"I think you're wrong that individuals will be forced out; remember that pursuing an individual past a C&D is also quite expensive,"
Except that it is not usually necessary, as many developers have discovered - see gnu.org,ffii.org and other sites for examples. Who would continue to develop software that has the ever present threat of legal action hanging over it?
"It is very rare for massive numbers of infringing individuals and small companies to get sued,"
That's probably because it's rare for "massive numbers" of individuals and small companies to simultaneously release software competitive enough with a commercial organisation's product to make it worth their while taking action. So ignoring a hypothetical and absurd scenario, what you're saying is that it's okay to have software patents because your project will only be destroyed by wealthy competitors if it gets too popular for their comfort. Great.
"The end result is that large companies with large portfolios and small groups of individuals with ad hoc agendas will be the centers of software development... in other words, nothing will change."
I don't know what you mean by "ad hoc agendas" but things have already changed - commercial and FOSS projects have already been closed down or aborted by the threat of patent trouble and whole areas of development have been closed off or are in the process of being closed off to all but the holders of large patent portfolios.
It is not just a matter of what might or will happen but what has already happened, is happening now and is getting progressively worse. It is unethical, detrimental to progress in the sciences and the arts and economically unjustifiable to continue granting software idea patents. What justification is there that there should be such "centers of software development" anyway and what justification is there that the existence of those centers should be enabled and reinforced by a legal right to exclude everyone else from pursuing those activities instead of by their own innovative excellence and productivity?
Actually I cut 'n' pasted it out of an email and it seems that the slashcode reproduction mechanisms mistakenly thought I intended an HTML tag. So I would say it was a non-obvious (to me at the time) quirk in the manufacturing process rather than any inherent non-triviality in the 'invention' itself.
"People need to either accept the idea of all patents or no patents. You can't legitimately have it both ways and be consistent."
What does that mean? All areas of human endeavour should be patentable? Maths, Science, Art, Literature... ? Or perhaps you mean all software patent appklications should be accepted by the patent offices? Either way, your statement is nonsensical.
1) Software patents are bad because they are unethical, unnecessary and arbitrary restrictions on the right to freedom of expression. There has never been a sound economic argument for them - quite the contrary - but even if there were it could not justify the deleterious effects on individual civil liberties of patents on pure ideas. Since when did the desires and conveniences of large commercial organisations and the financial interests of lawyers trump the fundamental rights of man?
2) Why then are there special protections for integrated circuit designs? Software ideas are not software are not software ideas - software is the expression of the idea and has copyright protection. Abstract ideas are now being patented, so excluding any expression of them at all and that is simply wrong, no matter how clever or original they are. Too many people seem to have forgotten that there is more to society and justice in society than the protection of the interests of lawyers and large commercial organisations.
3) Name one that is not simply the algorithmic expression of a mental process or idea made capable of computerised speed up by expression in a computer language. What is wrong with trade secrets in these exceptional cases anyway - if a proprietary software package contains the expression of an idea so fiendishly clever and complex that it is 'not obvious' it will not be obvious how to 'steal' it or easy to do so either.
"That said, it seems to me that the biggest reason a lot of geeks don't like software patents is that it is inconvenient for them. Every rationalization that disallows "software" patents but allows other types of patents has been uniformly weak and inconsistent. If you think patents are bad, then ALL patents are bad, not just the ones you wish you didn't have to abide by."
Again you are trying to get us to equate software idea patents with all patents and your assertion that the arguments against software patents have been "uniformly weak" is utterly risible. From Salin to Stallman, they have been uniformly strong - as strong as any arguments against the patentability of ideas in maths or art or music or literature and geeks don't like them because they are an outrageous infringement of their fundamental human rights - not just because they are an inconvenient nuisance. Yes they are 'inconvenient' but how does the convenience of large companies and lawyers outrank the freedom to engage in the arts and sciences of individual citizens? I know I don't regard fundamental human rights as mere conveniences, to be abandoned and abolished on the basis that to do so would make life easier and more profitable for large corporations and their lawyers.
Good grief! you're arguing for the exact reverse of that which is right, just and equitable. Algorithms are mental processes, mathematical ideas which when enacted in software cost next to nothing to manufacture and distribute. To use the patent system to exclude others from even using them is an assault on the very foundations of natural justice, not to mention economically unjustifiable. No-one ever sold a bare algorithm in a shrink-wrapped box - in case you hadn't noticed, each saleable software product is likely to contain many, many algorithms and ideas, and for that reason and others it is appropriately protected by copyright.
I don't think I really care whether the chemical companies consider patents on their expensive to build and operate production processes to be crazy or not - it's up to them. I do however think your idea of allowing ideas in mathematics and computer science to be patentable to be both crazy and evil.
I have never committed a crime on slashdot before, but I shall do so now, by manufacturing and distributing a patented invention:
echo -n $'__________\r' for ((i=0;i10;i++)) do echo -n "#" sleep 1 done echo
The patented progress bar, enacted in the shell in a few lines and all protestations that this one is trivial and should have been excluded by the patent office are meaningless unless you can describe specific criteria that the patent office bureaucrats can follow to allow them to discriminate. The RSA algorithm is just as trivial and yet I have heard many people mistakenly claim that it deserved a patent for it's originality and cleverness - but that cleverness was all in the maths - the idea, not the algorithmic expression of it and so a justification of patentability of software ideas is a justification of patentability of mathematics. An outrageous position.
This is the 2nd feeblest, most cripple minded argument I have ever seen as a justification for software patents. For a start, you've made the very stupid mistake of lumping everything together under the term 'intellectual property' and then gone and waffled on about copyright duration - you should've been modded offtopic, not insightful!
