Having recently (2000) installed a swimming pool, I spent a fair amount of time reading up on pool chemistry. One of the articles went into relative risks, including the relative risks of the use of chlorine in pools and water supplies. To put it simply, while there is some cancer risk to chlorine, but it's way down in the mud, especially compared to the bacterial risks of not using it. The cancer risks are in the one-in-a-million range, or less, while the bacterial risks are in the tens-in-a-hundred range.
There are ranges of chlorine for normal disinfection and superchlorination to get rid of eye-stinging chloramines. The "threshold" for trihalides is normally above either of these dosages, so with a little care you can avoid the bad stuff.
But clinical trials cost money, and if the patents on DCA have expired, then there's little opportunity to rake in big bucks on it, so who is going to fund clinical trials?
I wonder how many potential medications there are out there in the same boat. IMHO it would be interesting to begin mining "folk remedies" with clinical trials, to see how many have a grain of truth behind them, and how many are pure bunk. Theoretically, the reason for medical research is to improve the quality of lives, but we've introduced this pesky little profit motive right into the middle of it. Not that there's anything wrong with profit in and of itself, but when it contradicts the essential purpose, something in the business model needs some tuning.
The history of computing, up until Vista, has been mostly trying to make computer *work*, and work better. We're still not really there yet, because if you were to go up to anyone on the street and ask, "Does your computer work well enough yet, can you actually do what you want with it?" the answer would generally be "No," or at best, a qualified "Yes."
For the first time in a widely-deployed software product, Vista has had extensive work done to make it *not work*. That's really what DRM and SPP are, parts of the system that at "appropriate times" make the system not work, or work in a degraded fashion. Argue all you want about the philosophy of DRM and SPP, and even if you agree with those precepts, you must also agree that they are present in order to degrade the machine's operation.
Now if we can't really make a computer "work right" in the first place, what sort of hubris is at work to make us think that we can add software to make it "quit working right," and have that work right, too?
Too much money being made on petrochemicals and other aspects of the way things are being done now, which are taking a lion's share of the blame for global warming?
You immediately rule out Legos, and that may not be appropriate.
I would agree that depending on the class level, the RCS programming may be overly simplistic. But by no means is RCS the only way to program the Lego Mindstorms. A quick peek at Google and you'll see things like NQC, a Lego Mindstorm with Linux Mini-HowTo, and a whole wealth of free stuff that keeps the Mindstorm relevant to a more experienced class level.
The other nice thing about Mindstorms is that you can do things physical, not just things that you can see (LEDs, meters) or hear, and it's well integrated so you're fighting the task at hand, and not fighting making an ad-hoc collection of parts work together. (Though there is definitely value in the latter, it just depends on where you want to put the emphasis.)
Heck, you can even find hits searching on "mindstorm" and "gentoo". Emerge THAT!
I didn't know that, and I wouldn't be surprised if a significant number of the people in the wireless access business didn't either, especially their legal departments who are probably more concerned with IP than the FCC. For a corporate lawyer, the simple, easy, safe answer is always, "No!"
Only in appearance. From a fearful perspective, open source is just begging to be tweaked, of course irresponsibly into some illegal operating mode. That of course neglects the far more common act of simple driver maintenance.
As for the binary blob, if someone's taking the effort to pick it apart, I'd say that it's far more likely to be someone who *wants* to do something illegal, like increase the power.
I would guess the relative order of probability of actions would be: 1 - good maintenance of open source driver 2 - evil tweaking of binary blob 3 - evil tweaking of open source 4 - good hacks of binary blob
For the card supplier, they'd like to pretend that "1" doesn't have to exist. They can't really do squat about "2", nor can they be held responsible for it. But they may feel like they have complicity in "3", and some attorneys may exaggerate that fear.
As for why I have "2" more probable than "3", I'll go back to the Ham radio analogy. Open source developers are more likely to understand the limits, the FCC requirements, and as demonstrated by working with open source, a tendency to play by the rules.
