9 ms·

No Man's Sky sued over procedural generation algorithm patent

26 points by chriswwweb 10y ago

FreeFull 10y ago
I don't personally feel like something like this should be patentable, but Gielis is definitely following both the rule and spirit of the patent law. I hope Hello Games and Genicap can come to some sort of agreement. Edit: As another comment notes, the patent might have expired in 2006 due to fee non-payment, in which case Gielis doesn't have much of a standing. Edit 2: Nevermind, Gielis does seem to hold a valid patent: https://register.epo.org/application?number=EP00929732 https://register.epo.org/application?number=EP00929732
- Dylan16807 10y ago
  It describes a circle with parameterized wobbles? That is a pure mathematics problem, not in the realm of patents at all. Why are you certain they're acting validly? It seems misleading to even call it an algorithm.
  - jjgreen 10y ago
    It infringes on my x^2 + y^2 = 1 patent, as does the letter 'o'.
  - kale 10y ago
    I deal routinely with use patents in the medical field. There's a chemical that's been in use for 30+ years to stabilize polymers from ionizing radiation. It's used in practically all gallon milk jugs to protect them from UV degradation. A company has a bright idea in 2006 that this chemical would also stabilize medical device plastics during sterilization (usually gamma radiation), and get a use patent on this stabilizer for use in the medical industry. 30+ year old technology, used in several fields, but requires licensing to use it in medical devices. And this is a world patent. I'm guessing the patent would be a use patent. The patent is on a specific use of the math (although it doesn't sound all that specific from the claims that I've seen).
- sqldba 10y ago
  Well, if it's an algorithm as the article alleges - compression is an algorithm and so are video codecs. They can be patented. Why should this be treated differently?
  - galadran 10y ago
    Why should video codecs be patentable? I'm well aware they are currently but it seems ludicrous that you can patent the performance of a sequence of mathematical operations.
  - Dylan16807 10y ago
    Video codec patents are an odd example to use, as they are currently doing a great job of stifling the state of the art, rather than promoting it.
    - sqldba 10y ago
      I'm not up to date on the complexities of patent vs copyright, but I'll assume you're against both? It just seems to me if you can write software you should be able to own it. If you can own it then you get to say where and how it is used and what compensation you want. Isn't an algorithm or a codec a lot like saying a program? Why shouldn't people be able to own the fruit of their labors?
      - Dylan16807 10y ago
        People don't get to own the math they invent, or the mental processes they invent. I think the same should apply to algorithms in general. You don't get to control people's thoughts, and if someone is allowed to think something, they should be allowed to have a computer assist them in thinking it. Copyright is fine, if we ignore the recent length extensions. If you write software, you get to say who uses that software. But you don't get to say who uses similar software.
        seanp2k2 10y ago
        Consider also how this applies to AI. It's very possible that in our lifetimes (I'm assuming you'll live for the next 40 years) we'll see strong AI, possibly capable of inventing novel things by itself or with human collaboration. Who owns the patents which will inevitably result from this? What if the AI has an opinion on the matter? Should the AI own the patents? Is someone allowed to own the AI?
- mrmondo 10y ago
  Couldn't agree more, for f@&$! sake people, we're trying to evolve, form cross-community, intercontinental collaboration and move forward together as a species but a messed up financial model (particularly in the USA) is holding society back. We almost all know it, it's easy to say but hard to change. I think that the key is to keep the conversation current, transparent and relevant to the general public and that is the only way blockers like patents will ever go away.
  - stevenwiles 10y ago
    Are you saying that if I come up with a good idea, I don't have a right to protect that idea?
