Tweet
Considering the cost of patents, what are the thoughts of embedding the smarts of the system into the chipset and making the code and embedding near impossible to back engineer or to go for a patent, tell the world and hope the copy cats stay away? There is plenty of discoveries that can be patented or go the secretive way?.considering the Chinese got their hands on a detector companies design files a few years ago it seems that some entities are willing to go to all put on the route on intellectual theft. What are your thoughts on this?
-
-
You can get a patent for $5,000-$10,000 depending on how much of the work you want to do. But a patent only gives you the right to sue someone, and the cost of that starts at around $100,000. And if the perp is in China, there is no amount of money that will produce a win.
I generally oppose patents for individuals unless it is something truly worth a ton of money. You will never ever ever see a return on that investment, and you've just handed someone the key to your house. Expect it to be robbed. The exception to this is if you are wanting to sell an idea to a Big Detector Company. In that case, the US offers a provisional patent that is basically free and give you a year to negotiate. Australia may have something similar, I don't know. Also, patents are remarkably easy to bust because you are betting that you thought of something that no one else in the entire world has ever thought of. It's far more likely that prior art is lurking out there somewhere; you missed it, the patent examiner missed it, but the lawyer for the company you try to sue will find it. You can't win, you can't break even, and if you play the game you're sure to lose.
The alternative is to hide your IP the best you can. Keep in mind that there is nothing you can do to prevent someone from ripping your code out of the micro and disassembling it. Assume it will be done, because there are a dozen firms in China who will do it for $500 (I once used one). So even hiding your IP in code is not a safe bet.
However, Coca-Cola kept their formula a trade secret and outlasted patents by 120 years so far, so there is hope. -
As a patent holder myself, I can only back up Carl here. Either do it through your employer or forget about it. My impression is, that metal detector companies have a tendency to "over patent" things, probably to advertise with it and as a license to sue or threaten to counter-sue if they are threatened. For the individual, patents are all but useless. Just publish your invention for everybody to use and get the satisfaction of a company actually "stealing" your idea as a reward.Comment
-
Carl is right.Originally posted by searchbear View Post
I think instead of spending effort to hide the information;
put information on the market as soon as possible earn money.Comment
-
very few chips can not be readout, just depends on how much money you throw at it.... This was something I was very active in... but the the information got to be changing so fast the information became worthless, I have a friend who builds something like a detector, where he sells it they claim it can not work. he has been building this for about 20 years and still developing it now, used in gas / oil industry. he has not paintent on it but has loads on other stuff. not sure why.. might be even the Paintent gives away something ?Comment
-
Patents are a tool that the big companies can afford more so than the small producer, we have a patent for ground loop use for depth enhancement , its been an expensive exercise.Comment
-
Why did you decide to get a patent?Comment
-
Why? Looking back i would not do it again, but i feel it has hindered some companies bringing out various innovations. The reason was to have an edge over the opposition, but that is useless when you are not selling a product. A better way is to publish any discoveries so no one can patent and if they try then pop it up as prior art. I am currently sitting on a few discoveries that can be used to derive a patent but my
thoughts are leaning towards full public disclosure so everyone can use the information.Comment
-
Unless you already have a company making related products, I tend to agree. It's easy to think that one day you will get there but by then your discoveries may end up in someone else's patent, which blocks even you from using them.Originally posted by detectormods View PostComment
-
The safest method is to go and use companies expired patents, there is nothing that can be done unless you build a working unit on their printed circuit board artwork then they can have a copyright claim, The base designs are still very good, cannot change the laws of physics. Best to build a complete new input stage, go fully differential into 16 bit 2mbps SAR ADC, I found that switching the power supply off and back after the receive window causes as much issues as it solves, proper layout and shielding fixes particular issues, so some patents look good on the wall in a frame but not so important in the real world. That is just my opinion.Comment
-
Preventive publishing is the way to go. But if you have a golden egg, invest in production first, and while you are producing go with publishing - it becomes prior art.Comment
-
One of the gotchas about publishing prior art is that the "publication" must be reasonably accessible by anyone looking for it. I recall a case where someone wrote up a patentable idea and placed a copy of it in a random location in a huge library. Therefore making it technically "public," but not accessible enough according to a judge.
Suppose you have a cool new idea and it's all in DSP. So you build some detectors and sell them to the public, preferably on highly public Amazon or eBay where you can save the sales records. Along comes company X who patents the idea. I believe that your provable prior use of that idea protects you from infringement, but will that prior art invalidate the patent for others? I don't know the answer, just musing. If it might invalidate the patent, would the use of readout protection affect that decision? That is, you released the idea to the public in the form of a product but the idea was protected from reasonable access. Yes, I know that it's not hard to hack readout protection but in a court that may not matter.
I know that there are firms in China that will rip code from practically any micro for a pretty minimal fee. Once you have it, it's easy to run through a disassembler and end up with assembly code. That's where the real work begins, you now have to figure out what it all does. I also assume there are companies who will do this for a fee but I've never used one. As more & more ideas end up in firmware this is what would be needed to either prosecute patent infringement or defend a patent. For a company like Minelab it's nothing, for the individual it's probably suicide.Comment
Comment