If you buy and operate a ham radio, you have a license, and are obliged to know the difference between Good and Evil, and are sufficiently informed to be "responsible."
Network cards are "appliances", and therefore built for irresponsible idiots, who may download a "super-hot driver" and install it without understanding that they're breaking the law, or perhaps understanding that it's illegal, not understanding why, or caring. In any event, they don't pass the same muster as radio hams for being sufficiently informed.
For the recent times, I'd say that the "Linux-biased anti-Windows Slashdotter" is about as common as "The Liberal Media." In other words, the airwaves are full of conservative talking heads who cry out against, "The Liberal Media," but by and large those conservative talking heads are getting most of the airtime. Heck, pretty much until Katrina, the Bush Administration was getting an easy pass by the Media. Liberal Media, my (anatomical part).
Seems like recently, on Slashdot there's a decent amount of discussion about Linux shortcomings, and real (sometimes heated) discussion on both sides. But much of the time when the discussion becomes critical toward Windows, someone decries the "Linux-biased anti-Windows Slashdotter," kind of like Godwin's Law.
I'd suggest modifying your list, a little. As others have said, some of your options simply can't happen, and you missed one.
a) no driver for your hardware. b) GPL shim wrapped around stock Windows driver/blob (ndiswrapper) c) GPL shim wrapped around closed-source Linux driver/blob (nVidia, ATI) d) obfuscated code in kernel tree (kernel devs under NDA) e) uncommented code in kernel tree (kernel devs under NDA)
My bar would be a "c" or above.
To counter what others have said, it may not be problematic for "the" kernel developer for a given NDA driver to depart. The corporate contact would already be in place, and the NDA could be transferred to another kernel developer.
I would further suggest that NDAs carry an expiration date, after which fully commented code would be placed in the kernel. While patent protection may last for nearly 2 decades, very few patents are truly relevant for that long. Furthermore in this field, for most patents, most salient aspects have been reverse-engineered or worked around in far less than 1 decade.
Along this line, for nVidia and ATI, we should bill this as a savings to them. For instance, nVidia now has 2 lines of legacy drivers, the merely old and the *really* old. Presumably there is nothing technologically damaging about the *really* old drivers, yet nVidia currently has to maintain them. If they gave those drivers to the community under GPL, they could claim "Linux support" for that hardware with no ongoing effort or cost on their part.
Might Open Firmware offer a way past the FCC issues with Open Source and wireless cards? The reason given for the blobs in the Atheros driver is that the card can be operated in a way that violates FCC requirements - power levels, etc. A middle ground might be doing all of the high-speed work in an open source driver, using Open Firmware for the FCC-critical setups.
I more wonder who really waited. If Korea commissioned the ActiveX control in 1998, when was it actually delivered an deployed? (I didn't see it in the blog.) I can't imagine so significant a project happening in less than a year, from the word, "Go!" By the same token, I wonder how long it was from SSL standardization to deployable implementations. Assuming SSL followed an open-source type development model, I can't imagine that there weren't deployable solutions available prior to standardization, part of the "proof", and just waiting for the formal stamp of approval.
Again, which became usably deployed first, SEED or SSL?
Oh, but if your tinfoil had is screwed on really tight, and made of really heavy-gauge tinfoil, the answer is simple.
It was well within our 1969 technology to send small unmanned landers to the moon, carrying retroreflectors. (Didn't every one of the supposed Apollo missions carry one?) That's how they got there, and how we're able to bounce lasers off of them.
Beyond that, even once we do have decent cameras in-place in 2008, remember that based on one of the referenced articles, NONE of the so-called Apollo artifacts is bigger than about 16 pixels. The same landers that actually landed the retroreflectors had several spring-loaded compartments. Each shot a small balloon + gas canister a few feet away, and a minute later those inflated into the correctly-sized "shadows".
