On the other hand, no, I won't save you the embarassment. http://www.ejectionsite.com/eunderh2o.htm This is not one of the seats designed to eject underwater, but even it performed the feat in the end.
According to your link, he ended up with a broken back and lacerated arm from the attempt. Better than drowning, but hardly a ringing recommendation. (For underwater use, the ability to blow the canopy and detach from the seat without triggering the ejector charge would be much more use than any ejection mechanism...)
How much and how far can the creators be allowed to go ? What happens when "some payment" becomes "exorbitant payment"?
People stop using that product, and switch to one of the many rivals.
Your claim that redhat could easily have paid $50k can be easily voided with the counter-claim that this amount of money is 1 year's salary for a redhat developer, who can do lot more than just package an mp3 player in 1 year.
Yes, RedHat chose to spend the money some other way. So what? That doesn't "void" anything.
And mp3 is one of the most benign examples of software patents. Try getting a license to develop a player for MS MPEG4 on linux/*BSD or even MS windoze. Same for Apple. Let us know if Bill is happy with $50k.
WTF do you mean "MS MPEG4"? If you mean MPEG4, forget MS: it's the MPEG people you need to talk to. Bill isn't involved, nor is Apple - you would normally buy the license from "MPEG LA", on standard and open terms. (Tivo, for example, would pay $0.50 per unit shipped.) The prices are published - lots of legalese, but AIUI commercial distribution of an MPEG4 decoder - for Windows, FreeBSD, Linux or anything else - would cost $0.25 per unit (up to a maximum of $1m). Under 50,000, it's royalty free. (Microsoft and Apple would presumably opt for the flat-rate license.)
So, $50k doesn't buy unlimited distribution rights on MPEG4 software. Let us know how you get on developing your own video codec for $50k - Ogg have just about completed a commercially-usable audio codec, they might achieve video one day - but not on that budget.
In a patent-free and patent-filled world, an user is free to create one's own new format. But, in a patent-filled world, an user cannot choose to freely use a format that is patented, a choice the user has in a patent-free world. Under such conditions, it is obvious that in a patent-filled world, an user will have a greater need for creating one's own format (since thats all the choice the user has got). This means in a patent-filled world will have more number of mutually-incompatible formats than a patent-free world.
You mean, in a "patent-filled world", you need to get the patent-holder's permission to use their patented format - iff they choose to impose licensing requirements. (We'll ignore for now the fact you can't patent formats per se...) What is so wrong with that? As I said earlier, if you want to use somebody's invention in a way they don't like, invent your own damn product. Just like copyright: the inventor decides the terms.
About using using wine to play movies on linux, try the alternative of distributing mplayer's win32 port (oh ya, mplayer does play these patented MS media formats natively on win32) inside US and wait for your day in court.
Winamp seems to manage it OK - what does mplayer do wrong to cause a problem?
True enough. My point was that you're talking about real money, rather than the hypothetical money that the RIAA claims to be losing.
No - I was replying to a post which claimed more money would be made without patents (the scenario about Real making their app open source, and making money from installing servers for media sites.)
Clones are the real competition. If you need to play RealPlayer content, a video player from Apple or Microsoft doesn't do you any good when it can't play that content. You're locked into using RealPlayer. Since Real has a monopoly in this market, they can charge any price, no matter how exorbitant. And as the consumer, you can't go to a competitor offering a better price -- the competitor's product (while similar) doesn't meet your needs. You're stuck with the monopolist.
No, that's exactly how Apple and Microsoft got into the market. They didn't try to compete using their own Real clone, they built their own media client+server combo and started pushing that. Nobody was "stuck" with anything - they just had to have the appropriate (free) viewer for that site. The viewer doesn't have a choice of client software - but then, that isn't what anyone was competing for. The client software is just a loss-leader - the competition (and choice) is for the sites.
Yes, the monopolist will spend their money too. That doesn't mean it's a wash for society. Monopoly rents extract substantial costs from society at large (in various ways) for the benefit of the monopolist. When the monopolist spends that money, the value to society and the economy doesn't match the costs that were extracted. This is exactly why economists call it "inefficient" and are very wary of monopolists...
I agree true monopolies of a whole sector are bad - I don't think it's happened in computing via patents. Real don't have a monopoly on video or audio streaming. Google (with the PageRank patents) don't have a monopoly on searching. Microsoft have a [near] monopoly on desktop OS and office suites - but that has nothing to do with patents. A patent doesn't normally confer a monopoly on a whole market (which would be the harmful kind of monopoly economists oppose) - just a monopoly on a particular device. If Boeing patent some new wing design, or fly-by-wire system, it doesn't give them a monopoly on aircraft - it just makes their aircraft better, and forces competitors to develop something different to compete. (Or take more government subsidies, in Airbus's case - you know Concorde cost about the same per unit as the Space Shuttle?) Without patents, Airbus could sit and wait for Boeing to invent something, then produce a cheap clone - everyone loses.
Looking at the harmful monopolies of the past, none involve patents AFAICS. Most of the ones I'm familiar with were actually created by government mandate (in Britain, the government telephone company, railway and health service were all taken from their creators this way - and the resulting mess sucked badly.) Where have patents given any true monopoly?
Would you consider the market for CD players to be a monopoly, by the way? Would you change your answer if you knew they were all subject to a patent licensing fee to Sony and Philips, who co-developed CDs?
Unfortuneately, this is not true in the lastest GM products for the US market. Most of the data loging is now in an intregrated unit with common memory space. The systems that generate the trigger events are all stand alone but the sensors that feed the digital subsystems are shared.
The sensors are shared, of course, but the control systems are still largely isolated. In a tie-up with an earlier/. story, they often use QNX as the operating system, which is capable of keeping a b0rked logging daemon very firmly under control even on a shared CPU;-)
(The ones I've seen are all physically distinct boxes, but I agree they'll probably be integrated as long as GM is very very sure it'll work properly.)
You talking power steering? That is a hydralic feed back loop using a pilot valve in the steering box. Turning the control wheel moves the steering linkage that is built with a bit of flex in it. Deflection of the linkage moves the pilot valve. The pilot valve shuttles the control valve and moves the front wheels until the defection in the linkage centers the pilot valve. Loss of pressure in the control section or the pilot section of the system places all the load on the linkage so even with the pump dead you can still steer the car. No digital data to record.
That's why I said system rather than circuit;-)
Having said that, the manufacturers are planning to move to electric power steering in place of hydraulics - big weight and cost savings, apparently. And more digital data for Little Brother to log! (See the earlier story about 42V cars. I'd prefer 48V - made of -24,0,+24 - but they didn't consult me:P)
The money the owners get isn't created out of thin air. It's paid by those who license the patent. That's money they can't spend on other things -- it's quite legitimate to say that the license fees on a patent are a real cost to the economy.
You could say exactly the same about any purchase. The reality is, spending money is not a cost to the economy - it's exactly the transaction the economy is built from.
More importantly, the patent bars competition, which lowers prices -- another cost to the economy.
It doesn't bar real competition. Real has QuickTime and Media Player, MP3 has Ogg Vorbis, Cisco has Nortel, Netscape has Apache and Microsoft. Patents bar clones - you can't just go and write your own RealPlayer, assuming they've patented that, but you can go and write your own video player, just as Apple and Microsoft did.
