AMD wins Patent Infringement lawsuit against Vizio

[S w a t s o n]

Alright I'm confused, is vizio putting gpus inside of televisions?

[Drak3]

35 minutes ago, S w a t s o n said:

Alright I'm confused, is vizio putting gpus inside of televisions?

Yes. Vizio Smartcast and SmartTVs use their custom V series APUs.

[Sniperfox47]

5 hours ago, S w a t s o n said:

Alright I'm confused, is vizio putting gpus inside of televisions?

Any smart TV has a GPU. Most non smart TVs also have GPUs for accelerated 2D drawing of the interfaces.

[LAwLz]

20 hours ago, mr moose said:

I don't know how invested in this subject others are, but it has been a hobby (IP, copyright and corporate function in economics/society) of mine for a very long time.  We can always find the odd story or report here and there that supports whatever ideal we may have on the topic of copyright and IP, but at the end of the day the overwhelming consensus is that it grows economies and pushes development of new and better technologies.   Given it is a system we have had in use since the 18th century we can see demonstrably and emphatically that it does not prevent new inventors or startups from competing.

It's very difficult to argue for or against patents because on one hand, companies can claim that without patents they wouldn't invest in developing things.

On the other hand, you can't really measure or check what effect the removal of patents would have on competitors.

 

No patents = more companies can enter the market, thus causing more competition.

 

The question is, would the increased competition be enough to make up for the lack of protection your research? I don't think there is a definitive answer to that, because we can't peek at alternative universes.

 

However, I do think that our current 20 years of government issued monopolies does stifle innovation. I am very much for lowering it to maybe 10 years for technology related patents, plus completely removing the concept of software patents. They are completely illogical and spits in the face of what patents are meant to do and protect.

[D13H4RD]

On 8/26/2018 at 4:55 AM, yian88 said:

All patents should die, its the number one killer of tech advancements and cheaper products, competition killer and so on.

All patents are tyranical in their nature.

Fuck copyrights and patents, and im very serious.

Um, I wonder if you realize exactly what patents and copyrights are intended for. 

 

Let's say you make a fantastic art piece and post it on DeviantArt. Everyone loves it and you get recognition. But what if I take that art, lightly modify it so that it looks slightly different but still similar and claim it as mine? Or even worse, I take the art in its entirety and claim it's mine? 

 

That's precisely why copyrights and patents exist. It means that you legally have the right to use, reproduce or license that piece of work or tech that you have made. It's to ensure that such works are protected from abuse and theft from outside parties. 

 

Plus, let's be real. Every one of us has to earn money. 

 

I agree that some patents, like Apple's rounded corners one are silly to the point of absurdity. But there are other legitimate things that patents serve to protect 

[leadeater]

17 minutes ago, LAwLz said:

The question is, would the increased competition be enough to make up for the lack of protection your research? I don't think there is a definitive answer to that, because we can't peek at alternative universes.

And there are already too many giant mega corporations that would have the scales tipped massively towards them, an abrupt removal I would imagine would be rather bad.

[LAwLz]

20 hours ago, Aetheria said:

There's also an opportunity cost associated with having all your best and brightest working on solutions for purely legal problems. That's not progress. The fifty people who've been reinventing the wheel could have been inventing something genuinely new, better, and of more use than just allowing one company to avoid a licensing fee.

AV1 is a great example of this.

Netflix, Mozilla, Google, Microsoft and a ton of other companies has dedicated several teams of lawyers as well as encoding gurus just to figure out how to get around current patents.

They are spending thousands upon thousands of man-hours to avoid patent infringement. It's such a massive waste of resources for society.

 

 

19 hours ago, Jito463 said:

Would you feel as honored if someone took your code and used it without giving you any credit?

No, but that's why many copyleft licenses require giving credit.

I don't think AMD would have been pleased with just getting a "thank you" in a menu though.

 

 

19 hours ago, Trixanity said:

To be fair, in the last couple of centuries and especially in the last couple of decades it has become so much easier and faster to copy than it would have been a thousand years ago.

 

Access to information, access to education and access to talented people has never been easier.

 

The problem is finding the balance between the sharing of knowledge and protecting ideas and investments.

 

Right now it's a shit show but it'd be an even bigger shit show if we went to either extreme (abolishing all patents making it free for all or making patents easier to acquire and withhold).

 

Reform is necessary but if it was that easy it would have happened already and I imagine interest groups are lopsided in their power. So there is probably a vested interest and therefore backing in maintaining the current system as those who benefit are those who already have the money to maintain it.

Totally agree. I am not against patents (except software ones). I am against the current patent system though.

 

I can understand why some patents needs to last 20 years, but for technology it is way too long and it only benefits a handful of companies, while it limits everyone else. Shorten them to maybe 10 years. Even that might be too long to be honest.

Also, software patents should be completely abolished. They are illogical and goes against the original concept of patents.

 

19 hours ago, Stefan Payne said:

They do both!

They license the instruction set to everyone interested and they do their CPUs themselves (like Samsung Exynos, possibly Apple).

They also do reference implementations like the Cortex A53 for the CPU side.

And ARM also has the MALI Graphics cores as well.

I think you misunderstood his question.

ARM only sells reference implementations and other licenses. They do not manufacturer anything themselves, nor do they contract others to manufacture things for them. They only sell IP.

 

 

11 hours ago, jsjones008 said:

I'd much rather companies get their due for coming up with tech advancements than having cheap imitations that are only cheap because they didn't pay licensing. It's the only thing that keeps companies developing new tech.

 

Think of it from a business standpoint: If I come up with something and someone else can steal it with no repercussions and make money off of it, why should I even develop new technology?

