SoylentNews is people
SoylentNews
Roy Schestowitz at TechRights reports
Further reinforcing the current trend, software patents' demise in the United States has just been ascertained again.
[...] The latest decision[PDF] [...] from the Court of Appeals for the Federal Circuit (CAFC) is eye-catching, but either it hasn't caught the eye of legal firms or they're just trying to ignore it, so we'll be covering it more than we usually cover such decisions.
[...] The ruling is very important because it serves to demonstrate a loss for patent trolls and for software patents (or patent trolls that use software patents, which is typical). The decision criticises patent trolling as well.
[...] So far, based on our research, only one press article has been published about this decision. It's titled "Here's Why Software Patents Are in Peril After the Intellectual Ventures Ruling".
The end may be in sight for software patents--which have long been highly controversial in the tech industry--in the wake of a remarkable appeals court ruling that described such patents as a "deadweight loss on the nation's economy" and a threat to the First Amendment's free speech protections.
The ruling, issued on Friday [September 30] by the U.S. Court of Appeals for the Federal Circuit, found that three patents asserted against anti-virus companies Symantec [...] and Trend Micro were invalid because they did not describe a patentable invention. The patents were owned by Intellectual Ventures, which has a notorious reputation in the tech world as a so-called "patent troll", a phrase that describes firms that buy up old patents and wage lawsuits in order to demand payments from productive companies.
Software Patents as a Threat to Free Speech
Friday's ruling is also significant because Judge [Haldane Robert] Mayer eschews the insider baseball language that typically dominates patent law, and addresses patents in the broader context of technology and government monopolies.
Pointing out that intellectual property monopolies can limit free speech, Mayer notes that copyright law has built-in First Amendment protections such as "fair use" and that patent law must include similar safeguards. He suggests that the safeguard comes in the form of a part of the Patent Act, known as "Section 101", which says some things--including abstract ideas--simply can't be patented in the first place.
(Score: 2, Interesting) by SecurityGuy on Tuesday October 04 2016, @04:25PM
I find that very troublesome. The notion that a "computer" is technology, but software is not, and therefore doesn't deserve equal protection. If the purpose of patents is to spur invention, then we should also value and protect true innovation in software. The case in question seems to have hinged on one of the many "...on the internet" ideas, where someone takes a process that's been done in meatspace forever, does it on the internet, and thinks that should get them a 17 year license to preclude anyone else from doing it. That, of course, is idiocy. If you spend years toiling over a problem, a genuinely hard problem that others have been toiling over and failing to find a solution, and you're the one who finds it, why should you not be able to recoup the time, effort, and expense?
I don't think software patents in general are a problem. The problem is that 99% of software patents are issued for things that aren't novel or non-obvious.
(Score: 1, Interesting) by Anonymous Coward on Tuesday October 04 2016, @04:55PM
I've heard writing software compared to writing an assembly procedure for a piece of pre-fab furniture (a concept which is clearly not patentable).
Another argument says that rearranging a list of existing instructions is not a creative act (in line with writing a shopping list).
The double coverage of copyright and patent has already been mentioned in this (meta)thread.
-- OriginalOwner_ [soylentnews.org]
(Score: 1) by fubari on Tuesday October 04 2016, @05:25PM
Yeah, copyright seems like a good way to go.
Excerpt from the fine article:
He (Judge Mayer) adds that, because software is basically a form of language, its intellectual property protection should be limited to copyright — which covers things like musics and books—rather than patents.
So copyright, and maybe trade secrets ("secret sauce" if you will).
TFA is pretty freaking amazing, never thought I'd see the day something like this came up (then again, never thought I'd see the day of something so reasonable as marijuana legalization... it almost seems like an outbreak of common sense is sweeping the United States. I will stay tuned for the counter examples :-) ).