    - angrow 10y ago
      The purpose of copyright and patent laws is to give the greatest benefit to society (in the American constitution, the phrasing is "to promote the useful arts and sciences"). The granting of temporary monopolies to registrants is a common tactic used to achieve that end, but not the only possible one, and one that admits of many variations (terms, extensions, and limitations vary across jurisdictions and over time). Maybe some people think that the exact laws we have right now are the best possible at serving the underlying purpose of the long-term enrichment of the public domain, but many seem to think that improvements could be made.
    - ewzimm 10y ago
      Let's say you come up with a great idea, a universal cure for cancer, but it requires growing human stem cells in a lab and you're against that for religious reasons, so you patent it and make sure nobody ever uses it. There are a few issues here. First, you definitely didn't come up with that idea from nothing. You "stood on the shoulders of giants" and profited from thousands of years of scientific development. You almost certainly collaborated with lots of other people who might not have shared your personal values and contributed just as much or more than you did to its development. You happened to be the person who connected the final piece. Just how much control should that give you over the use of the idea? Some people would say it's hard to justify your absolute control over the use of that idea.
- chriswwweb 10y ago
  I totally aggree, patents on an algorithm are really not a good idea Yeah the patent is still registered in some european countries but not all of them, in some countries it has expiered, don't know what this means from a juridical point of view though Here is the full patent description: http://www.google.com/patents/EP1177529A1?cl=un http://www.google.com/patents/EP1177529A1?cl=un
MichaelBurge 10y ago
It looks like this is the patent, and that it was withdrawn because they stopped paying the fees: https://register.epo.org/application?number=EP05100648&lng=en&tab=main https://register.epo.org/application?number=EP05100648&lng=e... Assuming this is the same patent, either Gielis is so irresponsible and disorganized he didn't realize his patent expired, or he's just trying to rustle up some bad press to extort money out of them.
- FreeFull 10y ago
  Following the link from the article gives this: https://register.epo.org/application?number=EP00929732 https://register.epo.org/application?number=EP00929732
SlashmanX 10y ago
> creator Sean Murray acknowledged in a 2015 New Yorker interview that he had struggled with elements of procedural planetary generation, until he discovered an equation published in 2003 by Belgian plant geneticist Johan Gielis that he called “Superformula.” I sense a settlement (or partnership) in the near future
- runevault 10y ago
  So the fact he didn't even reinvent it but used the paper makes this an entirely different case to me. Hm.
speps 10y ago
The paper with the "Superformula" : http://www.amjbot.org/content/90/3/333.full http://www.amjbot.org/content/90/3/333.full
asherkin 10y ago
> Company claims No Man's Sky uses its patented equation without permission is pretty damn different from > No Man's Sky sued over procedural generation algorithm patent
anotheryou 10y ago
Why as this formular so cool? Cooking it the shapes on the wikipedia shapes, once unraveled from the circular shape the curves don't look too interesting to me. But these might be simple examples...
conmarap 10y ago
The superforumula is public knowledge and these guys are clowns. Dumb patent trolls... https://en.wikipedia.org/wiki/Superformula https://en.wikipedia.org/wiki/Superformula
- fsloth 10y ago
  Your linked wikipedia page states that it's patented. Yes, it's silly one can do that but it gives some legal credibility to the case.
chriswwweb 10y ago
UPDATE: Hello Games has finally posted a response to the patent infringement claims, this is what they tweeted: "No Man's Sky doesn't actually use this "superformula" thing or infringe a patent. This is a non-story... everybody chill (シ_ _)シ" Tweet is here: https://twitter.com/NoMansSky/status/756889227095318528 https://twitter.com/NoMansSky/status/756889227095318528