Because there is still a first-move advantage. It's what has driven technological progress for thousands of years.

Not saying that patents are useless, but I think some people are greatly overestimating their necessity.

Progress would not completely tank without them. It would probably slow down but not stop completely, not even from for-profit companies.

 

I mean, there are no food related patents (not counting GMO-related ones) but that is not stopping people from trying out new recipes and opening restaurants.

 

1 hour ago, leadeater said:

And there are already too many giant mega corporations that would have the scales tipped massively towards them, an abrupt removal I would imagine would be rather bad.

I don't think anything bad would happen if software patents were completely removed overnight.

Or well, quite a few patent troll companies would suffer but they deserve it if you ask me (and yes, only patent trolls would suffer, there is no legitimate use for software patents).

 

For regular patents though I think reform should be fairly gradual and the new laws only apply to new patents. Start by making technology related patents only last 15 years. Then in a few decades make them last 10 years. If we 100 years from now think 10 years are too long, reduce them to once more to maybe 5 years.

 

Technology progress has sped up exponentially so it doesn't really make sense to keep century old laws around and think that they don't need to be adapted to modern society. It will just become a bigger and bigger issue the longer we put it off.

[mr moose]

1 hour ago, LAwLz said:

It's very difficult to argue for or against patents because on one hand, companies can claim that without patents they wouldn't invest in developing things.

On the other hand, you can't really measure or check what effect the removal of patents would have on competitors.

 

No patents = more companies can enter the market, thus causing more competition.

 

The question is, would the increased competition be enough to make up for the lack of protection your research? I don't think there is a definitive answer to that, because we can't peek at alternative universes.

 

However, I do think that our current 20 years of government issued monopolies does stifle innovation. I am very much for lowering it to maybe 10 years for technology related patents, plus completely removing the concept of software patents. They are completely illogical and spits in the face of what patents are meant to do and protect.

There just needs to be a balance between what's good for IP owners and what's good for tech advancement overall.  Given everything I have read over the years I am a firm believer that reducing patent terms even for software can only do so much.  Fact of the matter is that humans are greedy by virtue of evolutionary survival instinct, not everyone wants their work copied.

[Stefan Payne]

13 minutes ago, LAwLz said:

I think you misunderstood his question.

ARM only sells reference implementations and other licenses. They do not manufacturer anything themselves, nor do they contract others to manufacture things for them. They only sell IP.

no, they also sell Blueprints for certain Manufacturers. So I'd suspect that they might have some sample Productions.

 

 

And with those Blueprints you can run to the manufacturer they sell it for and manufacture them.

 

[leadeater]

12 minutes ago, LAwLz said:

Because there is still a first-move advantage. It's what has driven technological progress for thousands of years.

Not saying that patents are useless, but I think some people are greatly overestimating their necessity.

Progress would not completely tank without them. It would probably slow down but not stop completely, not even from for-profit companies.

Old technological advancements were funded and driven by either the state/government or, going back really far, tribal needs so basically the same thing. Modern privatization and direct funding is a much newer concept but then so are patents, I view that is mutually inclusive because without both that paradigm as it is now would not exist. Good or bad I don't see the private funding vs state funding being as high as it is now, was a documentary I watched a while ago but it was something like 80% of research is privately funded where it used to be around 10% to 20%.

[LAwLz]

2 minutes ago, mr moose said:

There just needs to be a balance between what's good for IP owners and what's good for tech advancement overall.  Given everything I have read over the years I am a firm believer that reducing patent terms even for software can only do so much.  Fact of the matter is that humans are greedy by virtue of evolutionary survival instinct, not everyone wants their work copied.

Again, I don't want patents in general completely abolished.

I totally agree that they serve a very important purpose.

 

I just think that:

1) The driving force patents provide is greatly overestimated, since some people believe that technology progress would almost entirely die out without it even though there is no historical evidence for that.

 

2) That software patents are completely useless and goes against the purpose of patents in general. Software patents are patents for problems rather than specific solutions (the solutions are protected by copyright as well as having the code closed source). Problems should not be patentable. Only solutions.

 

3) Patents for technology related things, such as electronics, are too long today. It made sense some hundred years ago but it's extremely crippling today. Make them shorter. Maybe 10 years. Being 10 years behind is still very crippling but at least it opens up for companies to improve on things faster.

 

 

 

1 minute ago, leadeater said:

Old technological advancements were funded and driven by either the state/government or, going back really far, tribal needs so basically the same thing. Modern privatization and direct funding is a much newer concept but then so are patents, I view that is mutually inclusive because without both that paradigm as it is now would not exist. Good or bad I don't see the private funding vs state funding being as high as it is now, was a documentary I watched a while ago but it was something like 80% of research is privately funded where it used to be around 10% to 20%.

Again, you're arguing against a position I do not hold.

I am not saying that patents are useless. I completely understand why they are necessary. What I am saying is that they are not as important as some people think. Just read some of the people I have replied to which say technology progress would stop without it.

 

The system you refer to, with state/government sponsored research would still exist.

Product development between companies would still exist (for example Apple funding LG research for screens for iPhones).

Competition would still drive companies to try and one-up each other with first-mover advantages (this part would actually increase compared to what we have today).

 

It would most likely slow down a lot if all patents were completely removed, but I think it would not slow down as much as some people believe.

 

If I had to guess, by pulling numbers out of my ass, I would say progress in a completely patent-less system would slow down maybe 60%, but the pro-patent people on here seem to push for some 90% or more figure.

 

 

All of this is for patents is for normal patents, like the one for capacities touch screens for example. None of what I said applies to software patents, which I genuinely believe are one of the worst things that exist and it serve no purpose other than to harm society.

Filing a lawsuit based on a software patent is not a deadly sin, but it should be.