(Score: 0) by Anonymous Coward on Tuesday October 04 2016, @06:02PM
Roy Schestowitz used to be a regular at comp.os.linux.advocacy. [os.linux.advocacy]
He has since gotten his PhD in CS and found himself a hot wife (who is also a CS pro).
He's still the hardest-working guy you are likely to find, doing his day job plus trying to keep folks informed about FOSS and anti-FOSS topics.
His main topic for the last many months has been the completely-corrupt European Patent Office.
N.B. TechRights was originally called BoycottNovell (instituted after the Novell-M$ pact).
-- OriginalOwner_ [soylentnews.org]
(Score: 2) by SecurityGuy on Tuesday October 04 2016, @06:11PM
To me, that's like the argument that Microsoft Office isn't copyrightable because it can be represented as a number (albeit an incredibly large one).
I think that line of argument is basically reductio ad absurdum. Having written software and assembled pre-fab furniture, they're not at all the same thing, even if they share a common generic root: writing instructions.
It's like comparing Shakespeare to one of those metaphorical million monkeys banging on keyboards. Fundamentally, old Will was actually a kind of monkey banging on a keyboard (ok, scribbling with a quill, but you get the idea...). Still, are they the same? No.
(Score: 1, Insightful) by Anonymous Coward on Tuesday October 04 2016, @08:01PM
Microsoft Office
I like the example of "spreadsheet" better.
Those existed before computers.
Now, they were extremely cumbersome to do by hand and a computer made them infinitely easier BUT THEY WEREN'T NEW.
"With a computer" doesn't make something old into something new.
Pretty much **anybody** who had used a computer and had used a spreadsheet would know that combining the 2 is a logical thing to do.
Making the effort to produce the app is worthy of a **copyright** on that **specific** work product.
A PATENT would indicate that you came up with a wholly NEW thing--which you clearly didn't.
.
Having written software and assembled pre-fab furniture
You've distorted what I said.
The point is that one is a set of instructions for a -human- to follow and the other is a set of instructions for a -machine- to follow.
This is sometimes referred to as "a distinction without a difference".
Again, "with a computer" doesn't make something old into something new.
-- OriginalOwner_ [soylentnews.org]
(Score: 2) by sjames on Tuesday October 04 2016, @05:17PM
Currently, because someone got to the patent office 5 minutes before you did. Why indeed should you not be able to use all of your effort because someone else got the patent first, largely because of chance events?
So what would you say is the last time someone did anything on a computer that was so non-obvious that it truly deserved a patent?
(Score: 2) by SecurityGuy on Tuesday October 04 2016, @06:26PM
I completely agree with this.
I figured someone would ask that, and my honest answer is that I don't know. I'm not an IP lawyer, so I don't have my ear to the proverbial tracks listening as each patent is granted. Cryptographic algorithms (which are basically defined in pseudocode) might be reasonable, as long as they're truly unique in some way, not just a permutation of something that already exists. Page Rank didn't bug me too much, though I don't know that 17 years is a reasonable term for any technology patent today.
I just don't think we should invalidate an entire category of human endeavor (doing useful stuff with software) any more than we would another (doing useful stuff with a factory). Rather, we should limit it to non-obvious, novel inventions like the patent system was always intended for.
(Score: 2) by sjames on Tuesday October 04 2016, @08:44PM
Cryptography itself is explicitly forbidden to patent, it's "just" math. Implementing that math on a computer is clearly far from novel (and is largely mechanical translation).
Page rank is more or less a re-implementation of the impact factor applied to academic works with a few ad-hoc heuristics thrown on top to limit gaming. I'm not saying it wasn't good work, just that it's not that non-obvious.
I'm honestly not sure if something about software precludes novelty and non-obviousness sufficient to be worthy of a patent (perhaps because there's so little friction between idea and reduction to practice that everybody's doing it) or if it's just that too many people are too busy patenting the obvious or emulating the physical world, but the field of software seems to mostly produce bad patents. It could be that the really interesting things that take a long time tend to be academic and result in a lot of incremental publication followed by someone at the end commercializing it and claiming all the credit.
