High Court approves software patents Technology firms will be able to patent software programs following a High Court decision on Friday that could see the UK having closer ties with Europe when it comes to the handling of computer-related inventions. The Honourable Mr Justice Kitchin ruled in a case brought by five small UK businesses that the Intellectual …
The High Court is the final resting place of old fools unable to cope with the pressures of a real court. HC judges are inept, incompetant, and largely out of touch with any social needs not covered by Oliver Twist.
So, why exactly are buffoons like this old fart allowed to make up laws as they go along?
It says in the piggin' requirements that software as such isn't included.
So if it's software to control a unique bit of hardware, it isn't the patenting of the software to control it as such you're patenting. It's the hardware and how you control it.
Add to that the fact that there's nothing available to make this product from the patent applications, making the patent incomplete, there's no patent.
As someone pointed out, they are all:
A unique bit of maths changing one number into another number.
that's maths. And that's not patentable.
This is a sad trend - more and more people in the West realise it is easier to manipulate the legal system and lobby for skewed, preferential laws than it is to actually go and produce anythng useful and create value.
What that means is that an ever increasing part of the economy will thrive on the false pretense, eating money which is not earned, consuming product which is not prduced. That will translate into inflation and reduced competitive strength.
In the meantime, I am absolutely sure, the developing world will not abide by these stupid, made-up rules just to please Bush and Brown and co.
I don't like the way things are heading, eventually people will not be able to think of an idea unless they first buy a licence from another who has patented the idea of ideas. A computer program should be as patentable as a cake receipe, or the baking process itself, which is to say not at all.
Those who are pushing for software patents would surely also lobby to patent a method of mixing ingredients and applying heat to create an edible foodstuff. Surely they would then attempt to use the courts to wring money from every bakery in the land, at least until some other greedy person patented the idea of the oven (but not any specific kind of oven) and ursurped them.
Copyrights yes, patents no.
As a mathematician, that might make me rich! That would be a way for mathematicans to become rich, unlike the current situation where only those people who exploit our inventions can get rich.
On the other hand, software patents are obviously stupid. I'm in a quandry here...
... software should be patentable. Well, individual algorithm anyway. I am not talking about "one-click checkout". I think the algorithm should have to be:
1) an actual algorithm
2) non-obvious
3) have no prior art
So I think the mp3 codec would qualify, as it had no prior art, and if you were going to compress audio, the method doesn't just jump out at you (as evidenced by the fact that it took them so long to think it up). Btrees might arguably qualify. Doom-style BSP trees would be a definite. Like any other invention, I think algorithms need to be patentable, it is just the silly nonsense of patent offices that has been mucking up software patents; the concept itself is valid.
Anyone who thinks the patent system protects little people needs to try and join in a monopoly game 4 hours into play when every utility and almost every street has four hotels on it. You'll be bankrupt before you pass "Begin".
The only solution I can think of is patenting the patenting of the bloody obvious.
Aliens because perhaps their prior art will save us?
This isn't an approval of software patents. This is a reversal of a practice change made just over a year ago by the UK IPO when they misunderstood the landmark Macrossan judgment.
Under the old practice, only some software could be patented, and you could claim that software as a method or as a computer program. The practice change meant that it had to be claimed as a method but otherwise didn't change what software could be patented. This was rather silly of the IPO. If you grant patents for software, you grant patents for software and quibbling over the language used in the patent doesn't change that.
The decision doesn't change what's patentable at all. It just confirms the old practice of granting claims to computer programs as well as to a method so long as the invention isn't just a new computer program but, say, a more efficient method of using printer toner which, in practice, has to be controlled by a computer program - this being one of the inventions which has just been confirmed as patentable by this new judgment.
This is still miles away from Amazon's 1-click which neither the UKIPO nor the EPO have granted a patent for.
The test should NOT be whether the algorithm is implemented in cogs and wheels or bits and bytes.
The test should be whether the algorithm is trivial or not. So it appears the court made the right decision from a technical and engineering viewpoint.
Morely, if copyright protection worked on algorithms, inventors would go for it, since copyright protection lasts much longer and in some circumstances can be renewed.
Those looking for a free use of other people's inventions, like Paris, would be much much worse off if we could copyright our ideas.
