At this point I'm under the impression that either some standard gives us the right to to these patents provided we reverse engineer them or there's something else going on. If you know please enlighten me as I'll sleep better.
As to MPEG, yes, all the free software players (and encoders) are infringing on patents.
The risk is fairly small, since the patent holders have never expressed any interest in pursuing action against the free software-only implementations.
If you market an MPEG decoder or encoder chip without getting a license, though, they'll be knocking at your door.
Microsoft is pursuing a different strategy with their patents on the ASF file formats.
For example, the author of VirtualDub received cease-and-desist notification from MS, and removed support for ASF files from his GPL software.
This approach is an integral part of the MS pitch to the content companies to adopt ASF -- using an unlicensed ASF reader in the US is clearly illegal, which makes it much easier to crack down on such software.
It also makes it easier to crack down on copyright infringement -- if licensed ASF readers won't write MPEG files, having an MPEG version of a movie that was distributed only as ASF is almost an admission of illegal copying.
O.k. Now let's take this one step further. Is implementing a patent for non-commercial or just personal use illegal?
The short answer is "yes".
A patent-holder's rights are about as close to absolute as anything gets in law.
Type of use (eg, personal) doesn't matter.
Independent invention doesn't matter.
If you are using the invention without permission of the inventor, you are infringing.
Resources are limited, energy is limited, and our population is growing exponentially.
============
Wrong.
I'll support the first poster's position, at least with respect to energy.
"Developed" countries are rich because they can apply more and more outside energy to the production of goods and services.
If human labor is the only energy that's available, there are hard limits to how "rich" your society can be.
Harnessing animal power provides a step up.
So does the water wheel.
So does burning coal to power a steam engine.
And so on.
Other factors, such as accumulating capital and human knowledge are important, but the amount of applied energy is a critical enabler.
Use Japan as an example of a developed economy being reasonably efficient about the application of energy, since they have to import most of their important fuels.
According to the Economist's 2003 World in Figures, Japan consumes energy equivalent to 5,224 kg of coal per person per year (for comparison, the US is at about 10,900 kg, Canada is at 11,114 kg, and Qatar tops the list at about 50,000 kg).
By comparison, China consumes 907 kg and India 430 kg.
To become as "rich" as Japan, the billion people in China will need to increase their per person energy consumption by a factor of five.
The billion in India by a factor of twelve.
It is not clear that that much additional energy can be produced, and certainly it cannot be done cleanly in the near future.
As an example, consider the problems that some of the companies providing IT services in India face.
Many of them must build their own generating and delivery facilities in order to provide the power needed to run the computers (and the air conditioning those computers required).
The construction of the buildings requires much more energy per square foot (both the construction itself and the materials that are required) than is commonly used.
I hate to be the bearer of bad news, but minus some breakthrough in either energy production or energy efficiency, India and China are NOT going to become "rich" countries.
The issue is that the DMCA has essentially built a fence around copyright law. Your fair use rights are all there, perfectly intact, but they're inside the fence. There are ways through the fence, but they're either a) limited, or b) illegal.
Didn't say the DMCA was a good thing -- it's not, for the reasons you're citing and others.
Said I didn't think you could prevail in court on an anti-trust basis for DVDs.
Anti-trust law in the US forbids various specific behaviors by a company or a group of companies, and I haven't seen any reasoning that CSS and the terms under which it is licensed fall into any of those fairly narrow categories.
Which company/companies are the monopolist/trust?
What market do they control as a monopoly/trust?
Which of the activities that are illegal under anti-trust law are they conducting?
Illegal tying of products?
Illegal leverage of the monopoly product to gain a monopoly over another product?
Illegal barriers to entry in the market?
I am assuming (this is/., why would I actually READ the patent) the patent will be for "software" buttons on screens, and not the physical buttons.
And therein lies at least some of the problem with the current system.
Even if you're going to allow for software patents, the mere fact that you have implemented something in software that would not be patentable as a physical device ought to be disallowed.
The situation has gotten laughable.
Common business practice (referals, clerk looking up customer's shipping info) have been granted patents as software implementations.
And the EU is apparently going to follow the US down this idiot path.
However, fair use standards CLEARLY state that consumers are allowed to view copyrighted work however they please, as long as they have paid for it. There is no law or statute that allows copyright holders to force consumers to view their work only on certain devices. The DMCA's anti-circumvention provision has this effect, but it would be a blatant anti-trust violation to allow copyright holders to tell consumers they could only view their works on certain devices.
I think you would lose the anti-trust argument in court.
The MPAA doesn't tell you which device you must use, only that you must use a licensed device.
Any licensed device.
And, they argue, the necessary licenses are available on "reasonable and non-discriminatory terms" to all parties.
I don't know what the charges are for a CSS license these days, but say it's $1.00 per device.
I seriously doubt that any court would find that to be an unreasonable charge -- the licensing fees for the MPEG patents are at least that much (yes, the MPEG decoder you wrote from scratch from the published specification infringes on several patents, but the patent holders are only interested in pursuing commercial infringers).
The license says you have to protect the IP, which is also reasonable for a commercial venture, but rules out the possibility of a licensed open-source player.
What scares me is that the content people seem to be getting ready to get in bed with Microsoft and the Windows Media file formats.
Portions of those formats are covered by patents.
Anti-trust can't apply to patents -- the government is granting a constitutionally-approved temporary monopoly.