(Score: 2) by SecurityGuy on Tuesday October 04 2016, @09:04PM
http://www.emc.com/emc-plus/rsa-labs/standards-initiatives/important-patents-in-cryptography.htm [emc.com]
Honestly, I expect that part of it is that it's our field. If my job were producing things in a factory, I'd probably find a lot of those things obvious, too. Is something like Page Rank more or less obvious than pull tabs (patent #4530631), for example?
(Score: 2) by sjames on Tuesday October 04 2016, @10:05PM
There do exist patents in cryptography for much the same reason some guy in Australia managed to patent the wheel. Or if you prefer, the guy who patented teasing a cat with a laser pointer. But note how few there are and that they are all expired.
The pull tab patent is actually for the method of cheaply and reliably mass manufacturing the pull tab can tops. It may or may not have been worth a patent, but it very probably did involve a great deal more in time and materials to reduce it to practice than page rank.
(Score: 2) by DrkShadow on Wednesday October 05 2016, @01:10AM
Are you saying that a Factory is such a tangible idea that it should be patentable?
(Score: 0) by Anonymous Coward on Tuesday October 04 2016, @06:08PM
Complex software can include tens of millions of lines of code or more, and it is impossible to know how many software patents you're infringing upon given such a large project. Even small projects are subject to this, since no one is going to know about every patent in existence and it's easy to write software that violates a patent without even intending to. It's even possible to suddenly discover something very complex, which seems to be what your post focused on.
If you spend years toiling over a problem, a genuinely hard problem that others have been toiling over and failing to find a solution, and you're the one who finds it, why should you not be able to recoup the time, effort, and expense?
Rather, why should your method of recouping the time, effort, and expense be to have the government destroy competitors? The ends don't justify the means. Find a viable business model or fail.
Since software patents necessarily infringe upon free speech rights, they would not be morally justifiable even if they helped spur innovation. Freedom is more important.
(Score: 0) by Anonymous Coward on Tuesday October 04 2016, @06:18PM
And, if in the course of writing those millions of lines of code you can accidentally infringe on multiple patents, then how the hell can they be worth patenting?
(Score: 2) by SecurityGuy on Tuesday October 04 2016, @07:04PM
The pharmaceutical industry is a good parallel here. It might take me a billion dollars to find a drug that works for a particular condition. Once I've identified the drug and put it on the market, you can, with vastly less expense than a billion dollars, analyze the drug and produce it yourself, then sell it for a lot less because you didn't have to spend a billion to find it.
Why should the government (or rather why should we, as a society) grant me the ability to exclude people who would do that? Because otherwise I wouldn't have spent the $billion to find the drug in the first place.
That, really, is the test for me. We should have a patent system that successfully encourages people to create works that:
1) no one would create absent that artificial incentive
2) provide a net benefit to society (who grant them that artificial incentive)
(Score: 0) by Anonymous Coward on Thursday October 06 2016, @01:12PM
Why should the government (or rather why should we, as a society) grant me the ability to exclude people who would do that? Because otherwise I wouldn't have spent the $billion to find the drug in the first place.
As I said, the ends don't justify the means. Find a viable business model, or fail. Even if no viable business model even exists, that doesn't make it morally justifiable to take some amount of control over other people's property and speech (as software patents do). Censorship is especially never justified.
(Score: 3, Interesting) by DannyB on Tuesday October 04 2016, @07:00PM
I agree that software is technology.
But software does not deserve patent protection. Copyright protection is fitting. Someone should not be able to rip off your work.
But you should not be able to exclude someone from doing their own work, their own implementation.
Most new software gets the reaction, it is obvious how to implement that. It just takes the roll up your sleeves work to do it. Nothing magic.
It is rare that new software contains a secret sauce that is non obvious to most people. Which gets the reaction "how did they do that?"