As for "having to cope" with several sets of rules, well if you can't cope with a handful of different languages, national bureaucracies and patent application procedures, my heart just bleeds for you but how is your marketing and distribution going to fare if you are so poor at adaptation?.
Perhaps you could just hire some staff that are up to the job rather than pushing for laws that favour big companies with entrenched monopolies, destroy the business success chances of small companies and individuals and are an unwarranted assault on freedom of expression anyway?
"I've never really understood the practicality of software patents...Is there a point to software patents?"
Yes - they extend the enervating, tentacular reach of Lawyerdom yet further into the lives of ordinary people going about there ordinary business. Law and lawyers were established to protect the freedoms and rights of men from the ill effects of overweening military and financial might, to serve the public interest and maintain some semblance of justice in the World. The purpose of law has long since been forgotten though, and nowadays lawyers and legislators serve only themselves and those who offer to enrich them the most.
Not so - they are less of a target for patent parasite companies but sitting ducks for monopolisitically inclined companies seeking to exclude competition.
"Patents still expire in about 20 years, unlike say copyrights that will last longer than you will".
Copyrights don't prevent other people from expressing fundamental ideas and techniques in their own, independently created software. Copyrights don't stifle innovation in whole areas of research and development. Copyrights don't generate disincentive uncertainty. Copyrights don't encourage and enable monopolistic and unethical practices in business (not to the same extent anyway).
Software idea patents do all those things and more and it is feared that Microsoft is only holding back on it's assault on free/open source software until they manage to extend software patentability to Europe - if they had already begun attacking, it would have severely damaged their chances of getting the legislation they seek.
"I'd be curious if anyone can suggest a good rule for eliminating obvious patents."
It's very simple: 'Software' ideas should never have been made patentable in the first place. Look at the monitor in front of you and ask yourself this: "Is this a general purpose electronic computer I see before me or is it just another consumer appliance?" Are you free to use it as the invention it was originally intended to be or have large corporations now almost managed to metamorphose it into just another consumer appliance - a box into which you may plug only the software products that they deign to supply? Are you free to programme it as it was meant to be programmed or are there daily more and more restrictions on what code you can type in? Is this an acceptable state of affairs - do novelists and musicians have to put up with this kind of 'ownership' of the ideas of their crafts? Could 'inventions' like this one and many others like it only have been made by expert software designers or could a child have done it - or a law firm? And don't even think about the usual: "Well there are some clever mathematical algorithms that deserve to be patentable" - no there are not, they are mathematical ideas and belong to mathematics, which in turn belongs to us all. How many times need it be repeated that software is properly and appropriately a copyright protectable area of endeavour?
It's obvious because it's yet another trivial domain transfer of an ordinary practise from the real world to the internet and anyone who doesn't find this one obvious should avoid clicking on those "Test your IQ" ads since doing so could really spoil their day:
1990:
1) Check mail in morning, see targeted "Dear Customer" letter saying update to package X is available.
2) Pick up phone and order a copy.
3) Receive software update in mail.
4) Install it.
2000:
1) Go online, fire up software update tool.
2) See update to package X is available.
3) Select package X with mouse pointer.
4) Click on download/install button.
Applying for patents costs money and time - even large companies like Microsoft have to be somewhat selective about which trivial and obvious patents they choose to apply for. This is just one they let slip.
Well, they're at least an equal threat but in the case of 'IP parasite' companies, you're right - the idea is to maximise revenue from the patents by threatening commercial technology companies.
I think I understand now what motivated the USPTO to allow software patents in the first place: They saw that not every Sci-Fi novelist was capable of dreaming of warp-drives, alien beings and super-intelligent computers. Many were struggling to make a living because no-one wanted to read novels whose most exciting leaps of the imagination included one-click shopping and graphical progress bars.
This Reisman guy for example seems to have a history of trying to identify technology ideas that have a good chance of becoming common infrastructure in the near future and I suppose you could consider him the Arthur C. Clarke of technological mundanity but since no-one wants to read S.F. novels about software upgrade utilities, he would probably have starved to death in his garret.
So in an act of supreme benevolence and humanity, the USPTO, recognising the plight of Reisman and the other lesser SF novelists like him, has kindly provided an alternative way for them to make money from banal and commonplace ideas.
That's just the way it is - a patent "grants the holder the right to exclude others from manufacturing, using, selling or importing the invention". In all the patent rules and laws I've seen at various PTOs, I haven't actually seen any explicit mention of a prohibition on giving away copies of the invention but the manufacturing bit is enough anyway. Manufacturing copies of a software 'invention' is effectively unavoidable if you're giving it away.
The right of the software patent holder to exclude others from using the idea is just as sickening of course - it always struck me how outrageous it was that I could implement the RSA algorithm in a few commands piped together in the shell, and yet technically, the patent holders could legally prevent me from doing so.
Well it wasn't very hard for Thomas Jefferson to understand nearly two centuries ago (I'm sure you've read that letter) and I cannot understand how the entire legislative structure of Europe could have failed so miserably to grasp the essentials of the issue. Then again, given the UKPTO's behaviour I suspect the law schools are just handing out alternative MBAs to their graduates these days, teaching nothing about the law other than how to use it to make money.
As far as I can see, the legislation the Commission and Council are trying to push through is incompatible with Article 10 of the ECHR and is therefore illegal.
"Patent law is a purely economical law for X sake, so why shouldn't economical effects have precedence?"
Purely economical? - so everyone keeps saying but what bothers me about software patents is that they're not just economical are they? In fact if I had thought they were I would probably never have taken any interest in the issue.