[mr moose]

3 minutes ago, LAwLz said:

Again, I don't want patents in general completely abolished.

I totally agree that they serve a very important purpose.

 

I just think that:

1) The driving force patents provide is greatly overestimated, since some people believe that technology progress would almost entirely die out without it even though there is no historical evidence for that.

 

2) That software patents are completely useless and goes against the purpose of patents in general. Software patents are patents for problems rather than specific solutions (the solutions are protected by copyright as well as having the code closed source). Problems should not be patentable. Only solutions.

 

3) Patents for technology related things, such as electronics, are too long today. It made sense some hundred years ago but it's extremely crippling today. Make them shorter. Maybe 10 years. Being 10 years behind is still very crippling but at least it opens up for companies to improve on things faster.

 

 

 

1. Maybe not entirely die out but The idea that it wouldn't be majorly hindered is not born out in much of the literature on the subject.  Most of the research (ignoring the private stuff paid for by large companies) still points to patents being the key behind R+D.  It is also common that patents offer small companies and entry level businesses to protect their ides while they are still developing them.

 

https://link.springer.com/article/10.1007/s00712-010-0110-y

http://www.wipo.int/sme/en/documents/ip_innovation_development_fulltext.html

 

I think it is a fair conclusion that patents play a much larger role in development than many would be comfortable to accept.

 

2.That's just your opinion, there are many who feel the opposite.

 

3. referring back to the link I posted in response to 1. patents of a 20 year length have been noted to be sufficient to maintain R+D in furthering technology, longer and R+D wanes whilst shorter and companies stop developing new technologies.
 

Quote

extending the patent length beyond 20 years leads to a negligible increase in R&D despite equilibrium R&D underinvestment. In contrast, shortening the patent length leads to a significant reduction in R&D and consumption

 

 

And this is the conclusion majority of the research finds.   Government funding is no where near sufficient to replace private R+D and if private R+D drops so does government revenue (in the form of tax) thus the problem becomes self perpetuating.

[leadeater]

2 hours ago, LAwLz said:

Again, you're arguing against a position I do not hold.

That wasn't what I was meaning by that comment, not a counter argument just commentary, that without the protection of patents the private funding would be much less, even if protections are reduced. It's not like that is a bad thing but complete state funded research wasn't any better than what we have now either. To take a hot topic example would you want to use a state funded and researched encryption technology?

[LAwLz]

2 hours ago, mr moose said:

1. Maybe not entirely die out but The idea that it wouldn't be majorly hindered is not born out in much of the literature on the subject.  Most of the research (ignoring the private stuff paid for by large companies) still points to patents being the key behind R+D.  It is also common that patents offer small companies and entry level businesses to protect their ides while they are still developing them.

Just to be clear. I never said it would not be "majority hindered".

In fact, I said quite the opposite.

 

2 hours ago, mr moose said:

2.That's just your opinion, there are many who feel the opposite.

I don't think it's much of an opinion at this point. I can't think of a single valid argument for why software patents should exist.

Not only is it illogical to be able to own mathematical formulas, they are patents for problems rather than solutions.

Patents are meant to protect inventions. Software patents aren't written so that they protect the specific implementation of something. They are written so that they define the result.

That is bullshit.

The specific implementation, the thing that patents typically protect, is already protected through copyright as well as developers having the option to make something closed source. The patent is completely useless for protecting the specific implementation or method of performing something in software.

 

Copyright and making the source code closed already fulfill all the intended purposes of patents.

 

Patenting things like "rubberband effect when you hit the end of a list" is as bizarre as patenting movies with romantic elements in it. I am all for claiming copyright on a specific movie, but you shouldn't be able to patent the concept of something (which is what software patents are).

 

 

2 hours ago, mr moose said:

3. referring back to the link I posted in response to 1. patents of a 20 year length have been noted to be sufficient to maintain R+D in furthering technology, longer and R+D wanes whilst shorter and companies stop developing new technologies.

Not going to pay 44 euro to read what he has to say about shortening the 20 year monopoly granted by patents could harm innovation.

Do you have the part of the paper?

Please note that I have deliberately said that I think technology related patents should be shortened. I have no problem understanding why patents in other areas benefits from being longer. It's specifically software patents I want to get rid of completely, and specifically technology related patents that I believe should be shortened.

 

 

 

1 hour ago, leadeater said:

To take a hot topic example would you want to use a state funded and researched encryption technology?

Sure

AES is one of my favorite cryptos and it was state funded.

It was two researchers at KU Leuven (more specifically the COSIC department) in Belgium who developed it, submitted it to NIST and then got it approved.

 

I also think that Tor is a really well designed and useful tool, and it was also developed and funded by the US government.

 

As long as it is open source, transparent in how it works and doesn't rely on some chain-of-trust the government controls, then I am completely fine with using it.

[AlTech]

1 hour ago, leadeater said:

To take a hot topic example would you want to use a state funded and researched encryption technology?

No. I despise big brother already spying on me .

[mr moose]

8 hours ago, LAwLz said:

Just to be clear. I never said it would not be "majority hindered".

In fact, I said quite the opposite.

 

I don't think it's much of an opinion at this point. I can't think of a single valid argument for why software patents should exist.

Not only is it illogical to be able to own mathematical formulas, they are patents for problems rather than solutions.

Patents are meant to protect inventions. Software patents aren't written so that they protect the specific implementation of something. They are written so that they define the result.

That is bullshit.

The specific implementation, the thing that patents typically protect, is already protected through copyright as well as developers having the option to make something closed source. The patent is completely useless for protecting the specific implementation or method of performing something in software.

 

Copyright and making the source code closed already fulfill all the intended purposes of patents.