Microsoft can issue and revoke licenses as they see fit, and reverse-engineering is not a viable option.
It also makes it easy to go after people who transcode -- if the only officially released format is WMA, any MPEG version is clearly an illegal copy.
Indeed, an excellent set of bad examples.
The site also illustrates one of the difficulties of trying to teach physics to the masses:
as soon as you want to know how it really works, you have to use math.
Granted, working out that the victim of a shotgun shooting flying ten feet through the air violates conservation of momentum is a more interesting exercise than what you find in most physics texts, but you still have to work it out.
Just an opinion, but too many students these days have had everything made simple, to the point that they are not willing to persever through the hard things.
Up to a certain point you can wave your hands and get away with it; but at some point in physics you have to deal with calculus, differential equations, and so on.
The ads for the new movie "Van Helsing" have one of my favorite physics errors.
The hero swings down on the rope, snatches up the girl as he passes her, and continues on without any change in speed.
That's one reason why open source software (and open standard) is a good idea.:) It can still live on even when it's creator vanished!
Perhaps, although there's a potential chicken-and-egg problem:
here's a CD-ROM, and on it is the specification for how to read a CD;
here's a Word file, and in it is the documentation for how to parse a Word file.
If you can't read the CD, how do you build a reader?
If you can't parse the Word file, how do you code up a parser?
Some days I think I ought to go back to school and learn to be a librarian...
When Big Oil spends money researching renewable energy, I start imagining that their intent is to scuttle development. I could be wrong, though. Maybe they do want to develop new energy sources. I mean, they can still get a good firm strangle hold on supplies by patenting the new techniques.
Absolutely.
Imagine you're a large petroleum company in the US.
Your business has a variety of components -- collect raw hydrocarbons,
refine them into usable form,
move them around in pipelines and tank trucks and giant ships,
sell them at wholesale (to a natural gas distribution firm, for example) and at retail.
You know that the supply of raw hydrocarbons you are currently using is finite, and collecting them will continue to be more and more expensive.
It makes good business sense to spend some of your current income to investigate alternative sources of raw hydrocarbons.
Oil and natural gas are going to be around the US for a long time no matter what develops in the world of methanol.
Imagine that I have an infinite source of free methanol available starting today.
How long will it take to make the necessary changes so that the energy needs of a typical household are completely satisfied by methanol?
As several people have mentioned, cars and the filling stations that make convenient fueling of the cars feasible may need modifications to handle the more corrosive fuel.
The furnace and water heater that currently burn gas or oil must be replaced/modified to handle methanol.
For oil furnaces, will the current storage facilities handle methanol?
For gas, will the current natural gas network (pipes, compressors, meters, etc) handle methanol?
Since you obviously can't mix methane and ethanol in that delivery system, how does the conversion work?
How long will it take to convert existing coal- and gas-fired electric generating plants to methanol?
The core of the problem is getting past this silliness that anything that generates power == evil.
Precisely the right term for this attitude -- silly.
Most people do not think through the facts about how much energy ANY modern lifestyle requires.
Much of the "wealth" of the developed world is due to the productivity of labor which in turn is built on the ability to apply LARGE amounts of energy to the tasks at hand.
Japan is about as efficient as any developed economy in terms of energy use per capita, at about half the US level.
But even Japan uses five times as much per capita as China, and ten times as much per capita as India.
If you want to live with modern products and conveniences, you have to produce and use the energy that makes them possible.
The real problem will be that the file formats of today will be replaced in 10 years, and will be a legacy file format only readable with a compatibility layer in 20 years. In 50 years, that CD will be unreadable. Of course, storing it in ISO 9660 format would offer some protection. If nothing can read the CD 50 years from now, you could at least fall back to the standard spec write your own code to read it.
Not just the format of the file system (your example), but the format of the individual files.
Does anyone believe that, outside of a handful of people in museums, anyone will be able to read GIF files in 100 years?
Or MPEG-1 compressed video?
Or documents stored as Microsoft Word 97 files?
I've worked with computers for the past 25 years, and have encountered all of the problems that people have mentioned in this discussion:
tapes for which there are no drives available,
tapes and disks which degrade to the point that they are unreadable,
file systems on disks that are not supported by contemporary OSs, and
individual file formats for which no software (or specifications) exist.
I also have a box filled with the paper copies of 25 years' worth of writing, and even the oldest are in good shape and WORK.
If they were especially valuable, I'd make another paper copy and put it in the safe deposit box at the bank.
Audio and video are more difficult, since there's nothing as good as paper for them.
I will say that could be a decent amount to live on, depending on your local cost of living.
And depending on your individual circumstances, it could be a not-so-decent amount to try to live on.
Suppose your scumbag spouse has disappeared, leaving you to try to work and raise the kid.
Here's a guesstimate at some of the expenses you face on a $23000 annual income -- feel free to criticize the amounts.
1725 Social Security and Medicare taxes 1200 health insurance through your job at $100/month 2400 some sort of car payment (perhaps saving for when it dies) and maintenance 4200 rent at $350/month 1200 auto insurance at $100/month 1200 utilities and etc at $100/month 5000 child care at $20/day (good luck) 5475 food at $15/day, every day 22400 total
Didn't count clothing, out-of-pocket medical expenses (and with a little kid, you'll have some), much in the way of gasoline.
Nothing for entertainment.
No savings for a rainy day.
And still spent almost all of it.