On the other hand, if nobody has to pay monopoly rents to license a patent, that's real money they don't have to spend, and they'll certainly find another way to spend it.
The money still gets spent, just in a different place. No net gain, you're just taking money from Thompson and giving it to another company instead. What about when Thompson spends the money they get in license fees? What about the development it would otherwise have funded - where do you think MP3Pro came from?
Because others were forced to create incompatible formats to compete with them. A situation that's bad for consumers and for all the companies involved.
Yes. We all know how an open format spec will be used by everyone involved. *cough* deb RPM ebuild *cough*. Look at the different versions of HTML, where few sites bother to conform to the specs, and few browsers bother to render to them. Do you really think video and audio would have avoided that?
All the companies except Real might benefit, perhaps, but without their proprietary format, other people would be using their free player software with someone else's free or cheaper server. Real take a hit on the player (since it's a loss leader) then get screwed on servers (other companies with lower overheads - no player of their own - can undercut them).
The formats have no particular technical advantages or disadvantages, they're just different.
Not really true on either count. QuickTime and Media Player both had to offer better features to take market share from Real: because they couldn't just clone it, they had to compete on merit. Also, Real are now working on a unified server, meaning the player and format will be the viewer's choice. Looks to me like the best of both worlds: free players, and a choice of format.
Indeed, I don't think the US in the jet age has produced any really *beautiful* airliners, like the Comet or Concorde. I've always liked the B727's looks, and the B747 is impressive, but not beautiful.
The big ones are all the airborne equivalent of buses. From the right angle, 747s are pretty nice, but if you go for smaller jets, the Cessna Citation X is pretty nice IMO. Or the BBJ (Boeing Business Jet - a variant of the 737 Next Generation airframe), especially from the interior.
I tend to prefer the ones I've flown myself, but that's probably just my inner control-freak talking. Ironically, the AFB I first trained at is now the civilian airport I use most often - it still feels a bit weird flying out of there in a 737. Not being military, the flights are less exciting, but most of the passengers probably prefer that;-)
As other posters have said, I think that trying to mod the chip would be dangerous. This data feed is likely to be the same one that goes to important things like anti-lock brakes, traction control, etc.
The systems are totally separate, in fact. The car designers seem to follow the sort of Keep It Simple philosophy of good programming. The brakes, steering and airbag are absolutely critical, so you have a very very simple and totally dedicated system for them:
This circuit triggers the airbag whenever the car experiences an impact greater than X. You test the ass off it to make absolutely sure it does exactly that.
That circuit relaxes the brakes for X ms, whenever the wheels are slipping more than Y. Test the ass off it.
This system applies this much extra force in the direction you're turning the wheel. It gets tested to hell and back too.
Then there's a solid state recorder (usually sitting under the front passenger seat) which records whatever information it's fed from the various sensors. It's simple and designed to be solid, but it's deliberately not safety critical.
Ditto the engine management chip. Designers know customers will often swap those for an alternative to get better acceleration or whatever - so it only has access to the fuel mix controls, gear selection and so on. Whatever that chip does, it doesn't have any access to the brakes, steering, airbag - it can't affect the safety systems, the worst it can do is stall the car. (Or cause bad engine wear with dumb settings, but that doesn't kill anyone...)
If only MS put this kind of separation into their software: why exactly does a word processing document have the ability to email itself?
In a way, yes. It would be a short-lived monopoly, but yes. If Real had come out with a patent-free, Open Source format, and had sold servers (yes, you _can_ profitably sell Open Sorce software) and sold expertise in setting them up, they would've temporarily had a very lucrative monopoly.
Exactly that scenario happened anyway, until QuickTime and Media Player came along to muscle in on the market. Why would making the format patent-free have changed anything? They gave their own player software away anyway - including Linux support! - how would the absence of patents strengthened this control?
Eventually their prices would've had to come down because of competition from others, which they could do nothing to prevent. But, until others became credibly well-versed in the technology they created, they would've had their monopoly.
A monopoly patents could have strengthened - making life better for Real - how could Real's patents have weakened this control, or hastened the competition?!
If it's their system, then why do patent protections run out? You don't suddenly stop owning your car after 5 years.
You do if it's leased. A patent - like a lease - confers "ownership" for a finite time.
Also, it's probably costing the economy billions upon billions of dollars for these people to have their monopoly. It's inefficient and state granted.
Ah, the RIAA argument: it's "costing" billions, because you think billions might otherwise be made. Nice try. It's creating money for the owners. Perhaps some other system would make them more money, perhaps not - but you cannot claim it is "costing" the economy anything.
Yeah, and you are technically infrining on several patents using half those programs.
Such as? I'm pretty sure Real's player doesn't infringe Real's patents. Ditto Quicktime and Media Player. You thought I meant I was running some other player to view content in those formats? (Wine.)
The only reason those programs exist is because their existence serves other purposes of the creators of the format, and there's nobody with deep enough pockets to be worth suing.
The programs I use were created by the same people as the format they play - I'm infringing nothing.
And, just how much have they been actually helped by their patents? It's not as if ReplayTV didn't also exist. Also seems to me that Real's patents have actually hurt them. They'd have widespread adoption if it weren't for patents. They'd be able to make a ton more money installing and setting up servers for people.
Real do have widespread adoption, and they do make money on servers. They have competition from Microsoft and Apple, doing exactly the same thing, which is the only issue they face. So, you're claiming removing patents would be good, because it would give Real a monopoly patents currently prevent?!
No, actually, there is an important one that hasn't. Mojo nation used it to implement Mojo, but Mojo nation's libre spinoffs can't use it. I can find the references to the system if you'd like. The patents don't run out for a couple of years.
That's exactly how it's supposed to work. They developed a system, and now have a monopoly on their system. You want one? Fine - design your own.
Yeah, and that's why video sucks so badly under Linux.
Funny, it's fine for me, even Windows Media files.
Doesn't help developers... the only people it helps are people like Microsoft and Apple.
Microsoft and Apple are developers.
They lock out small competitors while the big ones quietly cross license. It's extremely damaging to competition and represents an unacceptably large barrier to entry.
Unacceptable to whom? Plenty of startups have managed to create something worthwhile, and benefit from patent protection for it. (Tivo, Real.)
And, the developers don't get paid a cent. They get their salaries sure, but no money for the patent.
You think their salary grows on a company money-tree? Where do you think the patent royalties go, if they don't go to the company owning the patent? What do you think boosts the company's stock price, making the developers' stock options worthwhile?
Patents... your government working hard to prevent any upset of the people who line the pocketbooks of its politicians.
Trying to prevent any upset of the economy. Sounds like a good idea.
I wouldn't be surprised if the system that lead to IBM and Unisys both being able to get patents covering the same algorithm also leads to one patent covering something ostensibly made available by the expiration of another patent.
No. If, for example, someone turns up with another US patent covering RSA (on which the original patent has expired), you could simply cite the expired patent as prior art. Goodbye second patent.
Internationally I think it's a different story--a US patent for LZW is expected to expire soon, but other patents covering the same algorithm abroad will expire later. Thus this algorithm will be simultaneously covered by patents and not covered by patents (assuming the LZW patent is not renewed) in countries which support software patents.