26 points by chriswwweb 10y ago

I don't personally feel like something like this should be patentable, but Gielis is definitely following both the rule and spirit of the patent law. I hope Hello Games and Genicap can come to some sort of agreement. Edit: As another comment notes, the patent might have expired in 2006 due to fee non-payment, in which case Gielis doesn't have much of a standing. Edit 2: Nevermind, Gielis does seem to hold a valid patent: https://register.epo.org/application?number=EP00929732 https://register.epo.org/application?number=EP00929732

Dylan16807 10y ago
It describes a circle with parameterized wobbles? That is a pure mathematics problem, not in the realm of patents at all. Why are you certain they're acting validly? It seems misleading to even call it an algorithm.
- jjgreen 10y ago
  It infringes on my x^2 + y^2 = 1 patent, as does the letter 'o'.
- kale 10y ago
  I deal routinely with use patents in the medical field. There's a chemical that's been in use for 30+ years to stabilize polymers from ionizing radiation. It's used in practically all gallon milk jugs to protect them from UV degradation. A company has a bright idea in 2006 that this chemical would also stabilize medical device plastics during sterilization (usually gamma radiation), and get a use patent on this stabilizer for use in the medical industry. 30+ year old technology, used in several fields, but requires licensing to use it in medical devices. And this is a world patent. I'm guessing the patent would be a use patent. The patent is on a specific use of the math (although it doesn't sound all that specific from the claims that I've seen).
sqldba 10y ago
Well, if it's an algorithm as the article alleges - compression is an algorithm and so are video codecs. They can be patented. Why should this be treated differently?
- galadran 10y ago
  Why should video codecs be patentable? I'm well aware they are currently but it seems ludicrous that you can patent the performance of a sequence of mathematical operations.
- Dylan16807 10y ago
  Video codec patents are an odd example to use, as they are currently doing a great job of stifling the state of the art, rather than promoting it.
  - sqldba 10y ago
    I'm not up to date on the complexities of patent vs copyright, but I'll assume you're against both? It just seems to me if you can write software you should be able to own it. If you can own it then you get to say where and how it is used and what compensation you want. Isn't an algorithm or a codec a lot like saying a program? Why shouldn't people be able to own the fruit of their labors?
    - Dylan16807 10y ago
      People don't get to own the math they invent, or the mental processes they invent. I think the same should apply to algorithms in general. You don't get to control people's thoughts, and if someone is allowed to think something, they should be allowed to have a computer assist them in thinking it. Copyright is fine, if we ignore the recent length extensions. If you write software, you get to say who uses that software. But you don't get to say who uses similar software.
      - seanp2k2 10y ago
        Consider also how this applies to AI. It's very possible that in our lifetimes (I'm assuming you'll live for the next 40 years) we'll see strong AI, possibly capable of inventing novel things by itself or with human collaboration. Who owns the patents which will inevitably result from this? What if the AI has an opinion on the matter? Should the AI own the patents? Is someone allowed to own the AI?
mrmondo 10y ago
Couldn't agree more, for f@&$! sake people, we're trying to evolve, form cross-community, intercontinental collaboration and move forward together as a species but a messed up financial model (particularly in the USA) is holding society back. We almost all know it, it's easy to say but hard to change. I think that the key is to keep the conversation current, transparent and relevant to the general public and that is the only way blockers like patents will ever go away.
- stevenwiles 10y ago
  Are you saying that if I come up with a good idea, I don't have a right to protect that idea?
  - angrow 10y ago
    The purpose of copyright and patent laws is to give the greatest benefit to society (in the American constitution, the phrasing is "to promote the useful arts and sciences"). The granting of temporary monopolies to registrants is a common tactic used to achieve that end, but not the only possible one, and one that admits of many variations (terms, extensions, and limitations vary across jurisdictions and over time). Maybe some people think that the exact laws we have right now are the best possible at serving the underlying purpose of the long-term enrichment of the public domain, but many seem to think that improvements could be made.
  - ewzimm 10y ago
    Let's say you come up with a great idea, a universal cure for cancer, but it requires growing human stem cells in a lab and you're against that for religious reasons, so you patent it and make sure nobody ever uses it. There are a few issues here. First, you definitely didn't come up with that idea from nothing. You "stood on the shoulders of giants" and profited from thousands of years of scientific development. You almost certainly collaborated with lots of other people who might not have shared your personal values and contributed just as much or more than you did to its development. You happened to be the person who connected the final piece. Just how much control should that give you over the use of the idea? Some people would say it's hard to justify your absolute control over the use of that idea.
chriswwweb 10y ago
I totally aggree, patents on an algorithm are really not a good idea Yeah the patent is still registered in some european countries but not all of them, in some countries it has expiered, don't know what this means from a juridical point of view though Here is the full patent description: http://www.google.com/patents/EP1177529A1?cl=un http://www.google.com/patents/EP1177529A1?cl=un

MichaelBurge 10y ago

It looks like this is the patent, and that it was withdrawn because they stopped paying the fees: https://register.epo.org/application?number=EP05100648&lng=en&tab=main https://register.epo.org/application?number=EP05100648&lng=e... Assuming this is the same patent, either Gielis is so irresponsible and disorganized he didn't realize his patent expired, or he's just trying to rustle up some bad press to extort money out of them.

FreeFull 10y ago
Following the link from the article gives this: https://register.epo.org/application?number=EP00929732 https://register.epo.org/application?number=EP00929732

SlashmanX 10y ago

> creator Sean Murray acknowledged in a 2015 New Yorker interview that he had struggled with elements of procedural planetary generation, until he discovered an equation published in 2003 by Belgian plant geneticist Johan Gielis that he called “Superformula.” I sense a settlement (or partnership) in the near future

runevault 10y ago
So the fact he didn't even reinvent it but used the paper makes this an entirely different case to me. Hm.

speps 10y ago

The paper with the "Superformula" : http://www.amjbot.org/content/90/3/333.full http://www.amjbot.org/content/90/3/333.full

asherkin 10y ago

> Company claims No Man's Sky uses its patented equation without permission is pretty damn different from > No Man's Sky sued over procedural generation algorithm patent

anotheryou 10y ago

Why as this formular so cool? Cooking it the shapes on the wikipedia shapes, once unraveled from the circular shape the curves don't look too interesting to me. But these might be simple examples...

conmarap 10y ago

The superforumula is public knowledge and these guys are clowns. Dumb patent trolls... https://en.wikipedia.org/wiki/Superformula https://en.wikipedia.org/wiki/Superformula

fsloth 10y ago
Your linked wikipedia page states that it's patented. Yes, it's silly one can do that but it gives some legal credibility to the case.

chriswwweb 10y ago

UPDATE: Hello Games has finally posted a response to the patent infringement claims, this is what they tweeted: "No Man's Sky doesn't actually use this "superformula" thing or infringe a patent. This is a non-story... everybody chill (シ_ _)シ" Tweet is here: https://twitter.com/NoMansSky/status/756889227095318528 https://twitter.com/NoMansSky/status/756889227095318528