 

Patenting things like "rubberband effect when you hit the end of a list" is as bizarre as patenting movies with romantic elements in it. I am all for claiming copyright on a specific movie, but you shouldn't be able to patent the concept of something (which is what software patents are).

The rubber band effect might be pushing the boundaries a bit I would agree, but software patents can't just be dismissed as math.  Unlike math, software is also an end product that does a specific job, it's existence is the result of hard work and carries a value that business who invest in it rely on to survive.  Copyright is IP for art (expression) if someone changes your code sufficiently then it is no longer covered by copyright even though the changes code still does the same job,  Patent covers function, so a combination of copyright and patent for software is needed when people create software as a functioning product.

 

8 hours ago, LAwLz said:

 

Not going to pay 44 euro to read what he has to say about shortening the 20 year monopoly granted by patents could harm innovation.

Do you have the part of the paper?

Please note that I have deliberately said that I think technology related patents should be shortened. I have no problem understanding why patents in other areas benefits from being longer. It's specifically software patents I want to get rid of completely, and specifically technology related patents that I believe should be shortened.

 

 

No I don't,  But it's not hard to sift through the springer library or any similar journal archive for more.

 

But again when it comes to software patents, that is a debate between opinions mostly,  I can see the argument against patents on things like rubber band effect, but I can also see the desire for people to want to protect their work and ensure they can enjoy the fruits of it.   Especially in context with this specific case where the patent is a process for a functioning product.  One that took years to develop and perfect. 

[LAwLz]

On 8/27/2018 at 10:54 PM, mr moose said:

The rubber band effect might be pushing the boundaries a bit I would agree, but software patents can't just be dismissed as math.  Unlike math, software is also an end product that does a specific job, it's existence is the result of hard work and carries a value that business who invest in it rely on to survive.  Copyright is IP for art (expression) if someone changes your code sufficiently then it is no longer covered by copyright even though the changes code still does the same job,  Patent covers function, so a combination of copyright and patent for software is needed when people create software as a functioning product.

The problem is that software patents are patents for problems rather than solutions.

 

I would have no problem with software patents if they actually patented the code, the function, which results in something, but that's not what software patents are. (Actually, I would still have a problem with them but not as big of a problem as I do right now).

Software patents, are for example "a rubber banding effect when you reach the end of a list". No matter what code you write or how you write it, if it makes the screen rubberband when you reach the end of a list, it infringes on Apple's patent.

 

I am all for Apple owning the code which creates that effect, and I think they are entitled to keeping it a secret (closed source) or not have it copied (copyright). What I don't agree with is that they should have a government issued monopoly on any code which creates a similar effect.

 

 

If you want an analogy.

A patent for a lock is a patent which describes exactly how the lock works, down to the mechanisms.

If a lock patent was written like a software patent it would say "method of preventing access to a location" and that would encompass every single lock ever created, as well as several other things which aren't locks.

 

If Apple wants to own their rubberbanding effect then let them. But they should only be able to own their own version of rubberbanding. Not every single version and method anyone can come up with.

 

 

Worth noting that some countries has completely abolished software patents (such as France and New Zealand). In the US it's a complete mess, with rulings such as the famous Alice case ruling that you can not patent abstract ideas (which is what software patents are), and has been used as precedent for hundreds of cases involving software patents (to successfully invalidate them).

 

Also, since you live in Australia I think it's worth pointing out that while software patents aren't explicitly invalid in Australia, it has only been discussed four times, and on all four occasions the software in question was determined to not be eligible for patenting.

 

 

I know that you like siding with companies, but software patents are a load of bollocks which only costs a ton of money and stifles innovation. The only companies that likes them are the massive ones like Google, Apple, Facebook, Microsoft etc, or patent trolls. Everyone else hates them.

 

 

If someone want some more reading, I strongly recommend this excellent article in the Washington Post, which goes over why software patents are so different from other patents, as well as a lot of statistics. It was written in 2013 and since then the problem has grown way bigger.

 

 

Another interesting read for those interested is this blog post from Joel Spolsky, the co-founder of Stack Overflow, titled "Victory Lap for Ask Patents". His conclusion is that almost all software patents are shit with a ton of prior art and deliberately written in obscure terms to hopefully slip through the cracks in the patent office, and be used to sue someone.

[mr moose]

13 minutes ago, LAwLz said:

The problem is that software patents are patents for problems rather than solutions.

 

I would have no problem with software patents if they actually patented the code, the function, which results in something, but that's not what software patents are.

Software patents, are for example "a rubber banding effect when you reach the end of a list". No matter what code you write or how you write it, if it makes the screen rubberband when you reach the end of a list, it infringes on Apple's patent.

 

I am all for Apple owning the code which creates that effect, and I think they are entitled to keeping it a secret (closed source) or not have it copied (copyright). What I don't agree with is that they should have a government issued monopoly on any code which creates a similar effect.

 

 

If you want an analogy.

A patent for a lock is a patent which describes exactly how the lock works, down to the mechanisms.

If a lock patent was written like a software patent it would say "method of preventing access to a location" and that would encompass every single lock ever created, as well as several other things which aren't locks.

 

If Apple wants to own their rubberbanding effect then let them. But they should only be able to own their own version of rubberbanding. Not every single version and method anyone can come up with.

 

 

Worth noting that some countries has completely abolished software patents (such as France and New Zealand). In the US it's a complete mess, with rulings such as the famous Alice case ruling that you can not patent abstract ideas (which is what software patents are), and has been used as precedent for hundreds of cases involving software patents (to successfully invalidate them).

 

Also, since you live in Australia I think it's worth pointing out that while software patents aren't explicitly invalid in Australia, it has only been discussed four times, and on all four occasions the software in question was determined to not be eligible for patenting.