That has always been the case, thanks to different patent systems. The change the original article was objecting to removes this conflict (at least between EU countries): instead of Unisys having a different patent on LZW for Japan, the US, Canada, the UK, France, Germany etc, it's reduced to Canada, US, Japan, EU. ISTR there's also a requirement to patent nearly simultaneously, so you couldn't get a US patent, then get an EU patent on the same invention just before the US one expires.
To be fair, we don't know how much RedHat has budgeted for this (assuming they would want to engage in purchasing patent licenses at all) and Thomson might choose to not sell RedHat a license.
I don't think they can: anti-trust laws prohibit that kind of discrimination. (This is how MS got into hot water: trying to use Windows license pricing as leverage against OEMs. That's illegal under the Sherman Act - if you sell Dell 10,000 OEM licenses for WinXP at $25 each, you can't charge HP $50 each for the same number.)
Also I would wonder if the decoder license is still incompatible with Free Software licenses because of something in the text of the patent license.
The GPL is clear about that: it doesn't try to override patents, or force licensing. Just like encryption: the existence of anti-encryption laws doesn't prevent GPLed crypto software. Those laws may prevent some people using the software, but that doesn't affect the GPL.
I think courting a patent infringement lawsuit is unwise.
I agree. That doesn't stop it happening...
I'm guessing RedHat agrees and that is why they removed MP3 software from their GNU/Linux distribution.
Yes: they took the slightly cheaper of their two legal options - removing MP3 players to save the $50k.
It's disappointing you aren't citing any reason for disobeying the law other than your immediate desire. Nothing about challenging bad laws, raising awareness of a system that prohibits competing on merit, or working on a better solution--any of the reasons that might encourage people to agree with why you advocate infringing patents. You appear to advocate infringing patents just because you want these features and you're not willing to consider the long-term ramifications of these choices.
I'm not advocating it, just stating it currently happens and is unlikely to stop because of the proposed changes in the law. The argument earlier about patents eliminating free software is like claiming a ban on crystal meth lab equipment would kill medicinal marijuana use...
No, and fortunately for this case another algorithm was not patent-encumbered. As patents cover more ideas more algorithms will be cut off from implementation and the strategy that led to gzip will be cut off.
No. By definition, all the current patent-free algorithms will remain so. New algorithms will be developed; some patented, some not. Meanwhile, the clock ticks away towards existing patented algorithms becoming patent-free - nothing will ever become patented.
For programmers in countries that observe software patents, there's no known way to legally encode or decode MP3s without acquiring a license from Thomson.
Or using an existing licensed tool to do it. RedHat could have taken the opportunity to remove the whole problem, at least for decoders: pay the $50k, and mpg123 and co are legit.
But you need not remind us of your solution--ignore the law because it gets in your way.
That's not "my solution". Someone claimed a whole category of free software would become illegal, hence extinct, under the new rules; I pointed out it is already illegal, and shows no sign of becoming extinct because of it.
I can only hope that any developer who doesn't have the US$50k or more to acquire an MP3 patent license somehow comes up with the money to defend themselves should they be sued for patent infringement.
I doubt they'd need it. If sued, a defense would be a waste of money: they'd just end up paying a big lawyer's fee on top of the license fee they should have paid in the first place.
Yes, their physical states were not determined by the television, and never has anyone ever made any claim that television is a teleportation device. But these kinds of crimes NEVER used to happen in Bhutan, so where are they getting the ideas from?
The article said the drunkenness did happen before, and there isn't much of an "idea" involved in hurting someone while drunk. More important is where did the drugs come from?
The fact that these people were intoxicated is insignificant: there were heroin addicts and drunks before television, but not this kind of crime.
I find that hard to believe. Murder is not an "idea", it's a violent act, which was committed as the result of the intoxication. Are you really trying to claim an aggressive drug addict would have had no idea how to kill or injure someone until he saw a TV program?
Now that the country of Bhutan is becoming corrupt and immoral just as our culture is being exported to it, you claim that television is a "convenient scapegoat"?
Yes. The article said it's one of many things which changed - they also have a drug and alcohol problem (which does result in violence in Western cultures - even among those with no access to TV).
As a counterpoint: Afghanistan had no TV either, until recently. The presence of Al Queda and multiple large armed factions in constant conflict suggests they still managed to be more violent than Bhutan, despite the absence of this "terrible device"...
A crucial part of mental development is learning to distinguish fantasy from reality, recognising that most of the programs on TV are the former. Western children develop this at a very early age; these people's isolated religious lives seem to have stunted this development. In time, it will come - they just have to be patient, and handle the adjustment carefully.
TV is indeed a powerful influence - like power tools, you should keep a careful eye on young kids with it, and gradually they will learn how to handle it properly. Try to take shortcuts, and you'll end up with red stains all over the place. That doesn't make either item a "terrible device", just one you need to learn to handle.
What wrong is that the security of the Internet as a whole was comprimised in the meantime.
How? SSL and co have been around for years.
What's wrong is that secure digital cash based systems aren't being deployed right now, instead of when the patents run out. That is a huge cost born by everybody.
They ran out years ago, and secure digital cash systems are already deployed.
Yeah, by people who can afford the infrastructure it takes to have lawyers and licensing agreements.
i.e. any software company. Yes. So?
It prohibits using the algorithm for me,
How? Use it yourself, and they will never know, let alone care or sue. Start distributing it - i.e. competing with the paying users - and they'll start caring, and rightly so.
and for a lot of other people. I'm unemployed, and writing encryption software right now. I can't afford a lawyer. I can't afford licensing fees, and I'm not going to burden myself with having to do those things if I should ever start selling my program.
So, you aren't willing to share any of the profit you make with the guys who developed the system you're using?
So, each individual user will have to purchase the license separately in order to use the program? That sounds workable.
This reminds me of the complaint about the GPL being "viral". Either comply with the license, or write your own damn algorithm. If you want to distribute an implementation of Thompson's MP3 algorithm, you'll have to pay them.
Patents are one of the largest stifling forces in the software world today. They do nothing to encourage innovation.
How do you justify that claim? MP3 was developed by a company in exchange for the license revenue they now receive. Ditto almost every other major format, from CDs to MPEG-4.
They are merely weapons for corporations to use against one another in the marketplace.
Bollocks. They're a source of revenue for the developers who create the systems in question. They are not "weapons", except in precisely the way they were intended: to lock out competitors for a limited time. Develop a new system, and you can either license it to others (making money from the patent licensing fee) or exploit it yourself (making money from your temporary monopoly). That's exactly how the system was intended to work from the outset.
Yes, there are some cases of people using patents to prevent an invention being sold; ISTR Macrovision did this (patent a Macrovision-defeating circuit) to keep anti-Macrovision devices off the market. That's a sort of abuse of the system, but hardly a big problem.
The $50K is for a decoder. An encoder like GRIP still has a unit price. None of that covers the legal complications in providing source code to end users. Technically people who compile mp3 software from source (without a license) are infringing the patent-- licenses for end users are not available.
I would disagree there. Where is it stipulated the license applies only to binary distribution, not source? Whether the program is distributed in binary, source or interpretive form, you need a license for that program. Also, RedHat distributed a binary decoder (which would be covered by the $50k license) and the source code. Compiling the source code may be a legal problem for end-users in theory, but not for RedHat.
You're right, the LAME acronym is meant to be cute, because technically it's not an encoder. It's just source code that still needs to be developed into an encoder by compiling it or incorporating the code into a different package.
