14 ms·

I still don't understand how this company managed to get a patent for what amounts to using coordinates to look up words in a table. It's also not the first to

by anfilt 5y ago

I still don't understand how this company managed to get a patent for what amounts to using coordinates to look up words in a table. It's also not the first todo this, there is prior art. For example this person was asking about what they would need to do to invalidate it because they built basically the same thing, but using 5 words before this company even had patents. https://patents.stackexchange.com/questions/13629/i-had-invented-and-published-before-this-patent-application-how-do-i-get-it-in https://patents.stackexchange.com/questions/13629/i-had-inve... Also on the OpenStreet Wiki, the EU seems to have doubts on the patents Validity. https://wiki.openstreetmap.org/wiki/What3words#Patent_Validity https://wiki.openstreetmap.org/wiki/What3words#Patent_Validi...

dleslie 5y ago
The American patent office assumes that if there is no prior patent then the new patent describes something new and novel. It's designed this way to encourage Americans to patent everything and anything, in order to dominate the intellectual property market.
- brobdingnagians 5y ago
  It's also done by political policy to make it look like America is still "very innovative". It would make politicians (and other people) look bad if the number of patents started falling precipitously. I believe America is still innovative, but not as much as the number of patents might suggest.
- kindall 5y ago
  The purpose of the patent system is to incentivize people to publish the details of their inventions instead of keeping them secret, thereby advancing the state of the art. The limited-time monopoly is the incentive.
  - yossarian1408 5y ago
    Except the limited-time monopoly becomes an until-the-end-of-time monopoly, and has a more toxic effect on innovation and creativity than what would exist in its absence.
    - throwaway3699 5y ago
      Patents expire all the time...
    - midasuni 5y ago
      You’re thinking of copyright which seems infinite. Patents aren’t
      - yossarian1408 5y ago
        you're right, my bad
- unityByFreedom 5y ago
  Woah, back up. The patent system is messed up for software but that wasn't the intent. You weren't supposed to be able to defend algorithm-based patents in court. Software is protected by copyright. The patent office doesn't do much background checking. Enforcement happens in the court system. And, some courts such as those in Texas have handed cases to patent trolls. Patents, by design, reward innovators and contribute to the public good by exchanging information for protection. When you see a big company owning a lot of patents, that's the system at work. IP rights allow the company to defend their interests and pay larger salaries. Without IP rights, individual creators really have little chance to move up because wealthy individuals or businesses will just copy and mass-produce the concept. In order to remain involved you would have to play by their rules, like what happened to Jack Ma. Everyone answers to the government there because there's no delegation of power to courts to sort out IP rights.
  - jfrunyon 5y ago
    > Patents, by design, reward innovators and contribute to the public good by exchanging information for protection. Design and practice are two different things. Patents, in practice, are hopelessly broken when you can buy a fairly general patent from someone else, sit on it without ever actually using it, and then sue someone else who had no way of knowing the patent even existed. And that's not just in software.
    - unityByFreedom 5y ago
      AFAIK it's primarily abused in software because software is so easy to "create". Again that's why algorithmic patents weren't supposed to be defendable in court. Outside of software patents which I don't think should exist, I don't see a problem with buying and sitting on a patent for something truly innovative. The original creator was rewarded, the knowledge becomes public, and the buyer can sit on it as they like for ~15 years. If it's a profitable concept they are wasting money by not developing it. The problem comes when courts uphold illegitimate patents. Fortunately there are companies who will stand up to this, such as Newegg and Cloudflare.
      - 5y ago
- gnicholas 5y ago
  > The American patent office assumes that if there is no prior patent then the new patent describes something new and novel. They don't just look for prior patents, they look for prior art (which includes any public research papers, presentations, products, etc.) as well. So if someone else did the thing you claim to have invented (even if they didn't patent it), you can't get a patent. And even if someone didn't do the thing you're claiming, but they did something that makes your claimed invention an obvious thing to do, you also can't get a patent. The obviousness standard is generally harder to overcome than the novelty standard.
  - Tor3 5y ago
    That's the theory. In practice it works as was described above - they look for existing patents. Looking for prior work to any degree would presumably make the cost (and time) astronomical (or at least not covering what they're paid), as a general rule. The world is very different from when the patent system was established.
    - gnicholas 5y ago
      It's possible that in some fields, the PTO focuses on patents because so many of the innovations are patented. But this doesn't fit my experience, having worked for well over a decade in Silicon Valley (first as a lawyer, and then as a founder). I could be wrong — this is just the sense I've gotten from talking with patent prosecutors at my old firm, and from talking with my outside counsel about our prosecution strategy.
killingtime74 5y ago
Patents can be challenged. You would need to file a lawsuit though if the filing challenge dates are past. Prospective patents are announced in a government gazette
- jillesvangurp 5y ago