What if the scale up is so you can have something the size of a mini fridge in your cellar that creates energy for just your house?
For a commercial power plant, you need 10s or 100s of megawatts.
For home, probably less than 10 kilowatts peak, less than that if I have a flywheel or other way to store power for peak demand periods?
Call it 4-5 orders of magnitude saved.
There are some additional potential savings in a distributed system;
you can in principle do away with the transmission losses we suffer in the current power grid.
Given that it's fusion, I do wonder about the point raised in other threads about how long it would take for my home fusion reactor to become a problem neutron source...
Fixed costs are, of course, enormous -- R&D, testing, compliance, advertising, sunk costs in the factory, etc
Also, the American car companies are now paying off on the promises they made to their unions 30-40 years ago.
IIRC, the Economist pointed out that every new car sold by GM includes $1300 for pension payments to retired workers.
Their article on the competitive position of the Big Three US car makers almost invited the reader to bet on which of them would be the first to declare bankruptcy in order to get out from under their enormous obligations to retirees.
To follow up further, this article suggests that the accumulation of buckyballs can cause brain damage in juvenile fish.
If such turns out to be the case, it will be interesting to understand the mechanism by which carbon, which is fairly non-reactive, and buckballs in particular, which are quite stable once formed, cause the damage.
For example when teaching solving triangles I visually measure off the angles and demonstrate that they all add to 180 degrees. Also teaching the pythagorean(sp?) theorum is helped by getting out a ruler and proving that in fact A^2 + B^2 = C^2 without just saying it's so.
I can understand examples as motivation, and even encourage it -- gee, in this triangle the angles add up to 180, and in this one, and in this funny-shaped one, too.
But the formal proof that such is always the case is also important.
I've seen too many people in college who really ought to know better claim that a few examples constitute a valid proof.
Sure, some people will never need to know it, or even know how to construct a logical argument like a proof -- but some will.
The fact that we don't necessarily know in advance which students are which means that we need to expose them all to it.
Several people have pointed out the "I'll never use this in my life" complaint from students.
I have often wondered why,
at least once you reach high school or so,
there is so much resistance to the idea of having math taught by people who have USED math all of their lives.
I know several people who are retired from engineering or an applied math discipline who would like to teach at the high-school level, but are put off by our current state requirements that would force them to complete the full battery of educational theory classes in order to get certified.
I can understand those requirements for grade school teaching -- I would never trust me with a class of third graders -- but high-school algebra and trig (and calculus in some schools) are a different matter.
Aren't they?
Re:No it's not.
on
A New Ice Age?
·
· Score: 2, Insightful
We mine, worldwide, billions of tonnes of coal every year (the US alone produces just under a billion). How much sulfur dioxide do you think all that lot produces?
Consider that China produces almost a billion-and-a-half tons, and that India is third in world production after China and the US.
In China, nearly 60% of that coal is used as a cooking fuel and to heat buildings -- applications that are not particularly amenable to clean burning techniques.
Because of the expense, little of China's coal is "washed" to reduce sulphur emissions.
Japan has a growing acid rain problem that can be traced directly to coal use in China.
India's position is similar to China's.
Before too many more years, effective limits on nasty emissions MUST consider the developing as well as developed economies of the world.
I have long maintained that one of the things that makes "developed" countries developed is that they produce and apply large amounts of energy per person.
The developing countries cannot catch up with the developed ones simply because there is not sufficient energy available worldwide.
From the Economist's 2003 World In Figures, per-capita energy consumption measured in kilograms of coal equivalent:
Canada 11,114 United States 10,900 Germany 5,626 Japan 5,224 China 907 India 430 Nigeria 178
Using Japan or Germany as the benchmark, China would need to increase its per-capita energy consumption by over five times, India increase by over ten times, and Nigeria by almost 30 times.
There's not that much clean energy available in the world (certainly not at today's prices), and very probably not that much energy of any sort.
What aspects of being "developed" must those other countries give up: mechanized production, large-scale transportation, climate control, sophisticated construction materials?
Microsoft is looking at it constant court costs and anti-trust fines as simply 'the cost of doing business,'
MS is not alone in this behavior.
Large local telephone companies are regulated by the states in which they operate, and many of those states require certain levels of company responsiveness when customers call -- eg, that 95% of calls be answered by a person in less than 30 seconds.
Staffing to the necessary level has historically been quite expensive, and the level of fine that the states can impose for non-compliance relatively small.
When you have to decide between spending $20M on additional staffing, or pay a $10M fine, the answer is fairly obvious.
I suppose extensive outsourcing to India or the Philipines will change the equation...
This incident (or series of incidents) points up that if you cannot control the physical security of your machines and networks, then a broader variety of attacks are possible.
Most universities have appalling physical security.
Some schools are also lax about enforcing appropriate user behavior -- activities that would get an employee escourted to the door at most companies may draw only a mild reprimand.
If you're in charge of the expensive cluster being used for research,
and if you have no choice but to attach said cluster to a network that should be regarded as "hostile",
then you need to take more extraordinary measures to secure your cluster.
I'm certainly no expert, but there are a variety of things that can be done:
put the cluster behind locked doors and change the access code regularly,
isolate the cluster behind a private firewall,
put machines that act as gateways for the broader public into a separate DMZ,
adopt physical authentication devices like SecurID,
etc.
Once you have "bad guys" on the inside, though, you're pretty much screwed.