"Tell me - in what way does having to cross-license their patent prevent them from profiting from it? "
It's very simple really - You get a patent to give you a temporary monopoly on the profitable use of your invention - no-one else can use yor idea in their products.
Then you are forced to cross-license with big companies X,Y and Z who can now use your idea in _their_ products and so you no longer have the monopoly you thought the patent was supposed to grant you. The advantage and marginal scope for profit that your idea initially gave you is lost.
"Big companies innovate. Little guys sometimes profit with submarine patents. Users pay a slightly higher price but still do get the innovations in the end."
Nonsense - even large companies have admitted that their own innovation is often slowed and even blocked by software patents (and let's stick to the subject of software patents - not all patents - brought up in freeduke's post). Here for example is what Alcatel said on the matter:
"Like other companies operating in the telecommunications industry, we experience frequent litigation regarding patent and other intellectual property rights. Third parties have asserted, and in the future may assert, claims against us alleging that we infringe their intellectual property rights. Defending these claims may be expensive and divert the efforts of our management and technical personnel. If we do not succeed in defending these claims, we could be required to expend significant resources to develop non-infringing technology or to obtain licenses to the technology that is the subject of the litigation. In addition, third parties may attempt to appropriate the confidential information and proprietary technologies and processes used in our business, which we may be unable to prevent.
Our business and results of operations will be harmed if we are unable to acquire licenses for third party technologies on reasonable terms.
We remain dependent in part on third party license agreements which enable us to use third party technology to develop or produce our products. However, we cannot be certain that any such licenses will be available to us on commercially reasonably terms, if at all."
And Joshua Kaplan of Intouch said in 2002:
"There are patents that come out today with hundreds of claims, unintelligible to almost anyone except the people who drew them. And yet, people who violate them jeopardize sometimes a lifetime of investment or their division or their product. That system doesn't work well to spur innovation or carry out the constitutional mandate.
Indeed, for those of you who were here this morning and listened to the people in the software industry talk about how threatening this is to their businesses, as I see it, patents today are often entrenching the established at the expense of allowing the newcomer to come in. I question today whether a Steve Jobs could start an Apple or a Bill Gates could start a Microsoft in view of the web and thicket of patents that is out there."
It is not so surprising that someone could hold and express such ill-informed and short-sighted views as you do, what did surprise me however is the depth of hypocrisy of a man who can proudly display on his website the work he has done in building on the GPLed work of others and yet simultaneously deem it necessary to 'apologise' for having to redistribute that work under the very same license that guaranteed his opportunity to do the work in the first place. May I remind you that your distaste for that license does not give you the right to breach it's conditions either, though I'm sure that's just a careless oversight on your part.
There are a thousand reasons why software patents are outrageous, and it is enervating to reiterate them over and over again especially when as far as I'm concerned the last words on the matter entered the historical record nearly two centuries ago. It also annoys me intensely that software is usually only ever considered in an economic context, as though the only purpose of writing software was in order to construct a saleable product and it was the exclusive domain of 'programme manufacturers'. Note especially what Jefferson says about the absurdity of patenting particular applications of an invention (eg. general purpose electronic computer), what he says about the fallacious concept of 'ownership' of an idea itself and what he says about what is the purpose of granting monopolies in the form of patents at all.
It matters not in the slightest whether an algorithm is obvious or not, it is a piece of mathematics and belongs to us all. If you find a new algorithm that is particularly efficient for some purpose then you are perfectly entitled to keep the matter secret. What you are not entitled to do (ethically speaking) is lay claim to ownership of that mathematical idea and prevent others from using it. Some time ago I read a comment in which it was claimed that the (unpatented) Viterbi algorithm was a pure invention, of use only in the field of digital communications and that consequently it would have been perfectly reasonable to patent it. This was obviously utter nonsense and it took me all of 30 seconds to discover that the Viterbi algorithm had already been adopted as an invaluable tool in several other fields. Sadly this is the kind of plausible fallacy often employed by pro-software patent activists and it can be very effective in convincing the innumerate and the scientifically illiterate.
Returning to the RSA algorithm - it can be understood by children, is about as sophisticated as Pythagoras's theorem and it was in fact patented by it's discoverers - or at least the first people known to have published it. That doesn't make it ethically justifiable though and it is certainly possible to patent algorithmic expressions of mathematics that is already well known, as for example Stanford University/Bracewell did with the DHT algorithm in 1987. I don't know offhand of any other uses for the RSA algorithm but if I had found one I would until recently have been legally prevented from using it. The mere possibility of that happening is enough to make software patents on algorithms absolutely unacceptable. A lot of maths research is done using computers these days and some fields probably wouldn't even exist without them (Fractal geometry - Mandelbrot and Julia sets etc. are the obvious and graphic example).
Economic arguments over the merits or otherwise of software patents are an irrelevance. I have purchased several general purpose electronic computers over the years and have used them for all sorts of things including writing programmes to implement bits of maths and communicating those ideas to others - sometimes in the form of explicit algorithms which for all I know may well have contained patented material. It is unconscionable to me that these largely non-economic activities be threatened by the existence of patents that benefit none but a handful of very large corporations and the parasitic community of patent lawyers. It is my computer and I will use it as I see fit and I will ignore the law if the law is enough of an ass to legalize exclusive ownership by others of the uses to which I might put my electronic machinery. I don't write fully fledged, saleable applications, I do not compete in the professional software market and if I did I would be well served by present day copyright law anyway. And when it comes to free software developer communities who could be said to compete in these markets, then if a company of full time professionals cannot carve out a profitable niche for themselves or keep up with
"For just one concrete example: where would we be if the basic FFT algorithms were patentable when first developed in the 60's? There have been thousands of variations in the basic algorithm itself due to the free sharing of the algorithm and the open development upon it."