But it gets better... By the time the tinfoil-hatters can go to the moon themselves and SEE the Apollo artifacts, they'll contend that they didn't get there in 1969-1972, but were in fact put there "only a few years ago", in whatever "now" we have lunar tourism.
For my part, I've imagined that someday the Apollo 11 landing site will have boardwalks so the original footprints remain undisturbed, along with some sort of lightweight transparent (polyethylene sheets?) shielding overhead to stop further micrometeorite erosion. The part I wonder is how they'll keep tourists on the boardwalk. (Unless we've done something stupid to ourselves prior to lunar tourism.)
A few days back, I was on your same point here: http://slashdot.org/comments.pl?sid=218160&cid=177 11614 and was told to "get a grip," presumably for trying to stick to such "unworldly" views that, even/especially for a corporation, there might be things more important than money.
I disagree. I think the US has gone off its rocker in pursuit of money. To put it in good old conservative religious terms, "The Great Whore of Babylon,' whoring after the almighty buck.
Take a concrete example... You've been married 1/5/10/25 years, and you want to take your wife out to dinner. Do you want to take her to the restaurant with the *best* chef, or the most profitable restaurant - the one with the chef who got into that business for the money?
Early in my career, I was visiting another semiconductor test center. The particular tester I was looking at had a "programmable knob" on the front, basically a shaft encoder whose function could be reassigned. As a joke, someone labeled it, "Yield". Of course on a tester, the only way you can deliberately crank up the "yield" is by cranking down the quality, assuming that the testing was correctly specified and implemented in the first place.
Same for that magical knob called, "Profit." There is no such knob, and pretending there is, is stupid and shortsighted. In today's world, it seems that CxOs look at the knob labeled "Cost" and mistakenly think it's really "Profit" with a backwards scale. It sorta works, but it ain't really so. While every successful company pursues revenue and profit, there must be a balance between that and the function being delivered. From MY perspective, Ford and GM are in business to make cars available to me, and I don't give SPIT about profits to stockholders, since I'm not one of them. Perhaps they're just a little too focused on stockholder value, and not enough on delivering cars that people want, and that's why they're sliding downward from #1 and #2.
IMHO, being too focused on making money may distract you from what the company is really supposed to be doing, and may not be good in the long term.
I would think it more likely that the "IP agreement" is not just a patent license. It's most likely a patent license plus some verilog netlist plus some driver code. It's that last part that likely precludes open source. While no doubt none of that drive code has actually survived into the final driver, it's all that "derived works" type of stuff. No doubt a cautious lawyer would consider the driver code to be partially derived from the licensed snippet. Beyond that, if they've licensed S3TC, it's not even clear that they can even document their texture compression hardware, since it is most likely considered a "derived work."
>If you ask me, the US has a long way to go before reaching the standards in terms of taste and healthiness (is that a word?) >that grocery food has set in the UK, Belgium, Netherlands, etc.
You seem to have the mistaken idea that "taste and healthiness" are goals of the US food industry.
Remember, the goals of industry in the US - any industry, are revenue and profit. Niceties such as "taste and healthiness" are merely means for achieving that goal. They may even even be simply costs to be reduced as much as possible, at least almost to the point where you'd buy a competitor's product, instead.
As long as sending email is "free" with no checks and balances, there will be spam. Yet the moment email becomes non-free and gains checks and balances, people are going to scream.
It's a philosophical problem, and requires philosophical underpinnings for an answer. Spam filters, RBLs, etc are just tactical responses. SPF, DKeys, hashcash and the like might be the beginnings of a true answer, but at the moment there's precious little being done at the philosophical level.
Slashdot Mirror
Last I heard, there were measure afoot to increase the number of hours a trucker can drive in a day - driven by the large trucking companies.
Having recently (2000) installed a swimming pool, I spent a fair amount of time reading up on pool chemistry. One of the articles went into relative risks, including the relative risks of the use of chlorine in pools and water supplies. To put it simply, while there is some cancer risk to chlorine, but it's way down in the mud, especially compared to the bacterial risks of not using it. The cancer risks are in the one-in-a-million range, or less, while the bacterial risks are in the tens-in-a-hundred range.