 

 

I know that you like siding with companies, but software patents are a load of bollocks which only costs a ton of money and stifles innovation. The only companies that likes them are the massive ones like Google, Apple, Facebook, Microsoft etc, or patent trolls. Everyone else hates them.

 

 

If someone want some more reading, I strongly recommend this excellent article in the Washington Post, which goes over why software patents are so different from other patents, as well as a lot of statistics. It was written in 2013 and since then the problem has grown way bigger.

That's great and all, but you are still conflating the math side of software patents with the product side. These are two separate components.  No patents stand for any device that is too general, all patents (software or otherwise have to be specific), it is of no surprise that people are having patents for general software solutions knocked back.  Your analogy about the lock stands very true, any product (software or otherwise) that is general and does not solve or perform a specific task will not get a patent (or at least shouldn't and the generic ones that are in the system need to be weeded out).  The issue is the grey area in between a product and a general result.   I.E rubber band effect is an artistic expression, it follows the same lines design and presentation of a product, which are patent-able (whether ewe like it or not).   Now I agree they are crossing the line to a degree with such things, but the fact is how a product looks is as unique as a trademark.

 

With regard to patent trolls, no one likes them, they are the bane of all industries. and to Me I feel if a company does not manufacture a product that uses a patent they own then they can't sue because others do.  That would stop majority of patent trolls.  

[LAwLz]

4 minutes ago, mr moose said:

That's great and all, but you are still conflating the math side of software patents with the product side.

No I am not. I think the problem here is that you think software patents are like regular patents. They aren't. Several countries has explicitly made distinctions between regular and software patents.

 

6 minutes ago, mr moose said:

No patents stand for any device that is too general, all patents (software or otherwise have to be specific)

No. That is where software patents differs from regular patents.

To me, a software patent that is specific would include the code. You can't be specific without actually saying how you accomplish something. However, software patents do not include the code (since that's a copyright thing, not patent thing).

To take the lock analogy again, you can't say the patent described as "method of preventing access to a location" is specific without actually showing how the mechanical component of such a system works. The "mechanical component" of software is the code.

 

How about this. In order to submit a software patent you need to also post the source code as an example of an implementation.

That sounds fair, right? When you submit anything else for patenting, such as chemical compositions or mechanical devices then you need to be specific in how they work. You used to actually need a working, physical version of what you patented. Make sense that software should have the same standards if they want the same type of protection, right?

 

 

13 minutes ago, mr moose said:

With regard to patent trolls, no one likes them, they are the bane of all industries. and to Me I feel if a company does not manufacture a product that uses a patent they own then they can't sue because others do.  That would stop majority of patent trolls.  

Wouldn't work because a lot of businesses entirely rely on licensing designs and patents.

For example with your proposal ARM would not exist. Their entire business is based on owning patents for things, and then licensing that to others. They don't manufacture anything themselves. They just sell IP.

(Important to note that what they sell are not software patent licenses).

[mr moose]

53 minutes ago, LAwLz said:

No I am not. I think the problem here is that you think software patents are like regular patents. They aren't. Several countries has explicitly made distinctions between regular and software patents.

 

I don't think that they are for the most part.  Some countries have extra conditions because their patent systems isn't sufficient for some of the more intrinsic characteristics of software, but there is no one definition that isolates software patents from other patents.  Most jurisdictions have rules that follow the same lines, i.e you can;t patent an abstract idea, your patent must have a technical effect outside of software or a process that is new/unique etc.  In fact the Full high court of Australia has effectively ruled that you cannot define a what is patent-able by terms like software or define what constitutes manufacturing because it severely limits the courts ability to determine fairness in applications regarding IP.

 

http://www.davies.com.au/ip-news/federal-court-reviews-patentability-of-business-methods-in-australia

 

Quote

No. That is where software patents differs from regular patents.

To me, a software patent that is specific would include the code. You can't be specific without actually saying how you accomplish something. However, software patents do not include the code (since that's a copyright thing, not patent thing).

To take the lock analogy again, you can't say the patent described as "method of preventing access to a location" is specific without actually showing how the mechanical component of such a system works. The "mechanical component" of software is the code.

Sometimes they do sometimes they don't, but we are talking about what actually happens not your perspective on what it should be.

Quote

How about this. In order to submit a software patent you need to also post the source code as an example of an implementation.

That sounds fair, right? When you submit anything else for patenting, such as chemical compositions or mechanical devices then you need to be specific in how they work. You used to actually need a working, physical version of what you patented. Make sense that software should have the same standards if they want the same type of protection, right?

Well, If the patent application meets the criteria for a patent (i.e not abstract, has a actual effect and can be specifically described) then I don't see why the code needs to be included, although enough of it probably already is in most cases.

 

Quote

 

Wouldn't work because a lot of businesses entirely rely on licensing designs and patents.

For example with your proposal ARM would not exist. Their entire business is based on owning patents for things, and then licensing that to others. They don't manufacture anything themselves. They just sell IP.

(Important to note that what they sell are not software patent licenses).

Except they do produce a product, they don't just own the IP and license it out,  they develop the actual processors and software that goes with it.  That design is a product in and of itself.   They are like those small business that design trailers and sell the plans online.  The plans are copyright because the plans constitute a product even though they don't physically manufacture the trailer.  

 

I am talking about patent trolls being the companies that treat patents as buying the right to sue people, they don't do anything with them except license and sue.   ARM design and validate their IP, Nvidia manufacture their IP, AMD manufacture their IP, IBM develop and validate their IP, so on and so forth.