Quite so, but even that field is not entirely clear of landmines. - though it's the only one I know of and I heard that Bracewell later regretted the patenting of this algorithm.
"Now, you ask the average open source advocate what s/he thinks about software patents, and s/he will be opposed to them, on the grounds that they stifle innovation."
Why not ask instead, the average mathematician, scientist or proud owner of what he or she once assumed to be a general purpose electronic computer? Just how is it you think anyone has the right to lay claim to exclusive ownership of any particular use of such a device and deny the owner of the machine his right to put it to that use?
"it's usually to copy a patented algorithm and incorporate it into an open source product."
'Usual circumstance' is no justification for allowing exclusive ownership of ideas that may and often do have only one algorithmic expression.
"the free equivalent to rsa"
Which is? - Please explicitly state an algorithm that would not have infringed the RSA patent and is also a valid algorithmic expression of the trivial corollary to Euler's congruence:
Given +ive integers a,k,k',m and primes p.q such that a) m=pq, b) kk' = 1 mod phi(m) and c) (a,m) = 1 then a^(kk') = a mod m
"Looking deeper, I don't see that it's consistent to be in favor of patents but opposed to software patents. This is because software blurs the line between a device and a description. For example, consider an integrated circuit. This is clearly a device, and hence patentable, right?"
Wrong, integrated circuits are primarily protected in Canada by the "Integrated Circuit Topography Act" of May 1993 and have similar protections elsewhere. And you are not "looking deeper" but instead making a shallow comparison and hollow analogy between software design and abstract software idea. Software designs are protected by copyright, abstract ideas are not and never should be.
Your 'drill printer' analogy attempts to further artificially blur the distinction between material design and abstract idea but it is ridiculous. Such a Trekkian replicator will still consume large quantities of matter and energy to produce the drill, a material object. Trade in material objects is fundamental to the operation of modern economies and such a replicating device would perhaps make it necessary to invent suitable protections for the designs of material objects but this has no bearing whatsoever on the question of patentability of abstract ideas expressed in software.
"this means that either all patents are valid (including software), or none are... though I would prefer a middle view that recognizes software as a distinct hybrid of an "idea" and a "device"."
No it does not, it is a non-sequitur and software is no more or less a 'hybrid of idea and device' than is a novel, musical score or work of visual art - all protectable by copyright.
"...some, such as rsa and lzw, are truly breakthroughs,..."
The RSA algorithm may seem like a 'breakthrough' to a software developer but as a mathematician I see it rather differently. The patenting of the RSA algorithm, like the DHT algorithm before it and any other piece of maths disguised as a software idea was a disgraceful act - effectively the partial privatisation of a piece of mathematical knowledge.
The patent on the RSA algorithm was particularly egregious as it is the obvious and *only* algorithmic expression of on an utterly trivial piece of number theory, a corollary to Euler's congruence. The idea that it should be a patentable entity is disgusting beyond belief and no - it does not matter that it's discoverers also call themselves mathematicians: No-one who maps some small part of the mathematical world yet subsequently attempts to fence it off for private use has any right to call themselves mathematician or scientist.
Let us remember that computers were themselves invented by mathematicians and are essential tools in modern mathematical and scientific research and I for one, strongly resent you programmers continually laying claim to 'ownership' of pieces of my subject just because you find them useful and profitable.
I do not enter your home and lay claim to ownership of your worldly possessions and I would be grateful if you would kindly reciprocate and keep your thieving hands off of my (and everyone else's) abstract mathematical possessions.
Not obvious? You cannot be serious; not only is it obvious but I've seen it at least once before (KDE???). Yet the worst thing about it is not it's obviousness or whether there's any prior art but that it's the kind of thing I can imagine a highly configurable desktop or window manager might allow a user to figure out how to do by him/herself.
It's the kind of thing that I can imagine the developers of said configurable desktop might not actually have explicitly programmed into it in the first place. I can imagine one or more smart power users figuring something like this out for themselves and telling their friends how to do it. Who is infringing on such a patent now?
It's not so much a technology patent as it is a patent on a particular *use* of a technology. Do you really want people to be able to patent the way you use things? Or use patents to prevent others from making flexible and configurable things for users who like that kind of power?
This is just the kind of patent that ought to be waking people up to the idea that software patents are perverse and egregious thieves of diversity,creativity and individual liberty.
"...I would love to play the ram feeds from the BBC..."
Unfortunately this news probably removes what little incentive the BBC may ever have had to provide their promised ogg streams. The BBC could have been a major force for good in this area but despite their size and power it seems they've surrendered to the irrational fears of third party rights holders and perversely, have even failed to make their own material freely and fully available.
Ever since I moved into a flat in which I can't get a decent EMR stream I hoped that the BBC would begin to make use of the pro bono publico technologies available to them. I tried their ram streams before I was aware of the issues surrounding such formats (and how inappropriate it was for the BBC to use them so exclusively) but the video streams were terrible and ate up all my memory before crashing X and the audio streams were of poor quality and often unreliable.
I should have known better than to expect the BBC to show any concerns about what is in the best interests of the public: They are bbc.co.uk and not bbc.org.uk after all and in recent years they seem to have made every effort to ensure that they better reflect that choice of domain name rather than the public service commitments of their charter.
Ironically, I now listen to the BBC World service streamed to me courtesy of WNYC.org and have found many of the other NPR programmes to be of equal quality to anything the likes of Radio 4 and 5 transmit. I have even found replacements for Radio 3, an especially good one being www.rozhlas.cz.