A few years ago I wrote and ran
(with appropriate permissions)
a piece of Linux software that allowed me to bridge the local Ethernet to any point on the Internet
(Ethernet frames were passed over a TCP connection).
The only capability I needed from the company network was to control a TCP connection or chain of TCP connections to the outside world over which I could pass arbitrary binary data in both directions.
Firewalls that allowed anyone on the inside to establish a TCP connection to port 80 on a machine on the outside made it simple.
Even with service proxies, it would have been straightforward.
I wrote the software for Linux because I needed to read/write raw Ethernet frames.
Now that Windows XP provides that capability, it should be possible to write a similar app to run on WinXP -- for me, a terrifying thought.
The world population is rapidly aging -- at least in developed countries. The number of seniors will explode in the next two decades.
This is not just a problem in the developed countries.
Forecasts indicate that by 2050 the elderly will be a bigger fraction of the population in China than the elderly in the US.
China's birth rate is lower than the US's, and they don't have immigration at the level that we do.
While the Chinese economy is currently growing faster than the US economy, they are starting from a MUCH smaller per-capita income base, and so are likely have more difficulty paying for their elderly.
Any country that has its birth rate under control will eventually have this problem (and if they don't have it under control, they will tend to have other problems).
Many countries are going to have to make some tough decisions about what part of their national income they wish to spend on the elderly.
In the context of this story, how many $100K robots will the young be willing to buy for the elderly (yes, the robots will get cheaper)?
Would you buy one for your Mom (note that male/female ratios are skewed towards women amongst the elderly)?
What would you give up to buy it -- a Porsche, owning your own house, helping your kid through college?
If you are young, would you let the government raise your taxes dramatically so the government can buy robots for all the Moms?
Where does the robot fall on the list of things that are important to buy for Mom?
Would you rather pay for the robot, or for the new $25K cancer medication?
How about the $50K hip replacement surgery?
Bill Gates and Scott McNealy can afford to buy their Moms all of these things -- can you?
I'm not elderly yet, but am almost certainly on the older end amongst the readership here, semi-early-retired and studying economics.
Young people really, really need to be worrying about these issues, because they are going to have to make some difficult choices.
And, I believe, they're going to have to start making them within the next ten years or so.
It is my theory that capitalism, or more precisely free markets, lead to monopolies and oligopolies. As long as you keep introducing good products, have good marketing, have a lot of capital, keep trying hard, and/or have good employees, you will aways dominate.
I suspect the same thing, but believe that there are some additional necessary conditions.
For example, there almost certainly need to be significant economies of scale in the industry in question -- if serving 100,000 customers costs more per customer than serving 10,000, firms will tend to stay small.
There probably need to be barriers to entry in the industry -- if it costs $1B to develop a competing product and get into the business, there will be few new entrants.
New entrants are important as an industry evolves over time, as there will be firms going out of business for a variety of reasons -- bad luck, bad planning, the founder sells out in order to retire, etc.
Continued introduction of new products is not as easy as it sounds over long periods -- US railroads got in trouble because they forgot they were in the transportation business and paid no attention to the evolution of competing car/truck technology.
It's lack of political and public will that's stopping us now
Now there you may be onto something, although I would have expressed it in terms of decisions about what we want to spend money on.
Compared to the 1960s, we (as a group -- sometimes living in a democracy is a bitch) have made some of the following choices:
We would rather live in larger houses, farther out in the suburbs, have all of the roads that entails, and have a third car. Home ownership is at an all-time high. The average house built today is 50% larger than the average from the 1960s. In some states, the number of registered vehicles now exceeds the number of licensed drivers.
Despite the larger house, we would rather spend a good chunk of our income so that Grandma can live out her old age in her own house, or live in some sort of care-provided housing, rather than have her come to live with us and have us take care of her as she becomes infirm.
The government is the transfer vehicle, and if you look at Social Security accounting, roughly 12% of younger people's income is going to the "collective" Grandma.
We would rather buy that third car or a second DVD player than save.
Heck, we would rather borrow money so that we can buy that third car or DVD player -- household debt as a fraction of income is at an all-time high.
The average person has to make a decision to give up something if they want to spend on outer space.
We would rather pay an increasingly large fraction of our income for health care.
It's nice that you can have a knee replaced rather than hobble about on a cane for the rest of your life (approx 280,000 knees replaced last year).
But knee-replacement surgery is a lot more expensive than a cane.
The state has had a huge role in creating new technologies. Half the stuff in the computer industry, a great deal of basic research in genetics, in physics. People seriously overestimate the contributions that the free market to science and knowledge as a whole... Private corporations are great at going the last mile, making a processor or a hard drive that's 10% better than last year.
Interesting.
I would have made almost exactly the opposite statement: people seriously overestimate the contributions that the state makes to science and knowledge as a whole.
Part of it may be that I think you underestimate the value of so-called applied research, the work that has to be done to reduce something from theory to real practice.
It is one thing to know that quantum theory predicts that flash memory is possible.
It is quite another to know how to fabricate such a device and make it work reliably.
Research in the private sector does tend to be more focused on understanding how to solve particular problems.
Yet those attempts produce a surprising amount of "pure" knowledge as well -- Bell Labs was not looking for the background radiation from the Big Bang, they were trying to solve a problem with static on microwave communications links.
Part of it may be that I think computers are a particularly bad example to use.
I think it is difficult to credit much of the current state of computer hardware and software technology to state-led research.