  They can but it is more expensive than filing them and you generally need deep pockets to do so. The issue with US patents specifically is that the whole system is rigged to big corporations flooding the system with patents that are not challenged because the patent offices are underfunded. This allows these big corporations to create a defensive buffer that they use against each other and much smaller competitors. Challenging that requires deep pockets. Companies like Apple, routinely settle patent licensing conflicts outside the courts. E.g. Nokia still receives a fair bit of revenue from them for patents the iphone and related products obviously infringe upon. Likewise Qualcomm taps into mobile phone sales revenue using their heavily patented proprietary solution for building 4G and 5G modems. But like patents or not, they are real and they are a factor to take into account when doing business. In this case, what3words seems to have genuinely come up with a way to translate coordinates into pronounceable words. You might not agree their method is particularly interesting, sophisticated, valuable, etc. But a similar enough method did not really exist prior to them. And they patented that method. And they got the patent. What that means legally is that the claims in the patent are legally enforceable against others doing a me too service. Unless they challenge all of those claims of course. The issue with prior art is that it usually only affects some of the claims. So, you might feasibly manage to weaken the patent somewhat in court at great cost but probably not in it's entirety. So, e.g. Apple or Google suddenly promoting a service that allows you to refer a map location on their using a bunch of words would be a problem as it would obviously infringe on one or more claims in the patent. Of course they have deep enough pockets that they could try to challenge the patent. But they might lose. It's more likely they'd attempt an acquisition and add the patent to their portfolio. That in a nut shell is why what3words might be considered a good investment by some: they are an obvious acquisition target with a customer base, some revenue, a brand name, etc. To me 80M raised in funding sounds excessive though. But it explains why they are eager to defend their IP: without that their valuation tanks. This lawsuit interestingly is not about the patent but about copyright. I assume the list of words used by their patented algorithm is the issue here. That list of words is not part of the patent. That copyright seems defensible as well. It's a list they came up with, that they own the copyright of, that is essential to them and somebody else is publishing that list without their permission. Copyright exists to allow people to put a stop to that sort of thing. Hard to argue against that.
oytis 5y ago
The threat doesn't mention the patent though, they claim copyright for the code and data has been violated.
- slaymaker1907 5y ago
  Yeah, that part makes this whole thing smell really fishy. The researcher claims it was done through reverse engineering which would not be a violation of copyright. They'll have to show that someone had the opportunity to see their code and then copied it. Patents would arguably be an easier defense since you only need to show similarity and independence of invention doesn't matter.
  - oytis 5y ago
    As I understand they designed it to give the same words as what3words does for each location. That's what w3w may consider violation of copyright for data by itself. A patent breach would be creating alternative implementation with a different dictionary or a different mapping
    - wccrawford 5y ago
      OTOH, the words aren't artistically chosen. They are chosen by a computer for each number. You can think of the words as numbers in another form. And interfacing with the same system would require the same words. I understand the argument that a curated list of words is copyrightable, but in this instance I don't think we're just talking about a list of words. We're talking about an algorithm that turns numbers into certain words. I could see it going either way, but I'd like to think it would fail to be copyrightable in the court.
      - oytis 5y ago
        Well, IDK it's not like there was an open standard for assigning words to locations, and w3w made an implementation for it. Both the mapping and the software for it was created by w3w and are proprietary to them. Interfacing with it would mean copying entire product verbatim, which I don't think is defendable. There are publicly available location methods like national addressing systems or coordinates, or one can create one's own and make it public. But honestly, while all my sympathies are with open source movement, and I'm generally critical to copyright, I don't understand why everyone seems to think you can treat things other people created as a commodity and require free access to it.
nevi-me 5y ago
Around 11 years ago, there was a South African startup called Waytag [0] [1]. It was started by some business consultant guys, who I assumed didn't understand the lack of novelty in their idea (from a patent perspective). Their idea was similar, minus the algorithm. You could assign names to places, and people would use their proprietary system to find them. They filed for patents in various countries, US included. Here's their European patent: [2], status withdrawn. The US equivalent is patent US20120246195A1. I can't find its history, nor a post on Patents Stackexchange, to detail the history. Someone applied to have it invalidated, on the premise that there was existing prior art. My reading of the Patents question, is that this person filed an opposition in the UK. Has anyone ever tried doing same in the US? Or has some legal time lapsed? [0] https://ventureburn.com/2010/06/waytag-changing-the-way-we-use-location-services/ https://ventureburn.com/2010/06/waytag-changing-the-way-we-u... [1] https://news.ycombinator.com/item?id=10961677 https://news.ycombinator.com/item?id=10961677 [2] https://patents.google.com/patent/EP2488969A1/en https://patents.google.com/patent/EP2488969A1/en
ealexhudson 5y ago
OpenStreetMap seem to have missed a trick. The GB patent was thrown out for being obvious, and the report on the exam is public. They also suggested withdrawing the application due to the EU one being filed.
suifbwish 5y ago
Couldn’t you get around this by just making it a trade secret and not publishing it? Your competitor can’t ask to see your source code (assuming it’s compiled). Patent trolls deserve whatever they have coming. It’s one thing if they have a novel idea but if they knowingly sit on a patent of a fundamental method that was given to them in error, they don’t deserve any consideration.