"GnuLinux" (guh-noo-lin-ux) - hardly more difficult than "Moulinex" (moo-lin-ex), the successful household appliance manufacturers and they probably paid $100,000s to a 'branding agency' for the name.
It would be even easier if the "Gnu" was pronounced "noo" (Chambers), like the real word referred to at http://www.gnu.org/gnu/gnu-history.html - unfortunately the pronunciation given at http://www.gnu.org/ conflicts with this.
Slashdot Mirror
"I think you're wrong that individuals will be forced out; remember that pursuing an individual past a C&D is also quite expensive,"
Except that it is not usually necessary, as many developers have discovered - see gnu.org,ffii.org and other sites for examples. Who would continue to develop software that has the ever present threat of legal action hanging over it?
"It is very rare for massive numbers of infringing individuals and small companies to get sued,"
That's probably because it's rare for "massive numbers" of individuals and small companies to simultaneously release software competitive enough with a commercial organisation's product to make it worth their while taking action. So ignoring a hypothetical and absurd scenario, what you're saying is that it's okay to have software patents because your project will only be destroyed by wealthy competitors if it gets too popular for their comfort. Great.
"The end result is that large companies with large portfolios and small groups of individuals with ad hoc agendas will be the centers of software development... in other words, nothing will change."
I don't know what you mean by "ad hoc agendas" but things have already changed - commercial and FOSS projects have already been closed down or aborted by the threat of patent trouble and whole areas of development have been closed off or are in the process of being closed off to all but the holders of large patent portfolios.
It is not just a matter of what might or will happen but what has already happened, is happening now and is getting progressively worse. It is unethical, detrimental to progress in the sciences and the arts and economically unjustifiable to continue granting software idea patents. What justification is there that there should be such "centers of software development" anyway and what justification is there that the existence of those centers should be enabled and reinforced by a legal right to exclude everyone else from pursuing those activities instead of by their own innovative excellence and productivity?
Actually I cut 'n' pasted it out of an email and it seems that the slashcode reproduction mechanisms mistakenly thought I intended an HTML tag. So I would say it was a non-obvious (to me at the time) quirk in the manufacturing process rather than any inherent non-triviality in the 'invention' itself.
"People need to either accept the idea of all patents or no patents. You can't legitimately have it both ways and be consistent."
What does that mean? All areas of human endeavour should be patentable? Maths, Science, Art, Literature... ? Or perhaps you mean all software patent appklications should be accepted by the patent offices? Either way, your statement is nonsensical.
1) Software patents are bad because they are unethical, unnecessary and arbitrary restrictions on the right to freedom of expression. There has never been a sound economic argument for them - quite the contrary - but even if there were it could not justify the deleterious effects on individual civil liberties of patents on pure ideas. Since when did the desires and conveniences of large commercial organisations and the financial interests of lawyers trump the fundamental rights of man?
2) Why then are there special protections for integrated circuit designs? Software ideas are not software are not software ideas - software is the expression of the idea and has copyright protection. Abstract ideas are now being patented, so excluding any expression of them at all and that is simply wrong, no matter how clever or original they are. Too many people seem to have forgotten that there is more to society and justice in society than the protection of the interests of lawyers and large commercial organisations.
3) Name one that is not simply the algorithmic expression of a mental process or idea made capable of computerised speed up by expression in a computer language. What is wrong with trade secrets in these exceptional cases anyway - if a proprietary software package contains the expression of an idea so fiendishly clever and complex that it is 'not obvious' it will not be obvious how to 'steal' it or easy to do so either.
"That said, it seems to me that the biggest reason a lot of geeks don't like software patents is that it is inconvenient for them. Every rationalization that disallows "software" patents but allows other types of patents has been uniformly weak and inconsistent. If you think patents are bad, then ALL patents are bad, not just the ones you wish you didn't have to abide by."
Again you are trying to get us to equate software idea patents with all patents and your assertion that the arguments against software patents have been "uniformly weak" is utterly risible. From Salin to Stallman, they have been uniformly strong - as strong as any arguments against the patentability of ideas in maths or art or music or literature and geeks don't like them because they are an outrageous infringement of their fundamental human rights - not just because they are an inconvenient nuisance. Yes they are 'inconvenient' but how does the convenience of large companies and lawyers outrank the freedom to engage in the arts and sciences of individual citizens? I know I don't regard fundamental human rights as mere conveniences, to be abandoned and abolished on the basis that to do so would make life easier and more profitable for large corporations and their lawyers.
Good grief! you're arguing for the exact reverse of that which is right, just and equitable. Algorithms are mental processes, mathematical ideas which when enacted in software cost next to nothing to manufacture and distribute. To use the patent system to exclude others from even using them is an assault on the very foundations of natural justice, not to mention economically unjustifiable. No-one ever sold a bare algorithm in a shrink-wrapped box - in case you hadn't noticed, each saleable software product is likely to contain many, many algorithms and ideas, and for that reason and others it is appropriately protected by copyright.
I don't think I really care whether the chemical companies consider patents on their expensive to build and operate production processes to be crazy or not - it's up to them. I do however think your idea of allowing ideas in mathematics and computer science to be patentable to be both crazy and evil.
I have never committed a crime on slashdot before, but I shall do so now, by manufacturing and distributing a patented invention:
echo -n $'__________\r'
for ((i=0;i10;i++))
do echo -n "#"
sleep 1
done
echo
The patented progress bar, enacted in the shell in a few lines and all protestations that this one is trivial and should have been excluded by the patent office are meaningless unless you can describe specific criteria that the patent office bureaucrats can follow to allow them to discriminate. The RSA algorithm is just as trivial and yet I have heard many people mistakenly claim that it deserved a patent for it's originality and cleverness - but that cleverness was all in the maths - the idea, not the algorithmic expression of it and so a justification of patentability of software ideas is a justification of patentability of mathematics. An outrageous position.