MP3 not obvious? In my third year at university (1990) someone asked me, out of the blue, how I'd compress music or audio? I said (after about 1/2 second thought) well, take the energy specrum of the music, remove areas where there's little energy and for the rest, reduce the data volume (bits per sample) for lower energy bands.
That's approximately what ATRAC does. MP3 uses somewhat similar approach: dynamic sampling and reducing the waveforms to a simpler one for the powerband they are in.
MP3 uses (if I'm right) *unpatented* algorithms from number theory and some mathematics from perceptual studies (also unpatented). The actual values used could have been printed in a research journal along with them. But were patented instead.
Nah, the problems were solved in open scientific journals.
You're wrong, a monopoly has a guideline figure in the EU: 60%. In actual fact, it is the ability to drive the market where the monopoly wants to go, rather than where the customers or market entries want it to go.
Did anyone get anything worthwhile from Vista? No. But they stop XP soon, stopped earlier version AGES ago and we all have to keep up.
MS have such a large share of the market, anyone wanting to enter must be compatible with Windows (as they had to be compatible with Intel, though Compaq and Via/AMD/Cyrix et al all were COMPATIBLE so IBM didn't have a monopoly per se) or they wouldn't get in the market. How well would a Mac-only game sell? What if it was Windows-only?
And isn't copyright and patent a monopoly grant anyway?
I'll tell you a story that shows how different software is from hardware (the figures aren't accurage, but ballpack correct, and the story is true).
You can get software that will design a circuit. Works great.
Someone thought that maybe the software was missing out on things that the real world had, so use FPGA's to design a circuit that would haev a stead voltage out either high or low for a high frequency or low frequency input.
Standard way in software would have a timing circuit that would take more than the 100 transistors to make on its own.
He used genetic algorithms to select the best 50 out of 100, interbred them and kept going. Eventually it got stable. And only 30 transistors out of the 100 availble were connected. However, if he disconnected all 70 redundant ones, the design didn't work. He needed 38, even though 8 of them weren't connected at all.
The reason why this story is needed is because when you get to implementing your idea in cogs and wheels, you find out that Real Life (tm) is a lot less helpful than software. You can find that your backlash is too high, or the teeth don't mesh or that there's a maximum rate you can use, or you need odd materials. Software works under it's own set of very simple laws (add one number to another number, repeat 1billion times a second) and it doesn't give you anywhere near the issues.
In the case of the cogs and wheels, you may find that it DOES work but not well. So someone comes along, solves the EXACT SAME problem but uses different shaped cogs (cams maybe) or a new material (that they invented) and it works better. And THEY get a patent on this "better mousetrap". With software, the only way you can make it "better" is by using more efficient code. But code isn't part of the patent, just the outcome required. Copyright is used on the code.
Lastly, the software patent version of "building a better mousetrap" would be something like:
A method to restrain by force, impediment or other method the continued translocation of a pest (rodent, insect or other) by means not limited to lethal action, trapdoor, latching or immobilisation, such situation to be occasioned on the ingress of the pest species into the invention. Such change in state of the trap being sufficient to render unavailable egress of the pest species from the invention's controlled space.
Now build a better mousetrap from this...
war?
Then any country going to war will have to pay me a hefty license fee!
I may have a problem with prior art (of war) though...
Oh well, back to thinking of things I could patent without getting off my lazy, unimaginative arse and implementing them then.
God, stifling creativity is so hard nowadays!
Good show Mark, I'll have to admit you're right on the MP3 thing (I was debating whether to use that as an example or not).
I think the point more or less works though. It would be silly to say you can't patent an algorithm in software, but then could implement the exact same algorithm in hardware and patent it. I can think of alot other examples that would be prime candidates for patents: the first artificial nearal network, wireless mesh net work routing protocols, the first phonem based speech recognition engine, hidden markov models, etc.
No there are no software ideas that are patent worthy it's an utterly mistaken idea that was done stupidly in the US now everyone has to patent everything to keep others from hijacking their business that camel is a bastard and he won't rest till he is all the way into your tent. All the styles and ways you can use code are all just that styles they compile or interpret and then compile to the same fucking opcodes over and over again you can't patent a literary form(novel, short story, anthology, long letter) and you shouldn't be able to patent code.