Integrated circuits were developed privately.
Microprocessors were developed privately.
I believe that hard disks were developed privately.
To expand on that, I believe that IBM and other companies put very large amounts of money into pure research on the magnetic properties of thin films -- and that there is very little state-led research in that area.
I would assert that the same situation is true on the software side.
Part of it may be that we disagree on who should get credit for some research results.
Suppose a professor at the local land-grant university develops an elegant new bit of mathematics -- does the state or private enterprise get credit?
Not necessarily a trivial question.
Suppose 10% of the professor's salary is paid by a grant from the NSF,
20% of that salary is paid from students' tuition, and
70% of that salary is paid from an endowment left by a wealthy alumni.
Which sector gets credit for the research results?
Perhaps the building in which he worked was built 50 years ago by the state government for an entirely different purpose.
It would still be hard for me to give much credit to the state government, since in all liklihood, if they have any concern about the professor, it is that he/she spends too much time on research and not enough time teaching (a commonly heard complaint in many states).
However, in some areas you are right, particularly those where basic research has become very expensive.
Private enterprise is unlikely to fund the kind of accelerators needed to do basic experimental research in particle physics.
Genetics may be a different issue -- wasn't it a race to see whether the government-led effort or a privately-funded one would sequence the human genome first?
Arbitrary Actor: You know Chris, I can't but think that this whole idea of yours is expensive and dangerous.
Christopher Columbus: Yeah, you're right actually. Sod this, let's go for a pint, someone else'll do it eventually anyway.
It's not the same thing.
Columbus was not running a scientific experiment or a "voyage of exploration."
It was a fairly coldly-calculated commercial undertaking, even if a somewhat risky one.
He could honestly answer the question by making several points:
Except for the religious zealots, we all know that the world is round.
Heck, the ancient Greeks made estimates of the diameter as a geometry exercise.
Ocean sailing has its dangers, but falling off the edge of the world isn't one of them.
The transport technology is well understood.
Portuguese sailors currently cover the same kinds of distances I'm talking about.
It's not "routine", but it's clearly possible.
On a national scale, this is NOT an expensive undertaking.
If I'm successful, the value of the cargo I will bring back ON THE FIRST TRIP will cover the entire cost.
Granted, if Columbus had done his sums right on the circumference, he probably would not have made the attempt, as the resulting open-ocean voyage would have been beyond the reach of the technology of the day.
As others have pointed out, much of the expense of doing things in space is the cost of getting significant masses as far as LEO.
As I have written before, if the government wants to see us make a go of things in space, spend the money to develop cheap ways to get to LEO.
If you could get stuff that far for, say, $2 per pound, all kinds of things become quite affordable and practical.
Beyond metastability, which always has a statistical probability to bite you in the foot, there is also the probability of a high energy photon, 'obeying the laws of physics' coming by and flipping a bit in your cpu's ALU. You can trust the machine to obey the laws of physics. You can't trust it to always be correct.
I can't lay my hands on the reference, but some years I read a paper based on work done at a company that built "duplex" processors using pairs of off-the-shelf microprocessors and a lot of custom hardware that compared the outputs of those processors.
IIRC, they were using pairs of 68040 processors and there was a transient error (the two processors produced different results) about once per month.
Lots of possible causes -- metastability, seldom exercised design "errors", high-energy photons.
Or bugs in the comparison hardware, although that was a lot simpler than the processors.
Assuming that we're dealing with deterministic stuff here -- no stochastic network behavior or such -- running the program more than once and getting the same results deals with "transients", at least on a statistical basis.
If you run it twice, and get the same results, the odds are enormously good that there were no transients during execution.
Probably at least as good as whether an undetected error in a proof slipped through a human review process.
Slashdot Mirror
As to MPEG, yes, all the free software players (and encoders) are infringing on patents. The risk is fairly small, since the patent holders have never expressed any interest in pursuing action against the free software-only implementations. If you market an MPEG decoder or encoder chip without getting a license, though, they'll be knocking at your door.
Microsoft is pursuing a different strategy with their patents on the ASF file formats. For example, the author of VirtualDub received cease-and-desist notification from MS, and removed support for ASF files from his GPL software. This approach is an integral part of the MS pitch to the content companies to adopt ASF -- using an unlicensed ASF reader in the US is clearly illegal, which makes it much easier to crack down on such software. It also makes it easier to crack down on copyright infringement -- if licensed ASF readers won't write MPEG files, having an MPEG version of a movie that was distributed only as ASF is almost an admission of illegal copying.
The short answer is "yes". A patent-holder's rights are about as close to absolute as anything gets in law. Type of use (eg, personal) doesn't matter. Independent invention doesn't matter. If you are using the invention without permission of the inventor, you are infringing.
I'll support the first poster's position, at least with respect to energy. "Developed" countries are rich because they can apply more and more outside energy to the production of goods and services. If human labor is the only energy that's available, there are hard limits to how "rich" your society can be. Harnessing animal power provides a step up. So does the water wheel. So does burning coal to power a steam engine. And so on. Other factors, such as accumulating capital and human knowledge are important, but the amount of applied energy is a critical enabler.