This is the 2nd feeblest, most cripple minded argument I have ever seen as a justification for software patents. For a start, you've made the very stupid mistake of lumping everything together under the term 'intellectual property' and then gone and waffled on about copyright duration - you should've been modded offtopic, not insightful!
As for "having to cope" with several sets of rules, well if you can't cope with a handful of different languages, national bureaucracies and patent application procedures, my heart just bleeds for you but how is your marketing and distribution going to fare if you are so poor at adaptation?.
Perhaps you could just hire some staff that are up to the job rather than pushing for laws that favour big companies with entrenched monopolies, destroy the business success chances of small companies and individuals and are an unwarranted assault on freedom of expression anyway?
"I've never really understood the practicality of software patents...Is there a point to software patents?"
Yes - they extend the enervating, tentacular reach of Lawyerdom yet further into the lives of ordinary people going about there ordinary business. Law and lawyers were established to protect the freedoms and rights of men from the ill effects of overweening military and financial might, to serve the public interest and maintain some semblance of justice in the World. The purpose of law has long since been forgotten though, and nowadays lawyers and legislators serve only themselves and those who offer to enrich them the most.
Not so - they are less of a target for patent parasite companies but sitting ducks for monopolisitically inclined companies seeking to exclude competition.
Copyrights don't prevent other people from expressing fundamental ideas and techniques in their own, independently created software. Copyrights don't stifle innovation in whole areas of research and development. Copyrights don't generate disincentive uncertainty. Copyrights don't encourage and enable monopolistic and unethical practices in business (not to the same extent anyway).
Software idea patents do all those things and more and it is feared that Microsoft is only holding back on it's assault on free/open source software until they manage to extend software patentability to Europe - if they had already begun attacking, it would have severely damaged their chances of getting the legislation they seek.
It's very simple: 'Software' ideas should never have been made patentable in the first place. Look at the monitor in front of you and ask yourself this: "Is this a general purpose electronic computer I see before me or is it just another consumer appliance?" Are you free to use it as the invention it was originally intended to be or have large corporations now almost managed to metamorphose it into just another consumer appliance - a box into which you may plug only the software products that they deign to supply? Are you free to programme it as it was meant to be programmed or are there daily more and more restrictions on what code you can type in? Is this an acceptable state of affairs - do novelists and musicians have to put up with this kind of 'ownership' of the ideas of their crafts? Could 'inventions' like this one and many others like it only have been made by expert software designers or could a child have done it - or a law firm? And don't even think about the usual: "Well there are some clever mathematical algorithms that deserve to be patentable" - no there are not, they are mathematical ideas and belong to mathematics, which in turn belongs to us all. How many times need it be repeated that software is properly and appropriately a copyright protectable area of endeavour?
It's obvious because it's yet another trivial domain transfer of an ordinary practise from the real world to the internet and anyone who doesn't find this one obvious should avoid clicking on those "Test your IQ" ads since doing so could really spoil their day:
1990:
1) Check mail in morning, see targeted "Dear Customer" letter saying update to package X is available.
2) Pick up phone and order a copy.
3) Receive software update in mail.
4) Install it.
2000:
1) Go online, fire up software update tool.
2) See update to package X is available.
3) Select package X with mouse pointer.
4) Click on download/install button.
Applying for patents costs money and time - even large companies like Microsoft have to be somewhat selective about which trivial and obvious patents they choose to apply for. This is just one they let slip.
Well, they're at least an equal threat but in the case of 'IP parasite' companies, you're right - the idea is to maximise revenue from the patents by threatening commercial technology companies.
I think I understand now what motivated the USPTO to allow software patents in the first place: They saw that not every Sci-Fi novelist was capable of dreaming of warp-drives, alien beings and super-intelligent computers. Many were struggling to make a living because no-one wanted to read novels whose most exciting leaps of the imagination included one-click shopping and graphical progress bars.
This Reisman guy for example seems to have a history of trying to identify technology ideas that have a good chance of becoming common infrastructure in the near future and I suppose you could consider him the Arthur C. Clarke of technological mundanity but since no-one wants to read S.F. novels about software upgrade utilities, he would probably have starved to death in his garret.
So in an act of supreme benevolence and humanity, the USPTO, recognising the plight of Reisman and the other lesser SF novelists like him, has kindly provided an alternative way for them to make money from banal and commonplace ideas.
That's just the way it is - a patent "grants the holder the right to exclude others from manufacturing, using, selling or importing the invention". In all the patent rules and laws I've seen at various PTOs, I haven't actually seen any explicit mention of a prohibition on giving away copies of the invention but the manufacturing bit is enough anyway. Manufacturing copies of a software 'invention' is effectively unavoidable if you're giving it away.
The right of the software patent holder to exclude others from using the idea is just as sickening of course - it always struck me how outrageous it was that I could implement the RSA algorithm in a few commands piped together in the shell, and yet technically, the patent holders could legally prevent me from doing so.
John Walker of AutoDesk fame and victim of the infamous XOR patent has had thoughts along those lines (PATO). See also these pages.
Well it wasn't very hard for Thomas Jefferson to understand nearly two centuries ago (I'm sure you've read that letter) and I cannot understand how the entire legislative structure of Europe could have failed so miserably to grasp the essentials of the issue. Then again, given the UKPTO's behaviour I suspect the law schools are just handing out alternative MBAs to their graduates these days, teaching nothing about the law other than how to use it to make money.