Why do you think an encoder in source form is "not an encoder"? Legally, it is. In practical terms, it is. As a source RPM, you can even (IIRC - it's a while since I used RedHat) install it like a binary one, it just takes longer.
You are over-simplifying the matter greatly and in dangerous ways to obfuscate the potential for real harm from software patents. That there is a case where the grey areas have been exploited to do an end-run around the licensing strictures does not provide significant evidence that people should be able to patent what is essentially math in the first place. This is a change to the traditional role of the patent that is neither necessary nor desirable in my view.
What change? The patents we've been talking about were issued under the existing rules. There's no "grey area" being exploited - the developers just ignore the patent issues. They ignore the existing rules, why would they start paying attention to changed ones? A few companies, like RedHat, pay enough attention to leave customers to download their own copy, rather than bundling one. Hardly a major problem. There's no change coming in this respect.
I do agree mathematical techniques shouldn't be subject to patents. (Patenting RSA - which essentially consists of a single equation about 7 characters long! - is insane.) I'm not so sure when it comes to complex systems like MPEG-4: why shouldn't the developers be entitled to some payment for it? Developing that system took serious work - anyone claiming to have developed an MPEG-4 codec without reference to any of the MPEG's work is obviously lying.
In June 1999, Bhutan became the last nation in the world to turn on television. The Dragon King had lifted a ban on the small screen as part of a radical plan to modernise his country
Call me naive, but I seriously doubt cable TV was the ONLY thing done to 'modernise his country'. But, telling the whole story never sells eyeballs, now does it?
Quite. If you look, you'll notice one of the murders being blamed on TV was this one:
Three days later in Thimphu, Bhutan's sedate capital,
where overindulgence in rice wine had been the only social vice, Dorje, a 37-year-old truck driver, bludgeoned his wife to death after she discovered he was addicted to heroin.
Yep, obviously all TV's fault. Or this one:
In Bhutan, family welfare has always come first; then, on April 28, Sonam, a 42-year-old farmer, drove his terrified in-laws off a cliff
in a drunken rage, killing his niece and injuring his sister.
The first one reminded us alcohol abuse was a problem before. We have two murders: one by a guy on drugs, the other by a drunk - how is either of these TV's fault? Did the first guy get his heroin via the TV? Did the other get drunk (an existing vice, as the article points out) from a program about alcohol?
TV is a convenient scapegoat. Dealing with the reality - that these people have led very sheltered lives so far, and are struggling to deal with life now some of that isolation has ended - is much harder.
Yeah, and how many years were people afraid to implement RSA until the big hullabaloo over the patent expiring?
None, they just had to give some money to the developers if they were making commercial use of it. "You can use my program if you give me some money for it. It's free for non-commercial use, too." What's wrong with that?
We might've had those technologies widely deployed as much as a decade earlier if those patents weren't there.
It's been widely used for years anyway.
A lot of people here seem to have missed something. A patent does not prohibit using an algorithm, it just allows the developer to set a price for it. Unless you're demanding that you get it "free as in beer", there's no problem: just pay the license fee! (It doesn't even prevent GPLed implementations, AIUI: your implementation may be freely copyable but not include a patent license.)
You can produce an entirely legitimate GPLed MP3 encoder - just point out "use of this software requires a license fee payment of $2.50 to Thompson". For a decoder, pay them $50k yourself and you can even make it free-as-in-beer for your users, too.
Your description glosses over a number of pertinant details. Unisys and IBM were both issued patents covering a compression algorithm commonly used in GIFs (U.S. patents 4,558,302 and 4,814,746, respectively). This presented a problem for those who wanted to deal in the patented algorithm because it meant there were two organizations to deal with to comply, not just one. Generally, as patent law is "harmonized", it becomes easier for corporations (who hold the vast majority of patents) to prevent the spread of Free Software worldwide.
It doesn't prevent anything of the sort. It just makes another patent law to be ignored. I provided plenty of examples which totally ignore the patent "issues" - and ignoring two patents is just as easy as ignoring one.
The one-time fees start at US$50,000. How many of the programs you listed have paid the appropriate one-time fee to Thomson so they can legally distribute their programs in countries that honor U.S. software patents?
None, AFAIK, which was precisely my point. Open source developers just ignore the patents completely anyway: why exactly will creating more patents for them to ignore make any difference? Both of those cases are patented in Europe already: how will this change?
RedHat received some press for removing MP3 decoding software from its GNU/Linux distribution. RedHat GNU/Linux is a very popular GNU/Linux distribution. So that means a lot of RedHat users lose unless they obtain a patent license or infringe upon Thomson's patent.
And RedHat knows very well their users will do the latter, which is why they didn't pay the $50k.
Whatever patent laws are passed, I'm sure open source developers will continue to ignore them completely, as they have done so far.
RMS tells the story of a compression algorithm that was about to be used in a compression program but couldn't be used because a patent had been issued covering the very same algorithm. RMS learned of this patent a week before that program was set to debut. I doubt that is the only time a program "died before it was born" (quoting RMS' description of the program that implemented the patented algorithm).
This is the only compression algorithm in existence? In the end, he changed to another algorithm, and gzip was born. Nothing really "died".
But the fact is, Red Hat takes the MP3 issues seriously enough that they stopped providing MP3 software.
They had a choice between doing what they did, or paying a one-off $50,000 license fee. They chose the former, because they knew their users could reenable MP3 playing with a few mouseclicks.
LAME apparently had enough concern that they emphasize that LAME is not an MPEG encoder.
No. It is an MP3 encoder; it's just another recursive acronym, like Gnu's Not Unix or Pine Is Not Elm. It's not the reference encoder published by MPEG (the group which developed the standard), it's a rewrite. They aren't trying to deny LAME being an MP3 encoder, just to have a geeky project name.
I think at some point the ROI changes to a positive number when it comes to pursuing infringers... and that's why you don't see large device makers like Sony using MP3s without a very costly license from Fraunhofer.
Very costly? For MP3 player software, it's a one-off payment of $50k. RedHat could easily have bought one, rather than disable MP3 playback. For an MP3 streaming service, it's 2% of revenue. Neither of those qualify as "very costly" for any commercial product. For hardware MP3 players, it's 75 cents per unit - again, small change. For an extra $10,000 (again one-off) they'll provide the software, too.
Actually, the thing that patents have done the most to hamper is encryption.
Yes. It's really difficult to find a copy of PGP/GPG, or RSA. *cough*. Most of the "big" algorithms are free; SHA-1, DES, AES and Serpent, courtesy of the US government and my old crypto supervisor, RSA (as of 3 years ago - free for non-commercial use prior to that). Diffie-Hellman's patent expired back in 1997, leaving - what? IDEA, RC5.
You can write a fully-compliant SSL implementation without any patent issues. Ditto a PGP-style encrypted email system. Where's the problem? Research wasn't affected - even the patented algorithms were published, with licenses only needed for commercial use.
And the reality of the gif patents is that Unisys just hadn't gotten around to extorting money from the authors of the programs you name yet.
"Yet"? They'd better hurry up: they have 6 days left! (In the US, that is; their European, Canadian and Japanese patents on the same algorithm will probably expire some time later...)
xv died because Unisys started demanding money from the author for its gif support.
Not "gif support" - just support for LZW compressed GIFs. Uncompressed GIFs are fine, as are the other bitmap formats. (For a website, for example, this is fine: mod_gzip or Tux's.gz support will provide better compression anyway.) Then, next week, the author could put it back anyway...