Use Japan as an example of a developed economy being reasonably efficient about the application of energy, since they have to import most of their important fuels. According to the Economist's 2003 World in Figures, Japan consumes energy equivalent to 5,224 kg of coal per person per year (for comparison, the US is at about 10,900 kg, Canada is at 11,114 kg, and Qatar tops the list at about 50,000 kg). By comparison, China consumes 907 kg and India 430 kg. To become as "rich" as Japan, the billion people in China will need to increase their per person energy consumption by a factor of five. The billion in India by a factor of twelve. It is not clear that that much additional energy can be produced, and certainly it cannot be done cleanly in the near future.
As an example, consider the problems that some of the companies providing IT services in India face. Many of them must build their own generating and delivery facilities in order to provide the power needed to run the computers (and the air conditioning those computers required). The construction of the buildings requires much more energy per square foot (both the construction itself and the materials that are required) than is commonly used.
I hate to be the bearer of bad news, but minus some breakthrough in either energy production or energy efficiency, India and China are NOT going to become "rich" countries.
Didn't say the DMCA was a good thing -- it's not, for the reasons you're citing and others. Said I didn't think you could prevail in court on an anti-trust basis for DVDs. Anti-trust law in the US forbids various specific behaviors by a company or a group of companies, and I haven't seen any reasoning that CSS and the terms under which it is licensed fall into any of those fairly narrow categories. Which company/companies are the monopolist/trust? What market do they control as a monopoly/trust? Which of the activities that are illegal under anti-trust law are they conducting? Illegal tying of products? Illegal leverage of the monopoly product to gain a monopoly over another product? Illegal barriers to entry in the market?
And therein lies at least some of the problem with the current system. Even if you're going to allow for software patents, the mere fact that you have implemented something in software that would not be patentable as a physical device ought to be disallowed. The situation has gotten laughable. Common business practice (referals, clerk looking up customer's shipping info) have been granted patents as software implementations. And the EU is apparently going to follow the US down this idiot path.
I think you would lose the anti-trust argument in court. The MPAA doesn't tell you which device you must use, only that you must use a licensed device. Any licensed device. And, they argue, the necessary licenses are available on "reasonable and non-discriminatory terms" to all parties. I don't know what the charges are for a CSS license these days, but say it's $1.00 per device. I seriously doubt that any court would find that to be an unreasonable charge -- the licensing fees for the MPEG patents are at least that much (yes, the MPEG decoder you wrote from scratch from the published specification infringes on several patents, but the patent holders are only interested in pursuing commercial infringers). The license says you have to protect the IP, which is also reasonable for a commercial venture, but rules out the possibility of a licensed open-source player.
What scares me is that the content people seem to be getting ready to get in bed with Microsoft and the Windows Media file formats. Portions of those formats are covered by patents. Anti-trust can't apply to patents -- the government is granting a constitutionally-approved temporary monopoly. Microsoft can issue and revoke licenses as they see fit, and reverse-engineering is not a viable option. It also makes it easy to go after people who transcode -- if the only officially released format is WMA, any MPEG version is clearly an illegal copy.
The ads for the new movie "Van Helsing" have one of my favorite physics errors. The hero swings down on the rope, snatches up the girl as he passes her, and continues on without any change in speed.
Perhaps, although there's a potential chicken-and-egg problem: here's a CD-ROM, and on it is the specification for how to read a CD; here's a Word file, and in it is the documentation for how to parse a Word file. If you can't read the CD, how do you build a reader? If you can't parse the Word file, how do you code up a parser? Some days I think I ought to go back to school and learn to be a librarian...
Absolutely. Imagine you're a large petroleum company in the US. Your business has a variety of components -- collect raw hydrocarbons, refine them into usable form, move them around in pipelines and tank trucks and giant ships, sell them at wholesale (to a natural gas distribution firm, for example) and at retail. You know that the supply of raw hydrocarbons you are currently using is finite, and collecting them will continue to be more and more expensive. It makes good business sense to spend some of your current income to investigate alternative sources of raw hydrocarbons.
Oil and natural gas are going to be around the US for a long time no matter what develops in the world of methanol. Imagine that I have an infinite source of free methanol available starting today. How long will it take to make the necessary changes so that the energy needs of a typical household are completely satisfied by methanol? As several people have mentioned, cars and the filling stations that make convenient fueling of the cars feasible may need modifications to handle the more corrosive fuel. The furnace and water heater that currently burn gas or oil must be replaced/modified to handle methanol. For oil furnaces, will the current storage facilities handle methanol? For gas, will the current natural gas network (pipes, compressors, meters, etc) handle methanol? Since you obviously can't mix methane and ethanol in that delivery system, how does the conversion work? How long will it take to convert existing coal- and gas-fired electric generating plants to methanol?
Precisely the right term for this attitude -- silly. Most people do not think through the facts about how much energy ANY modern lifestyle requires. Much of the "wealth" of the developed world is due to the productivity of labor which in turn is built on the ability to apply LARGE amounts of energy to the tasks at hand. Japan is about as efficient as any developed economy in terms of energy use per capita, at about half the US level. But even Japan uses five times as much per capita as China, and ten times as much per capita as India. If you want to live with modern products and conveniences, you have to produce and use the energy that makes them possible.
Not just the format of the file system (your example), but the format of the individual files. Does anyone believe that, outside of a handful of people in museums, anyone will be able to read GIF files in 100 years? Or MPEG-1 compressed video? Or documents stored as Microsoft Word 97 files? I've worked with computers for the past 25 years, and have encountered all of the problems that people have mentioned in this discussion: tapes for which there are no drives available, tapes and disks which degrade to the point that they are unreadable, file systems on disks that are not supported by contemporary OSs, and individual file formats for which no software (or specifications) exist. I also have a box filled with the paper copies of 25 years' worth of writing, and even the oldest are in good shape and WORK. If they were especially valuable, I'd make another paper copy and put it in the safe deposit box at the bank.
