That's really cynical. I did speak to Dan Lyons, and didn't manage to change his opinion. Indeed, I don't always manage to change the opinion of folks here. Go back and read my comments to this story.
No, I am not just being among extortionists. We have a seriously broken patent system, and there is no doubt whatsoever that anything as big as the kernel would have 300 or so patents that might apply to it. OSRM is a band-aid. In the two press tours I have done for them so far, my main point has been that the system needs fixing. Please also see my CNET editorial about them in which I explained that most people don't need the insurance.
Regarding how much I expect to make, I have helped create several companies so far without making anything. Maybe someday Progeny will be worth something, I don't know. So, I am not expecting anything. Nor do I see that my rep is suffering among those who take the time to examine the issue. The main reason I am participating in this is so that there will be a force able to defend Linux while being independent of hardware vendors and distributions, with their proprietary agendas. I am also working hard to fix the system, but politics takes years or decades to change. What else do you suggest?
Take a good look at the pharmecutical industry and the cost of drugs in the third world, and you will be convinced that patents do kill people. Doctors Without Borders has a good case on this, search them on the web.
I am on the board of this thing and would not allow them to goad patent holders into suing. Indeed, the entire thing is calculated to prevent people from suing, because a big insurance company with a $100M fund will defend against suits instead of a little customer.
I don't think you could really get from A to B in court. It would only be necessary to distinguish code as speech from its application in a device. No doubt an attorney could explain this better.
The most interesting part of OSRM is that they will maintain a standing legal defense force to protect Linux and GNU. Essentially, they pool the risk of deep-pockets defendants to support that force. Simply by existing and having customers, OSRM tells software patent holders that we will defend Open Source software from software patents asserted against it. The effect is to reduce nuisance suits.
But the patent problem is really scary, not just for Free Software but for any small-to-medium sized software manufacturer. It should not be discounted.
The least draconian solution would be to tremendously improve patent quality. This means good tests before patents are granted regarding triviality - give the problem to other engineers and they come up with the right solution right away - and prior art - probably requires an easy challenge process -, and preciseness of claims. We have a serious vague-claim problem right now, as people write vague claims so that they will be maximally applicable rather than specifically applicable. Then shorten the term to fit the field.
But I'd be just as happy to do away with them.
You are wrong about corporations paying more taxes. The burden is very slanted toward individuals, and within individuals toward lower income.
Me, I'd not allow corporations to be immortal, and I'd set a minimum tax rate for them and make sure it was enforced.
Lying on your patent application is perjury. It's a federal crime, and you can do serious jail time. But nobody does. It might be possible to show a federal prosecutor that some of the more blatant patents were filed even though the filer knew they were not inventions, and that the filer should be prosecuted.
Inconvenient is an understatement. We are moving toward a point beyond which only the largest companies will be able to engage in software development. Forget about individuals doing it when the cost of defending a single patent suit is about $3 Million (American IP Law Association estimate).
So, I'd suggest that "discriminatory" is a lot more accurate than "inconvenient".
And yes, hardware is software these days. Which means that all would better be protected with copyright. Applying both patent and copyright to the same material is too much.
First, going for our own patents in the Free Software community doesn't really help unless we have a huge legal fund behind us to 1) prosecute others and 2) defend ourselves from their patents.
Second, you should think through whether or not algorithms are mathematical in nature, and whether mathematics is discovered or invented.
I think the point about software being "mercurial" is vague, but what he was probably thinking about it that it's a medium better protected by copyright. Now having both patent and copyright apply to it is indeed strange.
Software patenting has been driven by court cases in the U.S. brought by a number of companies that wanted to be able to patent software and business methods. Do a web search for State Street Decision.
WIPO has been driven by the U.S. (reacting to its corporations) to get the same bad software patent system going elsewhere.
If you want to blame Clinton for some intellectual property matter connected to WIPO, digital rights management and associated treaties requiring anti-circumvention law would be more accurate.
A long time ago, at Pixar, I got an ARPA grant to work on an image-processing application for the feature film industry. The purpose of the grant was economic and military at the same time. I was to help create a market for multiprocessor computers (not really supercomputers) so that there would be U.S. manufacturers of them if when/if the Army needed them for military purposes. This is what often gets called corporate welfare, although I could see the defense purpose was valid. I don't know if ARPA still does this sort of grant. To do one would require an application that is interesting to more than just the folks on your list. And these days visual effects is much more of a solved problem.
One of the nice things about clusters is that they encourage people to consider how to decompose a problem so that it can work without a large high-speed shared data memory. Some of the older supercomputers were important because scientists hadn't done this work because there wasn't the economic incentive back then. Now there is one.
So, what tasks still require a high-speed shared data memory? Answer that, and you'll understand where you can still sell a supercomputer.
Well, these days I hear SABRE uses MySQL. Are you really sure that TRAMS was the only application where Interbase was ever used? Would you know if it was used in SABRE, for example, in the past?
It was a lot easier to make money in embedded systems when embedded operating systems were magic available only from a few.
The customer percieves that the value of his investment in the development of an embedded product depends on the embedded OS manufacturer remaining available to support the OS, and bringing the OS to new hardware, for the life of the embedded product and its successors. Embedded operating systems other than CE and Linux aren't getting the customer trust.
If you had to switch from your cash cow to a new product, your figures would suffer too. Expect this to shake out some of the embedded vendors.