I think it is sufficiently rare that software should not be patentable at all -- in order to avoid all the common abuses -- just to protect the exceedingly rare "how did they do that?" software. Keep it a trade secret and copyright it. Use a EULA.
But here's the other thing. It is simply a fact of life that sooner or later someone is going to re-implement it. Their own implementation effort. So if you've got something truly new and great, use your market lead advantage.
Also, if you could get a patent and licensed it at a reasonable price, everyone would just pay and be done with it. Or if you keep it to yourself and sell your product at a reasonable price (especially software only product), then everyone will just buy it. But the problem is that greed kicks in and seems to lead to this "I must have it all!" mentality.
Example: Apple. The Mac was made only by Apple. It was vastly ahead of competitors. It had, at one point, a commanding 25 % of the market -- and that over pathetic MS-DOS machines, not windows yet. The best thing Apple could have done would be to have licensed the OS. But that would mean some PC makers might make PCs that don't conform to Apple's sense of style and taste. Other PC makers aren't just going to go home and cry and not make PCs anymore. Once some other software, even inferior, comes along, they are going to use that. And Apple can't out compete everyone else put together. Sooner or later the innovation will gradually move to the other platform.
Next, the iPhone. Again, ahead of the game. They should have licensed it. But that would mean some phone makers would not conform to Apple's sense of style and taste. Other phone makers are not just going to go home and cry and stop making phones. Especially since some of them had been in the business long before Apple. When some alternative OS comes along, they're going to use that. And Apple can't out compete everyone else put together. Sooner or later the innovation will gradually move to the other platform.
Software patents are a bad idea. This is how we got to slide to unlock (yes really) and round rectangles (yes really) being worth over a billion dollars. Back in the classic Mac days, with Apple's regular tech notes, there was a tech note describing the coming software patents and the disaster they would be. You wouldn't be able to write software without having a lawyer. And every prediction in that tech note came true.
Hypothetical engineering problem: (prior to introduction of iPhone) You're handed this problem. This phone has one button. But you don't want a push of the button in your pocket to activate it. The only other input device is the screen. So how do you prevent accidental device activation in someone's pocket or purse? Do you really think some sort of slide gesture is not obvious?
To transfer files: right-click on file, pick Copy. Unplug mouse, plug mouse into other computer. Right-click, paste.
(Score: 2) by SecurityGuy on Tuesday October 04 2016, @07:15PM
I read a probably apocryphal example once. Guy's in a trial over whether his patent is obvious or not. As an example, he hands out eggs to the jury and tells them to balance them on one end without using any props. They all struggle for a while, and finally give up. Guy picks up an egg, smacks it on it's end into the table, and balances the apparently hard-boiled egg on it's now-crushed end. The test should not be "Is it obvious now that someone shows you how to do it?" Is it obvious BEFORE anyone shows you how to do it. Light bulbs are a great example. Lots of people were trying to create light bulbs. To us, it's obvious, but only because we've seen the method that works.
Two words: Reverse engineering. That's exactly the thing patents protect against in the real world. I can just take your physical product apart and see how it works. There's nothing to keep me from doing so...except patents.
Nope. I've already said a number of times I think most software patents fail the non-obvious test.
(Score: 2) by stormreaver on Tuesday October 04 2016, @08:17PM
Guy picks up an egg, smacks it on it's end into the table, and balances the apparently hard-boiled egg on it's now-crushed end.
At that point, the judge/jury should have invalidated the patent for failure to disclose pertinent conditions: the egg had to be hard boiled. Without that little, crucial, piece of information, the demonstration would not have worked. But this guy would have then sued anyone who crushed a hard boiled egg to balance it on its end.
This is the kind of stuff that supports the abolition of software patents.
(Score: 2) by FatPhil on Wednesday October 05 2016, @08:28AM
I have some topological background, and will simply lay it on the table carefully, and when it comes to rest define the point of contact as the "end". If he's going to say "no, only that point there is the end" then I'm going to say "your smashed end isn't just that point there".