[LAwLz]

8 hours ago, mr moose said:

you can;t patent an abstract idea

That is what 99.9% of all software patents are. Like the co-founder of Stack Overflow said, all software patents he has seen are bad. Lawyers just ask developers what they are working on, brings out a thesaurus, replace some words and shuffle them around a bit and then push a patent on it, hoping that nobody in the patent office notices what a load of bollocks that patent is.

Remember that patents can not have any prior art either, but almost all software do if you look hard and long enough.

 

Things like "rubberbanding effects" are abstract ideas unless they explicitly define how they work through for example code examples.

If you can't show how whatever you're patenting is implemented then it's an abstract idea.

If you can show it but refuse to do so and instead offer a vague description of the effect the formula has, rather than the actual formula, then you shouldn't be eligible for the patent in my opinion.

 

 

8 hours ago, mr moose said:

Some countries have extra conditions because their patent systems isn't sufficient for some of the more intrinsic characteristics of software, but there is no one definition that isolates software patents from other patents. 

While there is no universal definition for software patents, even you seem to agree that there are fundamental differences between a traditional patent and a software related one.

What I think needs to happen is that a consortium sit down, define software patents and then outlaw them. Maybe just adopt the New Zealand or French laws. The problem is that it is extremely heavily lobbied by a handful of companies, since those are the only ones benefiting from this completely broken system.

It would be a massive victory for progress and society as a whole.

 

 

8 hours ago, mr moose said:

Sometimes they do sometimes they don't, but we are talking about what actually happens not your perspective on what it should be.

You make it sound like it is common for them.

I've looked at quite a few software patents and none of them so far have had example code. If I had to guess, I'd say less than 1% do, and I think it should be mandatory just like it is mandatory to describe the intricate mechanical functions when applying for a patent for something like a lock.

Also, I don't appreciate that you're telling me I can't give my opinion on something.

 

8 hours ago, mr moose said:

Well, If the patent application meets the criteria for a patent (i.e not abstract, has a actual effect and can be specifically described) then I don't see why the code needs to be included, although enough of it probably already is in most cases.

Because without the source code it is an abstract idea.

Like the rubberbanding effect, or the lock example.

And no, code is next to never included. I haven't been able to find a single example (although I have not exactly been looking for it specifically either)-

 

If you can't show how something works, then it is abstract.

Abstract is defined as:

• Disassociated from any specific instance

If you can't show a specific instance of a software patent being implemented, that is to say, show the source code, then it is abstract. That is why the rubberbanding patent, and most other software patents, cover code that has not been written yet. Because they are abstract ideas which dramatically different code all infringe upon.

 

8 hours ago, mr moose said:

Except they do produce a product, they don't just own the IP and license it out,  they develop the actual processors and software that goes with it.  That design is a product in and of itself.   They are like those small business that design trailers and sell the plans online.  The plans are copyright because the plans constitute a product even though they don't physically manufacture the trailer.  

 

I am talking about patent trolls being the companies that treat patents as buying the right to sue people, they don't do anything with them except license and sue.   ARM design and validate their IP, Nvidia manufacture their IP, AMD manufacture their IP, IBM develop and validate their IP, so on and so forth.

It would be trivial to get around your proposal in that case.

A concept program using a software patent can often be created in like 15 minutes. So all a patent troll company would have to do to get around your proposal is create a snippet of code, claim that's their product and then seek licensing fees from other companies.

In practice, it is no different from what ARM does when they create a test chip to check if their design is working, and then license it to everyone else.

 

There is no practical difference between what ARM does and what patent trolls does.

The difference is all based on perceived intention. Basically, "I think they are being jerks" and that won't hold up as evidence in court.

[mr moose]

28 minutes ago, LAwLz said:

That is what 99.9% of all software patents are. Like the co-founder of Stack Overflow said, all software patents he has seen are bad. Lawyers just ask developers what they are working on, brings out a thesaurus, replace some words and shuffle them around a bit and then push a patent on it, hoping that nobody in the patent office notices what a load of bollocks that patent is.

Remember that patents can not have any prior art either, but almost all software do if you look hard and long enough.

 

Things like "rubberbanding effects" are abstract ideas unless they explicitly define how they work through for example code examples.

If you can't show how whatever you're patenting is implemented then it's an abstract idea.

If you can show it but refuse to do so and instead offer a vague description of the effect the formula has, rather than the actual formula, then you shouldn't be eligible for the patent in my opinion.

I think you better look up the definition of abstract, especially as it applies to patents, the  rubber band effect does have a physical presence/existence outside of thought. It is literally the implementation of what happens to the screen when you hit the end. Every time someone scrolls to the end of the menu on the ipad it bounces back, ergo a tangible result in a product.   Most software is the same, the patent is not the idea but the implementation.

 

28 minutes ago, LAwLz said:

 

While there is no universal definition for software patents, even you seem to agree that there are fundamental differences between a traditional patent and a software related one.

Of course, so do many of the legal and patents systems around the world, someone in the office stamps a patent with out considering the content probably happens  a lot more than we know, but that is a mistake on part of the person in the office not with the need for patents.

28 minutes ago, LAwLz said:

What I think needs to happen is that a consortium sit down, define software patents and then outlaw them.

Why bother having the consortium if the only outcome you would like is to have them outlawed?

28 minutes ago, LAwLz said:

Maybe just adopt the New Zealand or French laws. The problem is that it is extremely heavily lobbied by a handful of companies, since those are the only ones benefiting from this completely broken system.

It would be a massive victory for progress and society as a whole.

Only in your opinion.  software is as much a product as anything else, why shouldn't it be protected?

 

28 minutes ago, LAwLz said:

 

You make it sound like it is common for them.