Incidentally, this is not a bug in the US patent system compared to Europe: Unisys has LZW patents in the UK, Canada, France, Germany, Italy and Japan already, under the existing patent law.
Not really. NASA has already launched test probes that use Ion drives to great success. But the electricity was derived from their existing processes, so I guess you are correct in that this is the first attempt to marry the capability to generate lots of energy (nuclear) to much larger Ion drives.
First to combine nuclear+ion drive, yes. Cassini-Huygens used nuclear (Radioisotope Thermoelectric Generator) power for onboard systems, though, and Deep Space 1 used (solar-powered) ion drive.
Spacecraft aren't the only systems powered by RTGs; medical devices are as well.
The MP3 patent's licensing terms don't even prohibit legal Free Software implementations - you pay a one-off licensing fee, and you're fine. Thereseemseemtobeplenty.
The reality is, those patents haven't killed MP3 or GIFs. If anything, it's Ogg Vorbis and PNGs which are an endangered species - not from litigation, but disuse. (The MP3 patent, by the way, is Fraunhofer's - a German group, not US.) Patents or no patents, MP3 and GIF are still the format for that application, and supported by plenty of free/open source programs.
So the future for free software is (and this would be fine with me, except that it doesn't promote either fair use or interoperability) a ghetto of free formats like Ogg.
Unlikely. As I said, we've had MP3 and GIF patents for years without this result - the former from a German company, not a US one - without the result you predict. Why would Europe adopting the US system have this result, when it hasn't happened in the US?
That must have been a long time ago. The advertising clause went away not all that long after the USL vs. BSDI suit. Definitly before the last Linux TCP rewrite, and probbably before the one before that.
It was a long time ago - not only before the last Linux stack rewrite, IIRC this is how it came to be written in the first place!
TCP stacks are hard and it's a shame Linux had to write another (at least 3 times!), but it is kind of nice to have a second reasonable quality implmentation that one can read through.
It is a shame to see duplication of effort - OTOH, those two rewrites weren't forced by any licensing issues, but an engineering choice, so presumably they were considered a good idea by somebody. As a result, instead of one current open source IP stack, there are two - both being actively maintained and developed, and both exchanging ideas. (Sad the Linux one can't be propogated into the *BSD trees, but you can't have everything...)
Slashdot Mirror
According to your link, he ended up with a broken back and lacerated arm from the attempt. Better than drowning, but hardly a ringing recommendation. (For underwater use, the ability to blow the canopy and detach from the seat without triggering the ejector charge would be much more use than any ejection mechanism...)
People stop using that product, and switch to one of the many rivals.
Your claim that redhat could easily have paid $50k can be easily voided with the counter-claim that this amount of money is 1 year's salary for a redhat developer, who can do lot more than just package an mp3 player in 1 year.
Yes, RedHat chose to spend the money some other way. So what? That doesn't "void" anything.
And mp3 is one of the most benign examples of software patents. Try getting a license to develop a player for MS MPEG4 on linux/*BSD or even MS windoze. Same for Apple. Let us know if Bill is happy with $50k.
WTF do you mean "MS MPEG4"? If you mean MPEG4, forget MS: it's the MPEG people you need to talk to. Bill isn't involved, nor is Apple - you would normally buy the license from "MPEG LA", on standard and open terms. (Tivo, for example, would pay $0.50 per unit shipped.) The prices are published - lots of legalese, but AIUI commercial distribution of an MPEG4 decoder - for Windows, FreeBSD, Linux or anything else - would cost $0.25 per unit (up to a maximum of $1m). Under 50,000, it's royalty free. (Microsoft and Apple would presumably opt for the flat-rate license.)
So, $50k doesn't buy unlimited distribution rights on MPEG4 software. Let us know how you get on developing your own video codec for $50k - Ogg have just about completed a commercially-usable audio codec, they might achieve video one day - but not on that budget.
You mean, in a "patent-filled world", you need to get the patent-holder's permission to use their patented format - iff they choose to impose licensing requirements. (We'll ignore for now the fact you can't patent formats per se...) What is so wrong with that? As I said earlier, if you want to use somebody's invention in a way they don't like, invent your own damn product. Just like copyright: the inventor decides the terms.
About using using wine to play movies on linux, try the alternative of distributing mplayer's win32 port (oh ya, mplayer does play these patented MS media formats natively on win32) inside US and wait for your day in court.
Winamp seems to manage it OK - what does mplayer do wrong to cause a problem?
No - I was replying to a post which claimed more money would be made without patents (the scenario about Real making their app open source, and making money from installing servers for media sites.)
Clones are the real competition. If you need to play RealPlayer content, a video player from Apple or Microsoft doesn't do you any good when it can't play that content. You're locked into using RealPlayer. Since Real has a monopoly in this market, they can charge any price, no matter how exorbitant. And as the consumer, you can't go to a competitor offering a better price -- the competitor's product (while similar) doesn't meet your needs. You're stuck with the monopolist.
No, that's exactly how Apple and Microsoft got into the market. They didn't try to compete using their own Real clone, they built their own media client+server combo and started pushing that. Nobody was "stuck" with anything - they just had to have the appropriate (free) viewer for that site. The viewer doesn't have a choice of client software - but then, that isn't what anyone was competing for. The client software is just a loss-leader - the competition (and choice) is for the sites.
Yes, the monopolist will spend their money too. That doesn't mean it's a wash for society. Monopoly rents extract substantial costs from society at large (in various ways) for the benefit of the monopolist. When the monopolist spends that money, the value to society and the economy doesn't match the costs that were extracted. This is exactly why economists call it "inefficient" and are very wary of monopolists...
I agree true monopolies of a whole sector are bad - I don't think it's happened in computing via patents. Real don't have a monopoly on video or audio streaming. Google (with the PageRank patents) don't have a monopoly on searching. Microsoft have a [near] monopoly on desktop OS and office suites - but that has nothing to do with patents. A patent doesn't normally confer a monopoly on a whole market (which would be the harmful kind of monopoly economists oppose) - just a monopoly on a particular device. If Boeing patent some new wing design, or fly-by-wire system, it doesn't give them a monopoly on aircraft - it just makes their aircraft better, and forces competitors to develop something different to compete. (Or take more government subsidies, in Airbus's case - you know Concorde cost about the same per unit as the Space Shuttle?) Without patents, Airbus could sit and wait for Boeing to invent something, then produce a cheap clone - everyone loses.
Looking at the harmful monopolies of the past, none involve patents AFAICS. Most of the ones I'm familiar with were actually created by government mandate (in Britain, the government telephone company, railway and health service were all taken from their creators this way - and the resulting mess sucked badly.) Where have patents given any true monopoly?
Would you consider the market for CD players to be a monopoly, by the way? Would you change your answer if you knew they were all subject to a patent licensing fee to Sony and Philips, who co-developed CDs?
The sensors are shared, of course, but the control systems are still largely isolated. In a tie-up with an earlier /. story, they often use QNX as the operating system, which is capable of keeping a b0rked logging daemon very firmly under control even on a shared CPU ;-)
(The ones I've seen are all physically distinct boxes, but I agree they'll probably be integrated as long as GM is very very sure it'll work properly.)