As far as I can see, the legislation the Commission and Council are trying to push through is incompatible with Article 10 of the ECHR and is therefore illegal.
"Patent law is a purely economical law for X sake, so why shouldn't economical effects have precedence?"
Purely economical? - so everyone keeps saying but what bothers me about software patents is that they're not just economical are they? In fact if I had thought they were I would probably never have taken any interest in the issue.
"Componentized"??? - IMHO that's an anglobstipation excessively gratuitous even for you transatlanticated verbalizators. '-)
"Tell me - in what way does having to cross-license their patent prevent them from profiting from it? "
It's very simple really - You get a patent to give you a temporary monopoly on the profitable use of your invention - no-one else can use yor idea in their products.
Then you are forced to cross-license with big companies X,Y and Z who can now use your idea in _their_ products and so you no longer have the monopoly you thought the patent was supposed to grant you. The advantage and marginal scope for profit that your idea initially gave you is lost.
Nonsense - even large companies have admitted that their own innovation is often slowed and even blocked by software patents (and let's stick to the subject of software patents - not all patents - brought up in freeduke's post). Here for example is what Alcatel said on the matter:
"Like other companies operating in the telecommunications industry, we experience frequent litigation regarding patent and other intellectual property rights. Third parties have asserted, and in the future may assert, claims against us alleging that we infringe their intellectual property rights. Defending these claims may be expensive and divert the efforts of our management and technical personnel. If we do not succeed in defending these claims, we could be required to expend significant resources to develop non-infringing technology or to obtain licenses to the technology that is the subject of the litigation. In addition, third parties may attempt to appropriate the confidential information and proprietary technologies and processes used in our business, which we may be unable to prevent. Our business and results of operations will be harmed if we are unable to acquire licenses for third party technologies on reasonable terms. We remain dependent in part on third party license agreements which enable us to use third party technology to develop or produce our products. However, we cannot be certain that any such licenses will be available to us on commercially reasonably terms, if at all."
And Joshua Kaplan of Intouch said in 2002:
"There are patents that come out today with hundreds of claims, unintelligible to almost anyone except the people who drew them. And yet, people who violate them jeopardize sometimes a lifetime of investment or their division or their product. That system doesn't work well to spur innovation or carry out the constitutional mandate.
Indeed, for those of you who were here this morning and listened to the people in the software industry talk about how threatening this is to their businesses, as I see it, patents today are often entrenching the established at the expense of allowing the newcomer to come in. I question today whether a Steve Jobs could start an Apple or a Bill Gates could start a Microsoft in view of the web and thicket of patents that is out there."
It is not so surprising that someone could hold and express such ill-informed and short-sighted views as you do, what did surprise me however is the depth of hypocrisy of a man who can proudly display on his website the work he has done in building on the GPLed work of others and yet simultaneously deem it necessary to 'apologise' for having to redistribute that work under the very same license that guaranteed his opportunity to do the work in the first place. May I remind you that your distaste for that license does not give you the right to breach it's conditions either, though I'm sure that's just a careless oversight on your part.
It matters not in the slightest whether an algorithm is obvious or not, it is a piece of mathematics and belongs to us all. If you find a new algorithm that is particularly efficient for some purpose then you are perfectly entitled to keep the matter secret. What you are not entitled to do (ethically speaking) is lay claim to ownership of that mathematical idea and prevent others from using it. Some time ago I read a comment in which it was claimed that the (unpatented) Viterbi algorithm was a pure invention, of use only in the field of digital communications and that consequently it would have been perfectly reasonable to patent it. This was obviously utter nonsense and it took me all of 30 seconds to discover that the Viterbi algorithm had already been adopted as an invaluable tool in several other fields. Sadly this is the kind of plausible fallacy often employed by pro-software patent activists and it can be very effective in convincing the innumerate and the scientifically illiterate.
Returning to the RSA algorithm - it can be understood by children, is about as sophisticated as Pythagoras's theorem and it was in fact patented by it's discoverers - or at least the first people known to have published it. That doesn't make it ethically justifiable though and it is certainly possible to patent algorithmic expressions of mathematics that is already well known, as for example Stanford University/Bracewell did with the DHT algorithm in 1987. I don't know offhand of any other uses for the RSA algorithm but if I had found one I would until recently have been legally prevented from using it. The mere possibility of that happening is enough to make software patents on algorithms absolutely unacceptable. A lot of maths research is done using computers these days and some fields probably wouldn't even exist without them (Fractal geometry - Mandelbrot and Julia sets etc. are the obvious and graphic example).
Economic arguments over the merits or otherwise of software patents are an irrelevance. I have purchased several general purpose electronic computers over the years and have used them for all sorts of things including writing programmes to implement bits of maths and communicating those ideas to others - sometimes in the form of explicit algorithms which for all I know may well have contained patented material. It is unconscionable to me that these largely non-economic activities be threatened by the existence of patents that benefit none but a handful of very large corporations and the parasitic community of patent lawyers. It is my computer and I will use it as I see fit and I will ignore the law if the law is enough of an ass to legalize exclusive ownership by others of the uses to which I might put my electronic machinery. I don't write fully fledged, saleable applications, I do not compete in the professional software market and if I did I would be well served by present day copyright law anyway. And when it comes to free software developer communities who could be said to compete in these markets, then if a company of full time professionals cannot carve out a profitable niche for themselves or keep up with
Quite so, but even that field is not entirely clear of landmines. - though it's the only one I know of and I heard that Bracewell later regretted the patenting of this algorithm.
"Now, you ask the average open source advocate what s/he thinks about software patents, and s/he will be opposed to them, on the grounds that they stifle innovation."