There are ranges of chlorine for normal disinfection and superchlorination to get rid of eye-stinging chloramines. The "threshold" for trihalides is normally above either of these dosages, so with a little care you can avoid the bad stuff.
But clinical trials cost money, and if the patents on DCA have expired, then there's little opportunity to rake in big bucks on it, so who is going to fund clinical trials?
I wonder how many potential medications there are out there in the same boat. IMHO it would be interesting to begin mining "folk remedies" with clinical trials, to see how many have a grain of truth behind them, and how many are pure bunk. Theoretically, the reason for medical research is to improve the quality of lives, but we've introduced this pesky little profit motive right into the middle of it. Not that there's anything wrong with profit in and of itself, but when it contradicts the essential purpose, something in the business model needs some tuning.
The history of computing, up until Vista, has been mostly trying to make computer *work*, and work better. We're still not really there yet, because if you were to go up to anyone on the street and ask, "Does your computer work well enough yet, can you actually do what you want with it?" the answer would generally be "No," or at best, a qualified "Yes."
For the first time in a widely-deployed software product, Vista has had extensive work done to make it *not work*. That's really what DRM and SPP are, parts of the system that at "appropriate times" make the system not work, or work in a degraded fashion. Argue all you want about the philosophy of DRM and SPP, and even if you agree with those precepts, you must also agree that they are present in order to degrade the machine's operation.
Now if we can't really make a computer "work right" in the first place, what sort of hubris is at work to make us think that we can add software to make it "quit working right," and have that work right, too?
Too much money being made on petrochemicals and other aspects of the way things are being done now, which are taking a lion's share of the blame for global warming?
Worse... real work kept me from using my mod points, and they just expired today. But your right, this thread deserves them.
You immediately rule out Legos, and that may not be appropriate.
I would agree that depending on the class level, the RCS programming may be overly simplistic. But by no means is RCS the only way to program the Lego Mindstorms. A quick peek at Google and you'll see things like NQC, a Lego Mindstorm with Linux Mini-HowTo, and a whole wealth of free stuff that keeps the Mindstorm relevant to a more experienced class level.
The other nice thing about Mindstorms is that you can do things physical, not just things that you can see (LEDs, meters) or hear, and it's well integrated so you're fighting the task at hand, and not fighting making an ad-hoc collection of parts work together. (Though there is definitely value in the latter, it just depends on where you want to put the emphasis.)
Heck, you can even find hits searching on "mindstorm" and "gentoo". Emerge THAT!
I didn't know that, and I wouldn't be surprised if a significant number of the people in the wireless access business didn't either, especially their legal departments who are probably more concerned with IP than the FCC. For a corporate lawyer, the simple, easy, safe answer is always, "No!"
Only in appearance. From a fearful perspective, open source is just begging to be tweaked, of course irresponsibly into some illegal operating mode. That of course neglects the far more common act of simple driver maintenance.
As for the binary blob, if someone's taking the effort to pick it apart, I'd say that it's far more likely to be someone who *wants* to do something illegal, like increase the power.
I would guess the relative order of probability of actions would be:
1 - good maintenance of open source driver
2 - evil tweaking of binary blob
3 - evil tweaking of open source
4 - good hacks of binary blob
For the card supplier, they'd like to pretend that "1" doesn't have to exist. They can't really do squat about "2", nor can they be held responsible for it. But they may feel like they have complicity in "3", and some attorneys may exaggerate that fear.
As for why I have "2" more probable than "3", I'll go back to the Ham radio analogy. Open source developers are more likely to understand the limits, the FCC requirements, and as demonstrated by working with open source, a tendency to play by the rules.
If you buy and operate a ham radio, you have a license, and are obliged to know the difference between Good and Evil, and are sufficiently informed to be "responsible."