You talking power steering? That is a hydralic feed back loop using a pilot valve in the steering box. Turning the control wheel moves the steering linkage that is built with a bit of flex in it. Deflection of the linkage moves the pilot valve. The pilot valve shuttles the control valve and moves the front wheels until the defection in the linkage centers the pilot valve. Loss of pressure in the control section or the pilot section of the system places all the load on the linkage so even with the pump dead you can still steer the car. No digital data to record.
That's why I said system rather than circuit ;-)
Having said that, the manufacturers are planning to move to electric power steering in place of hydraulics - big weight and cost savings, apparently. And more digital data for Little Brother to log! (See the earlier story about 42V cars. I'd prefer 48V - made of -24,0,+24 - but they didn't consult me :P)
You could say exactly the same about any purchase. The reality is, spending money is not a cost to the economy - it's exactly the transaction the economy is built from.
More importantly, the patent bars competition, which lowers prices -- another cost to the economy.
It doesn't bar real competition. Real has QuickTime and Media Player, MP3 has Ogg Vorbis, Cisco has Nortel, Netscape has Apache and Microsoft. Patents bar clones - you can't just go and write your own RealPlayer, assuming they've patented that, but you can go and write your own video player, just as Apple and Microsoft did.
On the other hand, if nobody has to pay monopoly rents to license a patent, that's real money they don't have to spend, and they'll certainly find another way to spend it.
The money still gets spent, just in a different place. No net gain, you're just taking money from Thompson and giving it to another company instead. What about when Thompson spends the money they get in license fees? What about the development it would otherwise have funded - where do you think MP3Pro came from?
Yes. We all know how an open format spec will be used by everyone involved. *cough* deb RPM ebuild *cough*. Look at the different versions of HTML, where few sites bother to conform to the specs, and few browsers bother to render to them. Do you really think video and audio would have avoided that?
All the companies except Real might benefit, perhaps, but without their proprietary format, other people would be using their free player software with someone else's free or cheaper server. Real take a hit on the player (since it's a loss leader) then get screwed on servers (other companies with lower overheads - no player of their own - can undercut them).
The formats have no particular technical advantages or disadvantages, they're just different.
Not really true on either count. QuickTime and Media Player both had to offer better features to take market share from Real: because they couldn't just clone it, they had to compete on merit. Also, Real are now working on a unified server, meaning the player and format will be the viewer's choice. Looks to me like the best of both worlds: free players, and a choice of format.
The big ones are all the airborne equivalent of buses. From the right angle, 747s are pretty nice, but if you go for smaller jets, the Cessna Citation X is pretty nice IMO. Or the BBJ (Boeing Business Jet - a variant of the 737 Next Generation airframe), especially from the interior.
I tend to prefer the ones I've flown myself, but that's probably just my inner control-freak talking. Ironically, the AFB I first trained at is now the civilian airport I use most often - it still feels a bit weird flying out of there in a 737. Not being military, the flights are less exciting, but most of the passengers probably prefer that ;-)
The systems are totally separate, in fact. The car designers seem to follow the sort of Keep It Simple philosophy of good programming. The brakes, steering and airbag are absolutely critical, so you have a very very simple and totally dedicated system for them:
This circuit triggers the airbag whenever the car experiences an impact greater than X. You test the ass off it to make absolutely sure it does exactly that.
That circuit relaxes the brakes for X ms, whenever the wheels are slipping more than Y. Test the ass off it.
This system applies this much extra force in the direction you're turning the wheel. It gets tested to hell and back too.
Then there's a solid state recorder (usually sitting under the front passenger seat) which records whatever information it's fed from the various sensors. It's simple and designed to be solid, but it's deliberately not safety critical.
Ditto the engine management chip. Designers know customers will often swap those for an alternative to get better acceleration or whatever - so it only has access to the fuel mix controls, gear selection and so on. Whatever that chip does, it doesn't have any access to the brakes, steering, airbag - it can't affect the safety systems, the worst it can do is stall the car. (Or cause bad engine wear with dumb settings, but that doesn't kill anyone...)
If only MS put this kind of separation into their software: why exactly does a word processing document have the ability to email itself?
Exactly that scenario happened anyway, until QuickTime and Media Player came along to muscle in on the market. Why would making the format patent-free have changed anything? They gave their own player software away anyway - including Linux support! - how would the absence of patents strengthened this control?
Eventually their prices would've had to come down because of competition from others, which they could do nothing to prevent. But, until others became credibly well-versed in the technology they created, they would've had their monopoly.
A monopoly patents could have strengthened - making life better for Real - how could Real's patents have weakened this control, or hastened the competition?!
You do if it's leased. A patent - like a lease - confers "ownership" for a finite time.
Also, it's probably costing the economy billions upon billions of dollars for these people to have their monopoly. It's inefficient and state granted.
Ah, the RIAA argument: it's "costing" billions, because you think billions might otherwise be made. Nice try. It's creating money for the owners. Perhaps some other system would make them more money, perhaps not - but you cannot claim it is "costing" the economy anything.
Yeah, and you are technically infrining on several patents using half those programs.
Such as? I'm pretty sure Real's player doesn't infringe Real's patents. Ditto Quicktime and Media Player. You thought I meant I was running some other player to view content in those formats? (Wine.)
The only reason those programs exist is because their existence serves other purposes of the creators of the format, and there's nobody with deep enough pockets to be worth suing.
The programs I use were created by the same people as the format they play - I'm infringing nothing.
And, just how much have they been actually helped by their patents? It's not as if ReplayTV didn't also exist. Also seems to me that Real's patents have actually hurt them. They'd have widespread adoption if it weren't for patents. They'd be able to make a ton more money installing and setting up servers for people.
Real do have widespread adoption, and they do make money on servers. They have competition from Microsoft and Apple, doing exactly the same thing, which is the only issue they face. So, you're claiming removing patents would be good, because it would give Real a monopoly patents currently prevent?!
That's exactly how it's supposed to work. They developed a system, and now have a monopoly on their system. You want one? Fine - design your own.
Yeah, and that's why video sucks so badly under Linux.
Funny, it's fine for me, even Windows Media files.
Doesn't help developers ... the only people it helps are people like Microsoft and Apple.
Microsoft and Apple are developers.
They lock out small competitors while the big ones quietly cross license. It's extremely damaging to competition and represents an unacceptably large barrier to entry.
Unacceptable to whom? Plenty of startups have managed to create something worthwhile, and benefit from patent protection for it. (Tivo, Real.)
And, the developers don't get paid a cent. They get their salaries sure, but no money for the patent.
You think their salary grows on a company money-tree? Where do you think the patent royalties go, if they don't go to the company owning the patent? What do you think boosts the company's stock price, making the developers' stock options worthwhile?
Patents ... your government working hard to prevent any upset of the people who line the pocketbooks of its politicians.
Trying to prevent any upset of the economy. Sounds like a good idea.
No. If, for example, someone turns up with another US patent covering RSA (on which the original patent has expired), you could simply cite the expired patent as prior art. Goodbye second patent.
Internationally I think it's a different story--a US patent for LZW is expected to expire soon, but other patents covering the same algorithm abroad will expire later. Thus this algorithm will be simultaneously covered by patents and not covered by patents (assuming the LZW patent is not renewed) in countries which support software patents.