Hucksters shouldn't be given patents.
Great minds discuss ideas; average minds discuss events; small minds discuss people; the smallest discuss themselves
(Score: 2) by DannyB on Tuesday October 04 2016, @08:27PM
In the Reverse Engineering situation, this would be when the software does something which gets the reaction "how did they do that?". Rather than, I could build that same thing. Just need to roll up my sleeves and do the work. In other words, implementation is obvious.
The Reverse Engineering situation is what I argue is the rare case. Not worth giving that software patent protection. Not when balanced against the vast majority of cases (like today) where we have ridiculous obvious patents being granted and lawsuits for HUNDREDS OF MILLIONS or even over a BILLION dollars.
Too bad about that rare case. But letting it have a patent means that the entire rest of everyone else, every company, is hindered by in a morass of patent litigation and lawyers. Innovation is slowed down, just so that an occasional software item could get the patent that it might deserve.
I just don't see the patent system ever being truly fixed. And even in some fairytail world where it could be, it wouldn't stay fixed. It would quickly slide back to what we have now, or worse.
Sometimes there is a rare piece of software that deserves a patent. (Although I don't even know of a good example.) But the cost to everyone else for that to happen is just too high.
To transfer files: right-click on file, pick Copy. Unplug mouse, plug mouse into other computer. Right-click, paste.
(Score: 2) by SecurityGuy on Tuesday October 04 2016, @08:42PM
I can't really argue with that. Should software be patentable? IMO, yes. Is the current system, specifically that the USPTO grants patents to things they shouldn't, relying on the courts to sort it out, so unworkably broken that the cost of permitting software patents exceeds the benefit of them? Yeah, I'll grant that's possible. I certainly don't have a lot of hope of getting the USPTO to stop issuing bad patents.
(Score: 2) by DannyB on Tuesday October 04 2016, @08:35PM
What is a good software equivalent of that?
Like your light bulb, the vast majority of software, once you see a demonstration, it is obvious how to implement it yourself.
But it would have been as easy to implement yourself.
I would argue the software didn't take the kind of sweat and research that the light bulb took. I think the idea of incandescent bulbs were even known. It was just how to make it work -- which took a lot of experimentation. And inventions that require that kind of experimentation tend to be hardware. ("How to build a better flat panel display.", "how do we make these flip flops toggle faster?")
Unless we're talking about something that is a breakthrough (your manager says: "I need a working AI system with speech recognition and emotions."). Usually the case is your manager says "I need X". And it is just a matter of building X.
To transfer files: right-click on file, pick Copy. Unplug mouse, plug mouse into other computer. Right-click, paste.
(Score: 1) by terryk30 on Wednesday October 05 2016, @07:47AM
To further your contrasting of hardware invention as often requiring experimentation, the world of hardware throws up unexpected constraints from quirky materials and snags in working and tooling, unlike the world of software where the mere specification or outline of the problem essentially implies its own solution (to those skilled in the art) because the medium of realization is idealized general-purpose computing machinery.
IIUC linking the latter to math and Turing machines has been one basis for banning software patents as a class (rather than just banning them based on their generally lower degree of nonobviousness), but in persuading non-CS types of this (or ultimately the public) I've wondered if there also hasn't been enough explicit contrasting with the former - that is, with the constraints of the world of hardware. I.e.: the lack of physical constraints make software inventions intrinsically "less nonobvious" than physical inventions.
(Score: 2) by FatPhil on Wednesday October 05 2016, @08:20AM
No, not at all.
Round rectangles are a "design patent". They're at least as bogus as software patents, because at least software patents cover how a thing functions, but design patents only cover how it looks. Something which was already protected by the concept of trade dress (related to trademark law) and even copyright (if there was an actual artistic design being pretected).
Great minds discuss ideas; average minds discuss events; small minds discuss people; the smallest discuss themselves