by anfilt 5y ago

dleslie 5y ago

The American patent office assumes that if there is no prior patent then the new patent describes something new and novel. It's designed this way to encourage Americans to patent everything and anything, in order to dominate the intellectual property market.

brobdingnagians 5y ago
It's also done by political policy to make it look like America is still "very innovative". It would make politicians (and other people) look bad if the number of patents started falling precipitously. I believe America is still innovative, but not as much as the number of patents might suggest.
kindall 5y ago
The purpose of the patent system is to incentivize people to publish the details of their inventions instead of keeping them secret, thereby advancing the state of the art. The limited-time monopoly is the incentive.
- yossarian1408 5y ago
  Except the limited-time monopoly becomes an until-the-end-of-time monopoly, and has a more toxic effect on innovation and creativity than what would exist in its absence.
  - throwaway3699 5y ago
    Patents expire all the time...
  - midasuni 5y ago
    You’re thinking of copyright which seems infinite. Patents aren’t
    - yossarian1408 5y ago
      you're right, my bad
unityByFreedom 5y ago
Woah, back up. The patent system is messed up for software but that wasn't the intent. You weren't supposed to be able to defend algorithm-based patents in court. Software is protected by copyright. The patent office doesn't do much background checking. Enforcement happens in the court system. And, some courts such as those in Texas have handed cases to patent trolls. Patents, by design, reward innovators and contribute to the public good by exchanging information for protection. When you see a big company owning a lot of patents, that's the system at work. IP rights allow the company to defend their interests and pay larger salaries. Without IP rights, individual creators really have little chance to move up because wealthy individuals or businesses will just copy and mass-produce the concept. In order to remain involved you would have to play by their rules, like what happened to Jack Ma. Everyone answers to the government there because there's no delegation of power to courts to sort out IP rights.
- jfrunyon 5y ago
  > Patents, by design, reward innovators and contribute to the public good by exchanging information for protection. Design and practice are two different things. Patents, in practice, are hopelessly broken when you can buy a fairly general patent from someone else, sit on it without ever actually using it, and then sue someone else who had no way of knowing the patent even existed. And that's not just in software.
  - unityByFreedom 5y ago
    AFAIK it's primarily abused in software because software is so easy to "create". Again that's why algorithmic patents weren't supposed to be defendable in court. Outside of software patents which I don't think should exist, I don't see a problem with buying and sitting on a patent for something truly innovative. The original creator was rewarded, the knowledge becomes public, and the buyer can sit on it as they like for ~15 years. If it's a profitable concept they are wasting money by not developing it. The problem comes when courts uphold illegitimate patents. Fortunately there are companies who will stand up to this, such as Newegg and Cloudflare.
    - 5y ago
gnicholas 5y ago
> The American patent office assumes that if there is no prior patent then the new patent describes something new and novel. They don't just look for prior patents, they look for prior art (which includes any public research papers, presentations, products, etc.) as well. So if someone else did the thing you claim to have invented (even if they didn't patent it), you can't get a patent. And even if someone didn't do the thing you're claiming, but they did something that makes your claimed invention an obvious thing to do, you also can't get a patent. The obviousness standard is generally harder to overcome than the novelty standard.
- Tor3 5y ago
  That's the theory. In practice it works as was described above - they look for existing patents. Looking for prior work to any degree would presumably make the cost (and time) astronomical (or at least not covering what they're paid), as a general rule. The world is very different from when the patent system was established.
  - gnicholas 5y ago
    It's possible that in some fields, the PTO focuses on patents because so many of the innovations are patented. But this doesn't fit my experience, having worked for well over a decade in Silicon Valley (first as a lawyer, and then as a founder). I could be wrong — this is just the sense I've gotten from talking with patent prosecutors at my old firm, and from talking with my outside counsel about our prosecution strategy.