I've looked at quite a few software patents and none of them so far have had example code. If I had to guess, I'd say less than 1% do, and I think it should be mandatory just like it is mandatory to describe the intricate mechanical functions when applying for a patent for something like a lock.

Also, I don't appreciate that you're telling me I can't give my opinion on something.

They don't need sample code, if they can articulate what it does and how it does it you don't need the code. 

 

28 minutes ago, LAwLz said:

Because without the source code it is an abstract idea.

Like the rubberbanding effect, or the lock example.

And no, code is next to never included. I haven't been able to find a single example (although I have not exactly been looking for it specifically either)-

You don;t need the code to show what it does. You could write a program to say hello whenever the T button is pushed and goodbye for every other button, but as I have just described that function to you without showing you the code it stands to reason you can define and articulate the purpose of a program without presenting the code.  It does not suddenly become an intangible block of data just because the code isn't provided.  We all know what the rubber band effect does without seeing the code.

 

28 minutes ago, LAwLz said:

If you can't show how something works, then it is abstract.

Nope, not at all.

28 minutes ago, LAwLz said:

Abstract is defined as:

• Disassociated from any specific instance

If you can't show a specific instance of a software patent being implemented, that is to say, show the source code, then it is abstract. That is why the rubberbanding patent, and most other software patents, cover code that has not been written yet. Because they are abstract ideas which dramatically different code all infringe upon.

 

Except they can show an example of the rubber band effect on their product. That makes it not an abstract concept, it is not divorced from any specific instance because it is demonstrable real.  I don't even know how you can't see that.

 

28 minutes ago, LAwLz said:

It would be trivial to get around your proposal in that case.

A concept program using a software patent can often be created in like 15 minutes. So all a patent troll company would have to do to get around your proposal is create a snippet of code, claim that's their product and then seek licensing fees from other companies.

In practice, it is no different from what ARM does when they create a test chip to check if their design is working, and then license it to everyone else.

 

There is no practical difference between what ARM does and what patent trolls does.

The difference is all based on perceived intention. Basically, "I think they are being jerks" and that won't hold up as evidence in court.

What you have just said is illogical,  are you seriously trying to compare writing a 15 minute program to designing an ARM processor? let alone using that comparison as an argument as to why the ARM design is not a legitimate product?

[LAwLz]

21 minutes ago, mr moose said:

I think you better look up the definition of abstract, especially as it applies to patents, the  rubber band effect does have a physical presence/existence outside of thought. It is literally the implementation of what happens to the screen when you hit the end. Every time someone scrolls to the end of the menu on the ipad it bounces back, ergo a tangible result in a product.   Most software is the same, the patent is not the idea but the implementation.

The method used to achieve the result is abstract. It is not defined in the patent, only the result is defined.

The patent SHOULD be the implementation, but in the case of software patents they are patenting the result rather than the implementation.

 

21 minutes ago, mr moose said:

Why bother having the consortium if the only outcome you would like is to have them outlawed?

Because we would need a proper definition to determine exactly what should be outlawed. Like what happened in New Zealand but on a global scale.

 

21 minutes ago, mr moose said:

Only in your opinion.  software is as much a product as anything else, why shouldn't it be protected?

Because:

1) It is math, and you can't (or at least shouldn't) be able to patent math.

 

2) It leads to patent thickets.

 

3) It is extremely detrimental to progress and rather than encourage innovation, it actually discourages people from doing R&D because the risk of infringing someone's patent is too high.

 

4) There are already two true and tested ways of protecting your software from copycasts. Copyright and closing the source code off.

 

5) Software patents are defined entirely through their functions, unlike traditional patents which defines physical things (like the exact ingredients in a recipe, the specific components of a mechanical device, the exact molecular structure of a drug, and so on).

 

6) Software is a component of a computer. I can already do the things which future patents say I can't do because they invented it. Software patents are merely a new way of utilizing something that already exists. You can't patent the idea of let's say sleeping on a ladder, because the ladder already exists and can be used however someone choose to use it already.

 

7) It is by far the most commonly used tool by patent trolls.

 

8) Software patent usually has a very short time period where they are relevant. If someone patents a function meant to be used in for example a compression algorithm, chances are the program which uses that patent will only be relevant for a couple of years. On average it takes 2.5 years for a software patent to be approved. It is very likely that the entire product lifecycle is over before the patent is even approved, and at that point it can only be used for attacking others rather than defend your own product.

 

I can probably think of more reasons but they will more or less just be iterations of "it wastes time and money".

 

 

44 minutes ago, mr moose said:

They don't need sample code, if they can articulate what it does and how it does it you don't need the code. 

If you apply for a patent on let's say a soda, you need to disclose the exact recipe.

The source code is the recipe for a program. Why should we not hold software patents to the same standards as any other patent?

 

46 minutes ago, mr moose said:

You don;t need the code to show what it does. You could write a program to say hello whenever the T button is pushed and goodbye for every other button, but as I have just described that function to you without showing you the code it stands to reason you can define and articulate the purpose of a program without presenting the code.  It does not suddenly become an intangible block of data just because the code isn't provided.  We all know what the rubber band effect does without seeing the code.

We know what it does, but we don't know how it does it, which is in fact one of the requirements for patents. In order to have a patent granted, you need to fulfill the enablement requirement.

Without the source code, this is not possible. Someone can make something similar, but they can't make the same without pure guessing, unless the source code is provided.

 

51 minutes ago, mr moose said:

What you have just said is illogical,  are you seriously trying to compare writing a 15 minute program to designing an ARM processor? let alone using that comparison as an argument as to why the ARM design is not a legitimate product?

I never said anything even remotely close to that.

I never said designing an ARM processor was the same as a 15 minute coding job.

I never said ARM's designs aren't legitimate products.