"It would be silly to say you can't patent an algorithm in software, but then could implement the exact same algorithm in hardware and patent it."
I would say patent it in hardware and you're OK. The reason is that hardware can be tricksy and there may be unusual discoveries that can make a BETTER MP3 hardware circuit (one that is faster or cheaper or uses less power). And we already know that the full circuit diagram and process is patented with these elements on show (despite there being a copyright on blueprints, they are patented INSTEAD, unlike software where you don't get to see the code and it is full patented AND copyrighted AND a trade secret).
If someone can come up with a good way of mp3ing a datastream faster and the circuit is different, it can be patented. If someone comes up with a quicker more optimal code for mp3 encoding, they can't patent it because it's still doing the same thing the patent talks about.
Isn't THAT insane?
Don't get me wrong I understand all the comments about the patenting of mathematics. However here is the issue as seen by the small independant dev house.
We have an idea which we can take to market which gives companies what they want i.e improving efficiencies. The dev costs involved in the product are astronomical so the only way forward is to try to licence the tech.
Go to a large software apps supplier they like the idea but decide they can do it themselves for a fraction of what you are asking for the licence.
Copyright is not a protection. Patents are needed.
Concepts have value as with the magnetic shelf people hadn't thought of it therefore it has an intrinsic value that should be protected. (Tangental aside....sorry)
The reason the dev costs for a small independent house are astronomical, and not for a large software apps supplier are :
1) The SIH is crap at coding and business
2) The large supplier has more experience, pre-existing code and better tools
It may be that a large software house has an advantage in terms of infrastructure, but all things being equal, they're not an order of magnitude better at coding than anyone else.
If your product isn't selling it's either crap, unnecessary, or too expensive. If other people prefer to implement themselves, the idea is good but it's definitely too expensive and the implementation is either poor or insufficiently segmented for different needs.
Patents have nothing to do with it.
If I have a good idea, like, if I'm solving a problem and I come up with the only rational solution, I don't wish to find out a month later that someone else had the same idea as me and is entitled to charge me to implement my idea.
That's the arrogance of patents. I haven't got the time or the inclination to check the patent databases every time I have a good idea to make sure that someone hasn't had the same idea and patented it - if I had to do that, I would never get around to writing software.
I also don't have the arrogance to claim that I'm the only person capable of having such an idea. That's ludicrious.
Software patents are utterly wrong. Writing a program requires you to solve potentially thousands of problems, and it's wrong that someone could prevent you from solving some of the more essential ones.
Amazon-1-click was granted by EPO, though in a tuned-down version. EP927945 has only recently been revoked.
A horrifying thought: Had Emazon used the same 'gift ordering patent' method, they'd probably been bankrupted throughout all those years of the existence of a valid EP-B-927945. Now it would be too late to sue Amazon; jobs, families and growth lost. Public loss, taxpayers' loss.
I'd have only one outcome for fraudulent cheats like Bezos: 3 months of jailterm for fraud, and sue them into life-long bankruptcy.
I would think that the dozier internet law copyright would have a problem when I sue for using my name without paying me for the right to use it. If a business, like Cyrix for example, can sue a movie studio for using it's name without permission (6th day???), then I'd fall into the same category. I'd sue that my name be removed from said copyrighted work unless I was payed an obscene amount of money.
Sorry, but if you think allowing patents for software is going to benefit small business, you're clearly not aware of the millions big business puts into the patent box every month.
Heck, Microsoft, IBM, Oracle, Intel and all the rest - they have dedicated lawyers working in-house ! By the time the small business has tested its product and is ready to patent it, it will be to discover that any number of the big guys already have 20 variations that are patent-pending.
Sorry, but I cannot believe for one second that granting software patents is going to do anything but consolidate the revenues of the pigopolists. If it does help a little guy, we'll be sure to hear about it because it'll be the only case.
A patent is only valuable if you can defend an infringement in court, this process must be costly and far out of reach of much of the small business community. Therefore getting a return on the value of a patent must be realised by defending it in court a cost many small start-ups could not bare but large business use a patent system to control the market.
It’s the same situation as the supermarkets and their landbank, preventing the competition from entering the market by controlling where the competition can operate.
The Register 
Biting the hand that feeds IT
Copyright. All rights reserved © 1998–2025