That has always been the case, thanks to different patent systems. The change the original article was objecting to removes this conflict (at least between EU countries): instead of Unisys having a different patent on LZW for Japan, the US, Canada, the UK, France, Germany etc, it's reduced to Canada, US, Japan, EU. ISTR there's also a requirement to patent nearly simultaneously, so you couldn't get a US patent, then get an EU patent on the same invention just before the US one expires.
To be fair, we don't know how much RedHat has budgeted for this (assuming they would want to engage in purchasing patent licenses at all) and Thomson might choose to not sell RedHat a license.
I don't think they can: anti-trust laws prohibit that kind of discrimination. (This is how MS got into hot water: trying to use Windows license pricing as leverage against OEMs. That's illegal under the Sherman Act - if you sell Dell 10,000 OEM licenses for WinXP at $25 each, you can't charge HP $50 each for the same number.)
Also I would wonder if the decoder license is still incompatible with Free Software licenses because of something in the text of the patent license.
The GPL is clear about that: it doesn't try to override patents, or force licensing. Just like encryption: the existence of anti-encryption laws doesn't prevent GPLed crypto software. Those laws may prevent some people using the software, but that doesn't affect the GPL.
I agree. That doesn't stop it happening...
I'm guessing RedHat agrees and that is why they removed MP3 software from their GNU/Linux distribution.
Yes: they took the slightly cheaper of their two legal options - removing MP3 players to save the $50k.
It's disappointing you aren't citing any reason for disobeying the law other than your immediate desire. Nothing about challenging bad laws, raising awareness of a system that prohibits competing on merit, or working on a better solution--any of the reasons that might encourage people to agree with why you advocate infringing patents. You appear to advocate infringing patents just because you want these features and you're not willing to consider the long-term ramifications of these choices.
I'm not advocating it, just stating it currently happens and is unlikely to stop because of the proposed changes in the law. The argument earlier about patents eliminating free software is like claiming a ban on crystal meth lab equipment would kill medicinal marijuana use...
No, and fortunately for this case another algorithm was not patent-encumbered. As patents cover more ideas more algorithms will be cut off from implementation and the strategy that led to gzip will be cut off.
No. By definition, all the current patent-free algorithms will remain so. New algorithms will be developed; some patented, some not. Meanwhile, the clock ticks away towards existing patented algorithms becoming patent-free - nothing will ever become patented.
For programmers in countries that observe software patents, there's no known way to legally encode or decode MP3s without acquiring a license from Thomson.
Or using an existing licensed tool to do it. RedHat could have taken the opportunity to remove the whole problem, at least for decoders: pay the $50k, and mpg123 and co are legit.
But you need not remind us of your solution--ignore the law because it gets in your way.
That's not "my solution". Someone claimed a whole category of free software would become illegal, hence extinct, under the new rules; I pointed out it is already illegal, and shows no sign of becoming extinct because of it.
I can only hope that any developer who doesn't have the US$50k or more to acquire an MP3 patent license somehow comes up with the money to defend themselves should they be sued for patent infringement.
I doubt they'd need it. If sued, a defense would be a waste of money: they'd just end up paying a big lawyer's fee on top of the license fee they should have paid in the first place.
The article said the drunkenness did happen before, and there isn't much of an "idea" involved in hurting someone while drunk. More important is where did the drugs come from?
The fact that these people were intoxicated is insignificant: there were heroin addicts and drunks before television, but not this kind of crime.
I find that hard to believe. Murder is not an "idea", it's a violent act, which was committed as the result of the intoxication. Are you really trying to claim an aggressive drug addict would have had no idea how to kill or injure someone until he saw a TV program?
Now that the country of Bhutan is becoming corrupt and immoral just as our culture is being exported to it, you claim that television is a "convenient scapegoat"?
Yes. The article said it's one of many things which changed - they also have a drug and alcohol problem (which does result in violence in Western cultures - even among those with no access to TV).
As a counterpoint: Afghanistan had no TV either, until recently. The presence of Al Queda and multiple large armed factions in constant conflict suggests they still managed to be more violent than Bhutan, despite the absence of this "terrible device"...
A crucial part of mental development is learning to distinguish fantasy from reality, recognising that most of the programs on TV are the former. Western children develop this at a very early age; these people's isolated religious lives seem to have stunted this development. In time, it will come - they just have to be patient, and handle the adjustment carefully.
TV is indeed a powerful influence - like power tools, you should keep a careful eye on young kids with it, and gradually they will learn how to handle it properly. Try to take shortcuts, and you'll end up with red stains all over the place. That doesn't make either item a "terrible device", just one you need to learn to handle.
How? SSL and co have been around for years.
What's wrong is that secure digital cash based systems aren't being deployed right now, instead of when the patents run out. That is a huge cost born by everybody.
They ran out years ago, and secure digital cash systems are already deployed.
Yeah, by people who can afford the infrastructure it takes to have lawyers and licensing agreements.
i.e. any software company. Yes. So?
It prohibits using the algorithm for me,
How? Use it yourself, and they will never know, let alone care or sue. Start distributing it - i.e. competing with the paying users - and they'll start caring, and rightly so.
and for a lot of other people. I'm unemployed, and writing encryption software right now. I can't afford a lawyer. I can't afford licensing fees, and I'm not going to burden myself with having to do those things if I should ever start selling my program.
So, you aren't willing to share any of the profit you make with the guys who developed the system you're using?
So, each individual user will have to purchase the license separately in order to use the program? That sounds workable.
This reminds me of the complaint about the GPL being "viral". Either comply with the license, or write your own damn algorithm. If you want to distribute an implementation of Thompson's MP3 algorithm, you'll have to pay them.
Patents are one of the largest stifling forces in the software world today. They do nothing to encourage innovation.
How do you justify that claim? MP3 was developed by a company in exchange for the license revenue they now receive. Ditto almost every other major format, from CDs to MPEG-4.
They are merely weapons for corporations to use against one another in the marketplace.
Bollocks. They're a source of revenue for the developers who create the systems in question. They are not "weapons", except in precisely the way they were intended: to lock out competitors for a limited time. Develop a new system, and you can either license it to others (making money from the patent licensing fee) or exploit it yourself (making money from your temporary monopoly). That's exactly how the system was intended to work from the outset.
Yes, there are some cases of people using patents to prevent an invention being sold; ISTR Macrovision did this (patent a Macrovision-defeating circuit) to keep anti-Macrovision devices off the market. That's a sort of abuse of the system, but hardly a big problem.
I would disagree there. Where is it stipulated the license applies only to binary distribution, not source? Whether the program is distributed in binary, source or interpretive form, you need a license for that program. Also, RedHat distributed a binary decoder (which would be covered by the $50k license) and the source code. Compiling the source code may be a legal problem for end-users in theory, but not for RedHat.
You're right, the LAME acronym is meant to be cute, because technically it's not an encoder. It's just source code that still needs to be developed into an encoder by compiling it or incorporating the code into a different package.
Why do you think an encoder in source form is "not an encoder"? Legally, it is. In practical terms, it is. As a source RPM, you can even (IIRC - it's a while since I used RedHat) install it like a binary one, it just takes longer.
You are over-simplifying the matter greatly and in dangerous ways to obfuscate the potential for real harm from software patents. That there is a case where the grey areas have been exploited to do an end-run around the licensing strictures does not provide significant evidence that people should be able to patent what is essentially math in the first place. This is a change to the traditional role of the patent that is neither necessary nor desirable in my view.