Audio and video are more difficult, since there's nothing as good as paper for them.
And depending on your individual circumstances, it could be a not-so-decent amount to try to live on. Suppose your scumbag spouse has disappeared, leaving you to try to work and raise the kid. Here's a guesstimate at some of the expenses you face on a $23000 annual income -- feel free to criticize the amounts.
Didn't count clothing, out-of-pocket medical expenses (and with a little kid, you'll have some), much in the way of gasoline. Nothing for entertainment. No savings for a rainy day. And still spent almost all of it.
For a commercial power plant, you need 10s or 100s of megawatts. For home, probably less than 10 kilowatts peak, less than that if I have a flywheel or other way to store power for peak demand periods? Call it 4-5 orders of magnitude saved. There are some additional potential savings in a distributed system; you can in principle do away with the transmission losses we suffer in the current power grid. Given that it's fusion, I do wonder about the point raised in other threads about how long it would take for my home fusion reactor to become a problem neutron source...
Also, the American car companies are now paying off on the promises they made to their unions 30-40 years ago. IIRC, the Economist pointed out that every new car sold by GM includes $1300 for pension payments to retired workers. Their article on the competitive position of the Big Three US car makers almost invited the reader to bet on which of them would be the first to declare bankruptcy in order to get out from under their enormous obligations to retirees.
To follow up further, this article suggests that the accumulation of buckyballs can cause brain damage in juvenile fish. If such turns out to be the case, it will be interesting to understand the mechanism by which carbon, which is fairly non-reactive, and buckballs in particular, which are quite stable once formed, cause the damage.
I can understand examples as motivation, and even encourage it -- gee, in this triangle the angles add up to 180, and in this one, and in this funny-shaped one, too. But the formal proof that such is always the case is also important. I've seen too many people in college who really ought to know better claim that a few examples constitute a valid proof. Sure, some people will never need to know it, or even know how to construct a logical argument like a proof -- but some will. The fact that we don't necessarily know in advance which students are which means that we need to expose them all to it.
Several people have pointed out the "I'll never use this in my life" complaint from students. I have often wondered why, at least once you reach high school or so, there is so much resistance to the idea of having math taught by people who have USED math all of their lives. I know several people who are retired from engineering or an applied math discipline who would like to teach at the high-school level, but are put off by our current state requirements that would force them to complete the full battery of educational theory classes in order to get certified. I can understand those requirements for grade school teaching -- I would never trust me with a class of third graders -- but high-school algebra and trig (and calculus in some schools) are a different matter. Aren't they?
Consider that China produces almost a billion-and-a-half tons, and that India is third in world production after China and the US. In China, nearly 60% of that coal is used as a cooking fuel and to heat buildings -- applications that are not particularly amenable to clean burning techniques. Because of the expense, little of China's coal is "washed" to reduce sulphur emissions. Japan has a growing acid rain problem that can be traced directly to coal use in China. India's position is similar to China's. Before too many more years, effective limits on nasty emissions MUST consider the developing as well as developed economies of the world.
I have long maintained that one of the things that makes "developed" countries developed is that they produce and apply large amounts of energy per person. The developing countries cannot catch up with the developed ones simply because there is not sufficient energy available worldwide. From the Economist's 2003 World In Figures, per-capita energy consumption measured in kilograms of coal equivalent:
Using Japan or Germany as the benchmark, China would need to increase its per-capita energy consumption by over five times, India increase by over ten times, and Nigeria by almost 30 times. There's not that much clean energy available in the world (certainly not at today's prices), and very probably not that much energy of any sort. What aspects of being "developed" must those other countries give up: mechanized production, large-scale transportation, climate control, sophisticated construction materials?MS is not alone in this behavior. Large local telephone companies are regulated by the states in which they operate, and many of those states require certain levels of company responsiveness when customers call -- eg, that 95% of calls be answered by a person in less than 30 seconds. Staffing to the necessary level has historically been quite expensive, and the level of fine that the states can impose for non-compliance relatively small. When you have to decide between spending $20M on additional staffing, or pay a $10M fine, the answer is fairly obvious.
I suppose extensive outsourcing to India or the Philipines will change the equation...
Once you have "bad guys" on the inside, though, you're pretty much screwed. A few years ago I wrote and ran (with appropriate permissions) a piece of Linux software that allowed me to bridge the local Ethernet to any point on the Internet (Ethernet frames were passed over a TCP connection). The only capability I needed from the company network was to control a TCP connection or chain of TCP connections to the outside world over which I could pass arbitrary binary data in both directions. Firewalls that allowed anyone on the inside to establish a TCP connection to port 80 on a machine on the outside made it simple. Even with service proxies, it would have been straightforward. I wrote the software for Linux because I needed to read/write raw Ethernet frames. Now that Windows XP provides that capability, it should be possible to write a similar app to run on WinXP -- for me, a terrifying thought.