Their CEO would not be visibly spreading FUD that is so easily discounted if there was not something very seriously wrong there. My suspicion is that they have seen a big dip in new projects that is not reflected in their sales figures yet.
Green Hills is a failing company that is seeing its market go to Open Source. In contrast, Wind River, which is in the same market with the same customers, embraces Linux.
The fact is that Green Hills products are no more secure, and may well be less secure, because they don't have the "many eyes" looking at their source code. We've had trojan horse attempts in Open Source software. They get caught quickly. But even if the source is disclosed, nobody outside of their tiny company has an incentive to do productive work on the internals of a Green Hills operating system in the way that people who modify GNU/Linux do. And security audits by such a small company can't catch everything.
The best example of this has been the Borland Interbase database. This was used for airline reservations, and had a trojan horse buried in it for 6 to 9 years while it was a proprietary product. The door could have been found by anyone who did an ASCII dump of the product, but those who did kept it secret, and probably took a lot of free flights. An Open Source coder found the door some months after the database went Open Source, and had an incentive to report it - at that point he was one of the people doing productive work on the database and only wanted it to work better and more securely.
This "black hats" (people who are motivated for bad purposes) vs. "white hats" (good purpose) phenomenon is important to consider when you evaluate the security of Open Source. Generally the only people who would look for vulnerabilities in proprietary software, outside of its manufacturer, are looking to exploit them! This is hardly the case with Open Source.
Slashdot Mirror
Bruce
Regarding how much I expect to make, I have helped create several companies so far without making anything. Maybe someday Progeny will be worth something, I don't know. So, I am not expecting anything. Nor do I see that my rep is suffering among those who take the time to examine the issue. The main reason I am participating in this is so that there will be a force able to defend Linux while being independent of hardware vendors and distributions, with their proprietary agendas. I am also working hard to fix the system, but politics takes years or decades to change. What else do you suggest?
Thanks
Bruce
Bruce
Bruce
Bruce
Bruce
Bruce
Bruce
But the patent problem is really scary, not just for Free Software but for any small-to-medium sized software manufacturer. It should not be discounted.
Bruce
But I'd be just as happy to do away with them.
You are wrong about corporations paying more taxes. The burden is very slanted toward individuals, and within individuals toward lower income.
Me, I'd not allow corporations to be immortal, and I'd set a minimum tax rate for them and make sure it was enforced.
Bruce
Bruce
Bruce
So, I'd suggest that "discriminatory" is a lot more accurate than "inconvenient".
And yes, hardware is software these days. Which means that all would better be protected with copyright. Applying both patent and copyright to the same material is too much.
Thanks
Bruce
First, going for our own patents in the Free Software community doesn't really help unless we have a huge legal fund behind us to 1) prosecute others and 2) defend ourselves from their patents.
Second, you should think through whether or not algorithms are mathematical in nature, and whether mathematics is discovered or invented.
Thanks
Bruce
Bruce
WIPO has been driven by the U.S. (reacting to its corporations) to get the same bad software patent system going elsewhere.
If you want to blame Clinton for some intellectual property matter connected to WIPO, digital rights management and associated treaties requiring anti-circumvention law would be more accurate.
Bruce
A long time ago, at Pixar, I got an ARPA grant to work on an image-processing application for the feature film industry. The purpose of the grant was economic and military at the same time. I was to help create a market for multiprocessor computers (not really supercomputers) so that there would be U.S. manufacturers of them if when/if the Army needed them for military purposes. This is what often gets called corporate welfare, although I could see the defense purpose was valid. I don't know if ARPA still does this sort of grant. To do one would require an application that is interesting to more than just the folks on your list. And these days visual effects is much more of a solved problem.
Thanks
Bruce
So, what tasks still require a high-speed shared data memory? Answer that, and you'll understand where you can still sell a supercomputer.
Bruce
Thanks
Bruce
Bruce
The customer percieves that the value of his investment in the development of an embedded product depends on the embedded OS manufacturer remaining available to support the OS, and bringing the OS to new hardware, for the life of the embedded product and its successors. Embedded operating systems other than CE and Linux aren't getting the customer trust.
If you had to switch from your cash cow to a new product, your figures would suffer too. Expect this to shake out some of the embedded vendors.
Thanks
Bruce
Bruce
Thanks
Bruce
Bruce
The fact is that Green Hills products are no more secure, and may well be less secure, because they don't have the "many eyes" looking at their source code. We've had trojan horse attempts in Open Source software. They get caught quickly. But even if the source is disclosed, nobody outside of their tiny company has an incentive to do productive work on the internals of a Green Hills operating system in the way that people who modify GNU/Linux do. And security audits by such a small company can't catch everything.
The best example of this has been the Borland Interbase database. This was used for airline reservations, and had a trojan horse buried in it for 6 to 9 years while it was a proprietary product. The door could have been found by anyone who did an ASCII dump of the product, but those who did kept it secret, and probably took a lot of free flights. An Open Source coder found the door some months after the database went Open Source, and had an incentive to report it - at that point he was one of the people doing productive work on the database and only wanted it to work better and more securely.
This "black hats" (people who are motivated for bad purposes) vs. "white hats" (good purpose) phenomenon is important to consider when you evaluate the security of Open Source. Generally the only people who would look for vulnerabilities in proprietary software, outside of its manufacturer, are looking to exploit them! This is hardly the case with Open Source.
Thanks
Bruce