What change? The patents we've been talking about were issued under the existing rules. There's no "grey area" being exploited - the developers just ignore the patent issues. They ignore the existing rules, why would they start paying attention to changed ones? A few companies, like RedHat, pay enough attention to leave customers to download their own copy, rather than bundling one. Hardly a major problem. There's no change coming in this respect.
I do agree mathematical techniques shouldn't be subject to patents. (Patenting RSA - which essentially consists of a single equation about 7 characters long! - is insane.) I'm not so sure when it comes to complex systems like MPEG-4: why shouldn't the developers be entitled to some payment for it? Developing that system took serious work - anyone claiming to have developed an MPEG-4 codec without reference to any of the MPEG's work is obviously lying.
Quite. If you look, you'll notice one of the murders being blamed on TV was this one:
Yep, obviously all TV's fault. Or this one:
The first one reminded us alcohol abuse was a problem before. We have two murders: one by a guy on drugs, the other by a drunk - how is either of these TV's fault? Did the first guy get his heroin via the TV? Did the other get drunk (an existing vice, as the article points out) from a program about alcohol?
TV is a convenient scapegoat. Dealing with the reality - that these people have led very sheltered lives so far, and are struggling to deal with life now some of that isolation has ended - is much harder.
None, they just had to give some money to the developers if they were making commercial use of it. "You can use my program if you give me some money for it. It's free for non-commercial use, too." What's wrong with that?
We might've had those technologies widely deployed as much as a decade earlier if those patents weren't there.
It's been widely used for years anyway.
A lot of people here seem to have missed something. A patent does not prohibit using an algorithm, it just allows the developer to set a price for it. Unless you're demanding that you get it "free as in beer", there's no problem: just pay the license fee! (It doesn't even prevent GPLed implementations, AIUI: your implementation may be freely copyable but not include a patent license.)
You can produce an entirely legitimate GPLed MP3 encoder - just point out "use of this software requires a license fee payment of $2.50 to Thompson". For a decoder, pay them $50k yourself and you can even make it free-as-in-beer for your users, too.
It doesn't prevent anything of the sort. It just makes another patent law to be ignored. I provided plenty of examples which totally ignore the patent "issues" - and ignoring two patents is just as easy as ignoring one.
The one-time fees start at US$50,000. How many of the programs you listed have paid the appropriate one-time fee to Thomson so they can legally distribute their programs in countries that honor U.S. software patents?
None, AFAIK, which was precisely my point. Open source developers just ignore the patents completely anyway: why exactly will creating more patents for them to ignore make any difference? Both of those cases are patented in Europe already: how will this change?
RedHat received some press for removing MP3 decoding software from its GNU/Linux distribution. RedHat GNU/Linux is a very popular GNU/Linux distribution. So that means a lot of RedHat users lose unless they obtain a patent license or infringe upon Thomson's patent.
And RedHat knows very well their users will do the latter, which is why they didn't pay the $50k.
Whatever patent laws are passed, I'm sure open source developers will continue to ignore them completely, as they have done so far.
RMS tells the story of a compression algorithm that was about to be used in a compression program but couldn't be used because a patent had been issued covering the very same algorithm. RMS learned of this patent a week before that program was set to debut. I doubt that is the only time a program "died before it was born" (quoting RMS' description of the program that implemented the patented algorithm).
This is the only compression algorithm in existence? In the end, he changed to another algorithm, and gzip was born. Nothing really "died".
LAME apparently had enough concern that they emphasize that LAME is not an MPEG encoder.
No. It is an MP3 encoder; it's just another recursive acronym, like Gnu's Not Unix or Pine Is Not Elm. It's not the reference encoder published by MPEG (the group which developed the standard), it's a rewrite. They aren't trying to deny LAME being an MP3 encoder, just to have a geeky project name.
I think at some point the ROI changes to a positive number when it comes to pursuing infringers... and that's why you don't see large device makers like Sony using MP3s without a very costly license from Fraunhofer.
Very costly? For MP3 player software, it's a one-off payment of $50k. RedHat could easily have bought one, rather than disable MP3 playback. For an MP3 streaming service, it's 2% of revenue. Neither of those qualify as "very costly" for any commercial product. For hardware MP3 players, it's 75 cents per unit - again, small change. For an extra $10,000 (again one-off) they'll provide the software, too.
Yes. It's really difficult to find a copy of PGP/GPG, or RSA. *cough*. Most of the "big" algorithms are free; SHA-1, DES, AES and Serpent, courtesy of the US government and my old crypto supervisor, RSA (as of 3 years ago - free for non-commercial use prior to that). Diffie-Hellman's patent expired back in 1997, leaving - what? IDEA, RC5.
You can write a fully-compliant SSL implementation without any patent issues. Ditto a PGP-style encrypted email system. Where's the problem? Research wasn't affected - even the patented algorithms were published, with licenses only needed for commercial use.
And the reality of the gif patents is that Unisys just hadn't gotten around to extorting money from the authors of the programs you name yet.
"Yet"? They'd better hurry up: they have 6 days left! (In the US, that is; their European, Canadian and Japanese patents on the same algorithm will probably expire some time later...)
xv died because Unisys started demanding money from the author for its gif support.
Not "gif support" - just support for LZW compressed GIFs. Uncompressed GIFs are fine, as are the other bitmap formats. (For a website, for example, this is fine: mod_gzip or Tux's .gz support will provide better compression anyway.) Then, next week, the author could put it back anyway...
Incidentally, this is not a bug in the US patent system compared to Europe: Unisys has LZW patents in the UK, Canada, France, Germany, Italy and Japan already, under the existing patent law.
First to combine nuclear+ion drive, yes. Cassini-Huygens used nuclear (Radioisotope Thermoelectric Generator) power for onboard systems, though, and Deep Space 1 used (solar-powered) ion drive.
Spacecraft aren't the only systems powered by RTGs; medical devices are as well.
Ah yes, the GIF patent. That stopped any free software using GIFs.
The MP3 patent's licensing terms don't even prohibit legal Free Software implementations - you pay a one-off licensing fee, and you're fine. There seem seem to be plenty.
The reality is, those patents haven't killed MP3 or GIFs. If anything, it's Ogg Vorbis and PNGs which are an endangered species - not from litigation, but disuse. (The MP3 patent, by the way, is Fraunhofer's - a German group, not US.) Patents or no patents, MP3 and GIF are still the format for that application, and supported by plenty of free/open source programs.
So the future for free software is (and this would be fine with me, except that it doesn't promote either fair use or interoperability) a ghetto of free formats like Ogg.
Unlikely. As I said, we've had MP3 and GIF patents for years without this result - the former from a German company, not a US one - without the result you predict. Why would Europe adopting the US system have this result, when it hasn't happened in the US?
It was a long time ago - not only before the last Linux stack rewrite, IIRC this is how it came to be written in the first place!
TCP stacks are hard and it's a shame Linux had to write another (at least 3 times!), but it is kind of nice to have a second reasonable quality implmentation that one can read through.
It is a shame to see duplication of effort - OTOH, those two rewrites weren't forced by any licensing issues, but an engineering choice, so presumably they were considered a good idea by somebody. As a result, instead of one current open source IP stack, there are two - both being actively maintained and developed, and both exchanging ideas. (Sad the Linux one can't be propogated into the *BSD trees, but you can't have everything...)