 

What I said was that your proposal, that patents are only valid if the patent owners use them in a product, wouldn't help solve the issue of patent trolls at all.

All the software patent troll has to do is spend 15 minutes coding a simple program and then they would have circumvented your rule. And if you try and enforce stricter requirements like they have to actively sell the product then you run into issues where you're harming companies like ARM, whose entire business model is based on selling licenses.

 

In practice, there is no difference between ARM selling a license to someone, and a patent troll who spent 15 minutes developing a POC.

The only difference would be the amount of effort spent developing the "product" and the intention behind the development. Neither of those were taken into consideration in your proposal and neither of them can be objectively quantified and used as proper arguments in a court.

[suicidalfranco]

i find it laughable that in the US you can put a patent on the result instead of the solution. 

what i mean is:

problem: how to put a text that says "hello world!"

solution:

#include <iostream>
using namespace std;

int main() 
{
    cout << "Hello, World!";
    return 0;
}

result: "Hello, World!" <- patent this

 

then when someone else comes and decide to solve it like this:

#include <iostream>

int main() 
{
    std::cout << "Hello, World!";
    return 0;
}

result: "Hello, World!"

 

Somehow they can claim patent infringement.

[mr moose]

8 minutes ago, LAwLz said:

The method used to achieve the result is abstract. It is not defined in the patent, only the result is defined.

The patent SHOULD be the implementation, but in the case of software patents they are patenting the result rather than the implementation.

That's not how patents work though, many everyday patents refer to ideas that exist only as a plan, biscuits are an abstract by your definition.  There are many ways to cook biscuits, it's not just the process that you patent, it's the outcome too,  if you don't have an outcome or product at the end then it is abstract.

8 minutes ago, LAwLz said:

Because we would need a proper definition to determine exactly what should be outlawed. Like what happened in New Zealand but on a global scale.

 

Because:

1) It is math, and you can't (or at least shouldn't) be able to patent math.

 

2) It leads to patent thickets.

 

3) It is extremely detrimental to progress and rather than encourage innovation, it actually discourages people from doing R&D because the risk of infringing someone's patent is too high.

 

4) There are already two true and tested ways of protecting your software from copycasts. Copyright and closing the source code off.

 

5) Software patents are defined entirely through their functions, unlike traditional patents which defines physical things (like the exact ingredients in a recipe, the specific components of a mechanical device, the exact molecular structure of a drug, and so on).

 

6) Software is a component of a computer. I can already do the things which future patents say I can't do because they invented it. Software patents are merely a new way of utilizing something that already exists. You can't patent the idea of let's say sleeping on a ladder, because the ladder already exists and can be used however someone choose to use it already.

 

7) It is by far the most commonly used tool by patent trolls.

 

8) Software patent usually has a very short time period where they are relevant. If someone patents a function meant to be used in for example a compression algorithm, chances are the program which uses that patent will only be relevant for a couple of years. On average it takes 2.5 years for a software patent to be approved. It is very likely that the entire product lifecycle is over before the patent is even approved, and at that point it can only be used for attacking others rather than defend your own product.

1. Not just math, but carefully designed math to achieve a specific result using a specific platform.  Remember that design is also patent-able.

2. sometimes, but that's not a reason or else you could use the same argument to ban torrents, because it leads to piracy.

3. That's just one side of the argument, I have already posted enough evidence to show that is at best a heavily disputed issue let alone a fact.

4. I have already explained why the copyright systems fails software already. copyright has it's limits once something cease to be expressive or artistic.

5. no, traditional patents also include defining functions not just methods. 

6. that's a flawed concept, if you want to use that logic then you can't patent anything that uses or works with an already existing device.  That's just not a good reason to disallow a patent. 

7. So? 

8. which one is it?  is it only good for defending your product or is it not worth getting becasue it takes too long?  How can it be only good for the one thing it is designed to do if it takes too long to get making defending your product moot?

 

8 minutes ago, LAwLz said:

I can probably think of more reasons but they will more or less just be iterations of "it wastes time and money".

 

 

If you apply for a patent on let's say a soda, you need to disclose the exact recipe.

The source code is the recipe for a program. Why should we not hold software patents to the same standards as any other patent?

 

We know what it does, but we don't know how it does it, which is in fact one of the requirements for patents. In order to have a patent granted, you need to fulfill the enablement requirement.

Without the source code, this is not possible. Someone can make something similar, but they can't make the same without pure guessing, unless the source code is provided.

 

Patents can be for functions not just methods.

https://www.entrepreneur.com/encyclopedia/patents

 

So long as they perform a function not previously performed and said function is not obvious.  There is no clause that stipulates ingredients have to be listed, especially  as most software patents will come under design patents.

 

8 minutes ago, LAwLz said:

I never said anything even remotely close to that.

I never said designing an ARM processor was the same as a 15 minute coding job.

I never said ARM's designs aren't legitimate products.

 

What I said was that your proposal, that patents are only valid if the patent owners use them in a product, wouldn't help solve the issue of patent trolls at all.

All the software patent troll has to do is spend 15 minutes coding a simple program and then they would have circumvented your rule. And if you try and enforce stricter requirements like they have to actively sell the product then you run into issues where you're harming companies like ARM, whose entire business model is based on selling licenses.

 

In practice, there is no difference between ARM selling a license to someone, and a patent troll who spent 15 minutes developing a POC.

The only difference would be the amount of effort spent developing the "product" and the intention behind the development. Neither of those were taken into consideration in your proposal and neither of them can be objectively quantified and used as proper arguments in a court.

Again, ARM don't just own IP and license it, they actually do a low of work updating and maintaining the design of ARM processors, it is a lot more than just licensing.