killingtime74 5y ago

Patents can be challenged. You would need to file a lawsuit though if the filing challenge dates are past. Prospective patents are announced in a government gazette

jillesvangurp 5y ago
They can but it is more expensive than filing them and you generally need deep pockets to do so. The issue with US patents specifically is that the whole system is rigged to big corporations flooding the system with patents that are not challenged because the patent offices are underfunded. This allows these big corporations to create a defensive buffer that they use against each other and much smaller competitors. Challenging that requires deep pockets. Companies like Apple, routinely settle patent licensing conflicts outside the courts. E.g. Nokia still receives a fair bit of revenue from them for patents the iphone and related products obviously infringe upon. Likewise Qualcomm taps into mobile phone sales revenue using their heavily patented proprietary solution for building 4G and 5G modems. But like patents or not, they are real and they are a factor to take into account when doing business. In this case, what3words seems to have genuinely come up with a way to translate coordinates into pronounceable words. You might not agree their method is particularly interesting, sophisticated, valuable, etc. But a similar enough method did not really exist prior to them. And they patented that method. And they got the patent. What that means legally is that the claims in the patent are legally enforceable against others doing a me too service. Unless they challenge all of those claims of course. The issue with prior art is that it usually only affects some of the claims. So, you might feasibly manage to weaken the patent somewhat in court at great cost but probably not in it's entirety. So, e.g. Apple or Google suddenly promoting a service that allows you to refer a map location on their using a bunch of words would be a problem as it would obviously infringe on one or more claims in the patent. Of course they have deep enough pockets that they could try to challenge the patent. But they might lose. It's more likely they'd attempt an acquisition and add the patent to their portfolio. That in a nut shell is why what3words might be considered a good investment by some: they are an obvious acquisition target with a customer base, some revenue, a brand name, etc. To me 80M raised in funding sounds excessive though. But it explains why they are eager to defend their IP: without that their valuation tanks. This lawsuit interestingly is not about the patent but about copyright. I assume the list of words used by their patented algorithm is the issue here. That list of words is not part of the patent. That copyright seems defensible as well. It's a list they came up with, that they own the copyright of, that is essential to them and somebody else is publishing that list without their permission. Copyright exists to allow people to put a stop to that sort of thing. Hard to argue against that.

oytis 5y ago

The threat doesn't mention the patent though, they claim copyright for the code and data has been violated.

slaymaker1907 5y ago
Yeah, that part makes this whole thing smell really fishy. The researcher claims it was done through reverse engineering which would not be a violation of copyright. They'll have to show that someone had the opportunity to see their code and then copied it. Patents would arguably be an easier defense since you only need to show similarity and independence of invention doesn't matter.
- oytis 5y ago
  As I understand they designed it to give the same words as what3words does for each location. That's what w3w may consider violation of copyright for data by itself. A patent breach would be creating alternative implementation with a different dictionary or a different mapping
  - wccrawford 5y ago
    OTOH, the words aren't artistically chosen. They are chosen by a computer for each number. You can think of the words as numbers in another form. And interfacing with the same system would require the same words. I understand the argument that a curated list of words is copyrightable, but in this instance I don't think we're just talking about a list of words. We're talking about an algorithm that turns numbers into certain words. I could see it going either way, but I'd like to think it would fail to be copyrightable in the court.
    - oytis 5y ago
      Well, IDK it's not like there was an open standard for assigning words to locations, and w3w made an implementation for it. Both the mapping and the software for it was created by w3w and are proprietary to them. Interfacing with it would mean copying entire product verbatim, which I don't think is defendable. There are publicly available location methods like national addressing systems or coordinates, or one can create one's own and make it public. But honestly, while all my sympathies are with open source movement, and I'm generally critical to copyright, I don't understand why everyone seems to think you can treat things other people created as a commodity and require free access to it.

nevi-me 5y ago

Around 11 years ago, there was a South African startup called Waytag [0] [1]. It was started by some business consultant guys, who I assumed didn't understand the lack of novelty in their idea (from a patent perspective). Their idea was similar, minus the algorithm. You could assign names to places, and people would use their proprietary system to find them. They filed for patents in various countries, US included. Here's their European patent: [2], status withdrawn. The US equivalent is patent US20120246195A1. I can't find its history, nor a post on Patents Stackexchange, to detail the history. Someone applied to have it invalidated, on the premise that there was existing prior art. My reading of the Patents question, is that this person filed an opposition in the UK. Has anyone ever tried doing same in the US? Or has some legal time lapsed? [0] https://ventureburn.com/2010/06/waytag-changing-the-way-we-use-location-services/ https://ventureburn.com/2010/06/waytag-changing-the-way-we-u... [1] https://news.ycombinator.com/item?id=10961677 https://news.ycombinator.com/item?id=10961677 [2] https://patents.google.com/patent/EP2488969A1/en https://patents.google.com/patent/EP2488969A1/en

ealexhudson 5y ago

OpenStreetMap seem to have missed a trick. The GB patent was thrown out for being obvious, and the report on the exam is public. They also suggested withdrawing the application due to the EU one being filed.

suifbwish 5y ago

Couldn’t you get around this by just making it a trade secret and not publishing it? Your competitor can’t ask to see your source code (assuming it’s compiled). Patent trolls deserve whatever they have coming. It’s one thing if they have a novel idea but if they knowingly sit on a patent of a fundamental method that was given to them in error, they don’t deserve any consideration.