Network cards are "appliances", and therefore built for irresponsible idiots, who may download a "super-hot driver" and install it without understanding that they're breaking the law, or perhaps understanding that it's illegal, not understanding why, or caring. In any event, they don't pass the same muster as radio hams for being sufficiently informed.
For the recent times, I'd say that the "Linux-biased anti-Windows Slashdotter" is about as common as "The Liberal Media." In other words, the airwaves are full of conservative talking heads who cry out against, "The Liberal Media," but by and large those conservative talking heads are getting most of the airtime. Heck, pretty much until Katrina, the Bush Administration was getting an easy pass by the Media. Liberal Media, my (anatomical part).
Seems like recently, on Slashdot there's a decent amount of discussion about Linux shortcomings, and real (sometimes heated) discussion on both sides. But much of the time when the discussion becomes critical toward Windows, someone decries the "Linux-biased anti-Windows Slashdotter," kind of like Godwin's Law.
I'd suggest modifying your list, a little. As others have said, some of your options simply can't happen, and you missed one.
a) no driver for your hardware.
b) GPL shim wrapped around stock Windows driver/blob (ndiswrapper)
c) GPL shim wrapped around closed-source Linux driver/blob (nVidia, ATI)
d) obfuscated code in kernel tree (kernel devs under NDA)
e) uncommented code in kernel tree (kernel devs under NDA)
My bar would be a "c" or above.
To counter what others have said, it may not be problematic for "the" kernel developer for a given NDA driver to depart. The corporate contact would already be in place, and the NDA could be transferred to another kernel developer.
I would further suggest that NDAs carry an expiration date, after which fully commented code would be placed in the kernel. While patent protection may last for nearly 2 decades, very few patents are truly relevant for that long. Furthermore in this field, for most patents, most salient aspects have been reverse-engineered or worked around in far less than 1 decade.
Along this line, for nVidia and ATI, we should bill this as a savings to them. For instance, nVidia now has 2 lines of legacy drivers, the merely old and the *really* old. Presumably there is nothing technologically damaging about the *really* old drivers, yet nVidia currently has to maintain them. If they gave those drivers to the community under GPL, they could claim "Linux support" for that hardware with no ongoing effort or cost on their part.
Might Open Firmware offer a way past the FCC issues with Open Source and wireless cards? The reason given for the blobs in the Atheros driver is that the card can be operated in a way that violates FCC requirements - power levels, etc. A middle ground might be doing all of the high-speed work in an open source driver, using Open Firmware for the FCC-critical setups.
I have discovered a truly remarkable proof^H^H^H^H^HVista DRM crack which this margin is too small to contain.
Isn't pixellated, it's merely overexposed to the point of showing about the same detail.
Much as I would like to see lunar tourism, preferably participate, I'm not betting on either.
I more wonder who really waited. If Korea commissioned the ActiveX control in 1998, when was it actually delivered an deployed? (I didn't see it in the blog.) I can't imagine so significant a project happening in less than a year, from the word, "Go!" By the same token, I wonder how long it was from SSL standardization to deployable implementations. Assuming SSL followed an open-source type development model, I can't imagine that there weren't deployable solutions available prior to standardization, part of the "proof", and just waiting for the formal stamp of approval.
Again, which became usably deployed first, SEED or SSL?
Oh, but if your tinfoil had is screwed on really tight, and made of really heavy-gauge tinfoil, the answer is simple.
It was well within our 1969 technology to send small unmanned landers to the moon, carrying retroreflectors. (Didn't every one of the supposed Apollo missions carry one?) That's how they got there, and how we're able to bounce lasers off of them.
Beyond that, even once we do have decent cameras in-place in 2008, remember that based on one of the referenced articles, NONE of the so-called Apollo artifacts is bigger than about 16 pixels. The same landers that actually landed the retroreflectors had several spring-loaded compartments. Each shot a small balloon + gas canister a few feet away, and a minute later those inflated into the correctly-sized "shadows".