This is not just a problem in the developed countries. Forecasts indicate that by 2050 the elderly will be a bigger fraction of the population in China than the elderly in the US. China's birth rate is lower than the US's, and they don't have immigration at the level that we do. While the Chinese economy is currently growing faster than the US economy, they are starting from a MUCH smaller per-capita income base, and so are likely have more difficulty paying for their elderly. Any country that has its birth rate under control will eventually have this problem (and if they don't have it under control, they will tend to have other problems). Many countries are going to have to make some tough decisions about what part of their national income they wish to spend on the elderly.
In the context of this story, how many $100K robots will the young be willing to buy for the elderly (yes, the robots will get cheaper)? Would you buy one for your Mom (note that male/female ratios are skewed towards women amongst the elderly)? What would you give up to buy it -- a Porsche, owning your own house, helping your kid through college? If you are young, would you let the government raise your taxes dramatically so the government can buy robots for all the Moms? Where does the robot fall on the list of things that are important to buy for Mom? Would you rather pay for the robot, or for the new $25K cancer medication? How about the $50K hip replacement surgery? Bill Gates and Scott McNealy can afford to buy their Moms all of these things -- can you?
I'm not elderly yet, but am almost certainly on the older end amongst the readership here, semi-early-retired and studying economics. Young people really, really need to be worrying about these issues, because they are going to have to make some difficult choices. And, I believe, they're going to have to start making them within the next ten years or so.
I suspect the same thing, but believe that there are some additional necessary conditions. For example, there almost certainly need to be significant economies of scale in the industry in question -- if serving 100,000 customers costs more per customer than serving 10,000, firms will tend to stay small. There probably need to be barriers to entry in the industry -- if it costs $1B to develop a competing product and get into the business, there will be few new entrants. New entrants are important as an industry evolves over time, as there will be firms going out of business for a variety of reasons -- bad luck, bad planning, the founder sells out in order to retire, etc. Continued introduction of new products is not as easy as it sounds over long periods -- US railroads got in trouble because they forgot they were in the transportation business and paid no attention to the evolution of competing car/truck technology.
Now there you may be onto something, although I would have expressed it in terms of decisions about what we want to spend money on. Compared to the 1960s, we (as a group -- sometimes living in a democracy is a bitch) have made some of the following choices:
Interesting. I would have made almost exactly the opposite statement: people seriously overestimate the contributions that the state makes to science and knowledge as a whole.
Part of it may be that I think you underestimate the value of so-called applied research, the work that has to be done to reduce something from theory to real practice. It is one thing to know that quantum theory predicts that flash memory is possible. It is quite another to know how to fabricate such a device and make it work reliably. Research in the private sector does tend to be more focused on understanding how to solve particular problems. Yet those attempts produce a surprising amount of "pure" knowledge as well -- Bell Labs was not looking for the background radiation from the Big Bang, they were trying to solve a problem with static on microwave communications links.
Part of it may be that I think computers are a particularly bad example to use. I think it is difficult to credit much of the current state of computer hardware and software technology to state-led research. Integrated circuits were developed privately. Microprocessors were developed privately. I believe that hard disks were developed privately. To expand on that, I believe that IBM and other companies put very large amounts of money into pure research on the magnetic properties of thin films -- and that there is very little state-led research in that area. I would assert that the same situation is true on the software side.
Part of it may be that we disagree on who should get credit for some research results. Suppose a professor at the local land-grant university develops an elegant new bit of mathematics -- does the state or private enterprise get credit? Not necessarily a trivial question. Suppose 10% of the professor's salary is paid by a grant from the NSF, 20% of that salary is paid from students' tuition, and 70% of that salary is paid from an endowment left by a wealthy alumni. Which sector gets credit for the research results? Perhaps the building in which he worked was built 50 years ago by the state government for an entirely different purpose. It would still be hard for me to give much credit to the state government, since in all liklihood, if they have any concern about the professor, it is that he/she spends too much time on research and not enough time teaching (a commonly heard complaint in many states).
However, in some areas you are right, particularly those where basic research has become very expensive. Private enterprise is unlikely to fund the kind of accelerators needed to do basic experimental research in particle physics. Genetics may be a different issue -- wasn't it a race to see whether the government-led effort or a privately-funded one would sequence the human genome first?
It's not the same thing. Columbus was not running a scientific experiment or a "voyage of exploration." It was a fairly coldly-calculated commercial undertaking, even if a somewhat risky one. He could honestly answer the question by making several points:
Granted, if Columbus had done his sums right on the circumference, he probably would not have made the attempt, as the resulting open-ocean voyage would have been beyond the reach of the technology of the day. As others have pointed out, much of the expense of doing things in space is the cost of getting significant masses as far as LEO. As I have written before, if the government wants to see us make a go of things in space, spend the money to develop cheap ways to get to LEO. If you could get stuff that far for, say, $2 per pound, all kinds of things become quite affordable and practical.
I can't lay my hands on the reference, but some years I read a paper based on work done at a company that built "duplex" processors using pairs of off-the-shelf microprocessors and a lot of custom hardware that compared the outputs of those processors. IIRC, they were using pairs of 68040 processors and there was a transient error (the two processors produced different results) about once per month. Lots of possible causes -- metastability, seldom exercised design "errors", high-energy photons. Or bugs in the comparison hardware, although that was a lot simpler than the processors.
Assuming that we're dealing with deterministic stuff here -- no stochastic network behavior or such -- running the program more than once and getting the same results deals with "transients", at least on a statistical basis. If you run it twice, and get the same results, the odds are enormously good that there were no transients during execution. Probably at least as good as whether an undetected error in a proof slipped through a human review process.