... though I would prefer a middle view that recognizes software as a distinct hybrid of an "idea" and a "device"."
Why not ask instead, the average mathematician, scientist or proud owner of what he or she once assumed to be a general purpose electronic computer? Just how is it you think anyone has the right to lay claim to exclusive ownership of any particular use of such a device and deny the owner of the machine his right to put it to that use?
"it's usually to copy a patented algorithm and incorporate it into an open source product."
'Usual circumstance' is no justification for allowing exclusive ownership of ideas that may and often do have only one algorithmic expression.
"the free equivalent to rsa"
Which is? - Please explicitly state an algorithm that would not have infringed the RSA patent and is also a valid algorithmic expression of the trivial corollary to Euler's congruence:
Given +ive integers a,k,k',m and primes p.q such that a) m=pq, b) kk' = 1 mod phi(m) and c) (a,m) = 1 then a^(kk') = a mod m
"Looking deeper, I don't see that it's consistent to be in favor of patents but opposed to software patents. This is because software blurs the line between a device and a description. For example, consider an integrated circuit. This is clearly a device, and hence patentable, right?"
Wrong, integrated circuits are primarily protected in Canada by the "Integrated Circuit Topography Act" of May 1993 and have similar protections elsewhere. And you are not "looking deeper" but instead making a shallow comparison and hollow analogy between software design and abstract software idea. Software designs are protected by copyright, abstract ideas are not and never should be.
Your 'drill printer' analogy attempts to further artificially blur the distinction between material design and abstract idea but it is ridiculous. Such a Trekkian replicator will still consume large quantities of matter and energy to produce the drill, a material object. Trade in material objects is fundamental to the operation of modern economies and such a replicating device would perhaps make it necessary to invent suitable protections for the designs of material objects but this has no bearing whatsoever on the question of patentability of abstract ideas expressed in software.
"this means that either all patents are valid (including software), or none are
No it does not, it is a non-sequitur and software is no more or less a 'hybrid of idea and device' than is a novel, musical score or work of visual art - all protectable by copyright.
"...some, such as rsa and lzw, are truly breakthroughs,..."
The RSA algorithm may seem like a 'breakthrough' to a software developer but as a mathematician I see it rather differently. The patenting of the RSA algorithm, like the DHT algorithm before it and any other piece of maths disguised as a software idea was a disgraceful act - effectively the partial privatisation of a piece of mathematical knowledge.
The patent on the RSA algorithm was particularly egregious as it is the obvious and *only* algorithmic expression of on an utterly trivial piece of number theory, a corollary to Euler's congruence. The idea that it should be a patentable entity is disgusting beyond belief and no - it does not matter that it's discoverers also call themselves mathematicians: No-one who maps some small part of the mathematical world yet subsequently attempts to fence it off for private use has any right to call themselves mathematician or scientist.
Let us remember that computers were themselves invented by mathematicians and are essential tools in modern mathematical and scientific research and I for one, strongly resent you programmers continually laying claim to 'ownership' of pieces of my subject just because you find them useful and profitable.
I do not enter your home and lay claim to ownership of your worldly possessions and I would be grateful if you would kindly reciprocate and keep your thieving hands off of my (and everyone else's) abstract mathematical possessions.
Thankyou.
Not obvious? You cannot be serious; not only is it obvious but I've seen it at least once before (KDE???). Yet the worst thing about it is not it's obviousness or whether there's any prior art but that it's the kind of thing I can imagine a highly configurable desktop or window manager might allow a user to figure out how to do by him/herself.
It's the kind of thing that I can imagine the developers of said configurable desktop might not actually have explicitly programmed into it in the first place. I can imagine one or more smart power users figuring something like this out for themselves and telling their friends how to do it. Who is infringing on such a patent now?
It's not so much a technology patent as it is a patent on a particular *use* of a technology. Do you really want people to be able to patent the way you use things? Or use patents to prevent others from making flexible and configurable things for users who like that kind of power?
This is just the kind of patent that ought to be waking people up to the idea that software patents are perverse and egregious thieves of diversity,creativity and individual liberty.
"...I would love to play the ram feeds from the BBC ..."
Unfortunately this news probably removes what little incentive the BBC may ever have had to provide their promised ogg streams. The BBC could have been a major force for good in this area but despite their size and power it seems they've surrendered to the irrational fears of third party rights holders and perversely, have even failed to make their own material freely and fully available.
Ever since I moved into a flat in which I can't get a decent EMR stream I hoped that the BBC would begin to make use of the pro bono publico technologies available to them. I tried their ram streams before I was aware of the issues surrounding such formats (and how inappropriate it was for the BBC to use them so exclusively) but the video streams were terrible and ate up all my memory before crashing X and the audio streams were of poor quality and often unreliable.
I should have known better than to expect the BBC to show any concerns about what is in the best interests of the public: They are bbc.co.uk and not bbc.org.uk after all and in recent years they seem to have made every effort to ensure that they better reflect that choice of domain name rather than the public service commitments of their charter.
Ironically, I now listen to the BBC World service streamed to me courtesy of WNYC.org and have found many of the other NPR programmes to be of equal quality to anything the likes of Radio 4 and 5 transmit. I have even found replacements for Radio 3, an especially good one being www.rozhlas.cz.
"GnuLinux" (guh-noo-lin-ux) - hardly more difficult than "Moulinex" (moo-lin-ex), the successful household appliance manufacturers and they probably paid $100,000s to a 'branding agency' for the name.
It would be even easier if the "Gnu" was pronounced "noo" (Chambers), like the real word referred to at http://www.gnu.org/gnu/gnu-history.html - unfortunately the pronunciation given at http://www.gnu.org/ conflicts with this.