But it gets better... By the time the tinfoil-hatters can go to the moon themselves and SEE the Apollo artifacts, they'll contend that they didn't get there in 1969-1972, but were in fact put there "only a few years ago", in whatever "now" we have lunar tourism.
For my part, I've imagined that someday the Apollo 11 landing site will have boardwalks so the original footprints remain undisturbed, along with some sort of lightweight transparent (polyethylene sheets?) shielding overhead to stop further micrometeorite erosion. The part I wonder is how they'll keep tourists on the boardwalk. (Unless we've done something stupid to ourselves prior to lunar tourism.)
You mean you don't keep a cat5 patch cable in your laptop case?
A few days back, I was on your same point here: http://slashdot.org/comments.pl?sid=218160&cid=177 11614 and was told to "get a grip," presumably for trying to stick to such "unworldly" views that, even/especially for a corporation, there might be things more important than money.
I disagree. I think the US has gone off its rocker in pursuit of money. To put it in good old conservative religious terms, "The Great Whore of Babylon,' whoring after the almighty buck.
Take a concrete example... You've been married 1/5/10/25 years, and you want to take your wife out to dinner. Do you want to take her to the restaurant with the *best* chef, or the most profitable restaurant - the one with the chef who got into that business for the money?
Early in my career, I was visiting another semiconductor test center. The particular tester I was looking at had a "programmable knob" on the front, basically a shaft encoder whose function could be reassigned. As a joke, someone labeled it, "Yield". Of course on a tester, the only way you can deliberately crank up the "yield" is by cranking down the quality, assuming that the testing was correctly specified and implemented in the first place.
Same for that magical knob called, "Profit." There is no such knob, and pretending there is, is stupid and shortsighted. In today's world, it seems that CxOs look at the knob labeled "Cost" and mistakenly think it's really "Profit" with a backwards scale. It sorta works, but it ain't really so. While every successful company pursues revenue and profit, there must be a balance between that and the function being delivered. From MY perspective, Ford and GM are in business to make cars available to me, and I don't give SPIT about profits to stockholders, since I'm not one of them. Perhaps they're just a little too focused on stockholder value, and not enough on delivering cars that people want, and that's why they're sliding downward from #1 and #2.
IMHO, being too focused on making money may distract you from what the company is really supposed to be doing, and may not be good in the long term.
I would think it more likely that the "IP agreement" is not just a patent license. It's most likely a patent license plus some verilog netlist plus some driver code. It's that last part that likely precludes open source. While no doubt none of that drive code has actually survived into the final driver, it's all that "derived works" type of stuff. No doubt a cautious lawyer would consider the driver code to be partially derived from the licensed snippet. Beyond that, if they've licensed S3TC, it's not even clear that they can even document their texture compression hardware, since it is most likely considered a "derived work."
>If you ask me, the US has a long way to go before reaching the standards in terms of taste and healthiness (is that a word?)
>that grocery food has set in the UK, Belgium, Netherlands, etc.
You seem to have the mistaken idea that "taste and healthiness" are goals of the US food industry.
Remember, the goals of industry in the US - any industry, are revenue and profit. Niceties such as "taste and healthiness" are merely means for achieving that goal. They may even even be simply costs to be reduced as much as possible, at least almost to the point where you'd buy a competitor's product, instead.
And herein lies the problem.
As long as sending email is "free" with no checks and balances, there will be spam. Yet the moment email becomes non-free and gains checks and balances, people are going to scream.
It's a philosophical problem, and requires philosophical underpinnings for an answer. Spam filters, RBLs, etc are just tactical responses. SPF, DKeys, hashcash and the like might be the beginnings of a true answer, but at the moment there's precious little being done at the philosophical level.
(same as terrorism)
But that wouldn't be Patriotic!
The terrorists WANT us to get mired in this recount morass!
(If you can't detect satire, never mind.)