZ compression that led to the GIF format issues? Algorithms have been and are patentable if they meet the requirements of novelty and prior art. And I would argue they should be. It takes a lot of research effort to come up with these methods, and after investing in so much research, why shouldn't you be able to license it as broadly as possible in the market, so that everyone in the world can benefit? If your only protection were copyright/trade secrets, then reverse engineering would kill your ability to profit, and would kill many other people's ability to use it, because you wouldn't publish your research. Instead, it would foster an environment of secrecy, the unintended consequences of which would be that new and unimagined uses for your technique would not materialize, and we couldn't benefit from collaboration the way we do.
Academics may publish openly, but ultimately somebody has to pay for their research time. And we want more of this kind of investment than just what can be afforded by charitable contributions, educators, and government. Basic research funded by capitalistic enterprises is responsible for many many innovations that would not have been possible without a patent system.
I disagree. The biggest problem with software patents is that the USPTO cannot seem to make a fair measure of novelty and obviousness. If they could, it might be an okay system. For example, if some organization invests several years of research to produce a new algorithm that does something wonderful, say a heuristic for the bin packing problem that's an order of magnitude faster than its next competitor, then why shouldn't they get the benefit of licensing that technology to everybody in the world who could benefit from it? We need to encourage this kind of investment in creating new technologies.
The problem isn't in the law, it's in the implementation of it. The USPTO needs to alter its methods of reviewing patent applications. One solution might be to solicit feedback from the community in an open forum. Feedback could include arguments why it's obvious as well as references to prior art.
Like many posts, I recommend you find a way to alter your career path. Many many jobs out there need someone who understands programming but does not do much of it. One person mentioned management. You could consider this path: learn project management. My ideal job (because I'm such a ADHD scatter-brain) is one that has a lot of variety. My current job is close to it, too.
I work in a lab with people who are building a system. There are scientists, engineers, business people, clients, etc. Each has needs, and I try to be available to interface with many as I design and write the software. Well, the software is a little behind schedule. So? The other people have gotten help from me, so the overall project is doing better than it would. And everyone comes to me when they need to brainstorm a solution to their problem, whether it's related to science, engineering, marketing, logistics. I get a lot of breaks from the code, and have a nice bit of variety in my day.
Pick a direction of interest for you other than computers. Look for (or work to create) a position that allows you to do more of the new interest and less programming. Ultimately, some hot shot will come along with more and up-to-date skills than you have in computers. But competition will be scarce for someone with both of your specialties. Perhaps, down the road, you'll grow tired of this one. So, pick a related third. Specialize and diversify. You may find that this leads you into projects at work that are so stimulating to you that you have a hard time leaving the office! With a little focus and determination, you can get there.
That's my 2 cents.
Thanks for the thoughtful reply. Sorry I was vague.
Yah, I'd love to pick up and use O'Caml, but the biggest barrier is interfacing with other libraries. I'm an embedded programmer, and often, I have to get control of a device with a canned software library. Correct me if I'm wrong, but the current state of O'Caml is that interfacing to these kinds of 3rd party libraries (using standard DLL or shared library interfaces) is almost unworkable. The thing that makes Python and Perl (and Java, too) shine as languages is their rich access to huge libraries. Without that, they'd never be the successes they are.
The researchers at INRIA have their noses in the air and refuse to understand that without good native interface support, their wonderful invention is destined to languish in obscurity, while other "lesser" languages experience greater success.
I'll keep looking at the community of O'Caml to see if they wake up and build this in. Until then, I'll be destined to stick with those that have the complete package.
Actually, Green Hills market share is one of the fastest growing in the embedded software industry. They are one of the few consistently profitable embedded tools vendors.
MOD PARENT UP!
You said what I intended so much better. Thanks!
Your example does not contradict my basic point. These Professional Inventors do actually seek to have their ideas marketed, whether they do it themselves or sell it to someone who will. The scumbags in question do not do so. They lie in wait for some unsuspecting hardworking Joe to actually succeed using their (obvious, but) patented idea. (This, IMHO, is the biggest problem with software patents, testing obviousness.) Then, they turn loose the ravenous wolves.
As a previous post mentioned, it is very difficult to overturn an issued patent, especially on the obviousness criteria. About the only way is to prove prior art, which is not always doable, and even when you do, you are not guaranteed success. A patent is a powerful weapon. Even the IBM's and Microsoft's of the world are not immune, because their only tool to fight back (cross-licensing) does not work with these kinds of companies. The BSVs do not actually do anything for which they'd need a cross-license. All they do is hold a patent portfolio and sue the pants off anybody unlucky enough to make money using one of their obvious ideas. They are basically using weaknesses in the legal system to exploit others, which is very troubling to me.
We want those who succeed to be rewarded, and we want to punish those who seek to do nothing but exploit the weaknesses of others. The law or process needs a fundamental change.
One of my former company's customers in California had a site source license from us. They were able to port our application (>500,000 lines C/Informix) from SCO to Linux in less than a week. Of course, our application ran on more Unix's than SCO. But, I only found out about it afterwards, because we had a support agreement, and they were having one problem they couldn't resolve. After a half-day, I resolved that one, and they've moved entirely to Linux. It's a network of 10-20 computers running in excess of 1000 remote devices on a WAN.
The big benefit for them was to be able to use fewer servers (they had been planning 50 cpu's), and the application ran operationally the same as before, only faster.
Oh, and they're running on IBM hardware now, too.