For someone with a science background, such as a chemist or chemical engineer, how difficult would it be to produce a patented product without the original inventor alive and only the patent document to guide off of? This is assuming the patent rights were available of course. Patents get licensed out frequently to other companies so I assume it gets done. I am just not in a scientific field so it’s hard for me to visualize how tough that would be.
It’s supposed to be easy enough to be practical. The patenting inventor has to provide teachings that can be followed by one skilled in the relevant art. That’s his part of the deal, in return for which he gets a monopoly for 20 years.
Thread title edited to indicate subject.
Thanks, sorry about that.
To the previous poster, yes you are correct that the patent must explain alot and even has illustrations. Was just curious though if it would be a case of “much easier said than done.” 
Patents can be pretty confusing to read though. I think often the reviewer lets a lot of vague things get by. Even with a detailed procedure there is trial and error. It is simply impossible to write down every pitfall and every detail of every procedure, and often the publishor or writer has no interest in making it easy. Sometimes things only work due to impurities or other anomalies that weren’t really predicted in the first place.
It may not be as true of patents, but with journal articles I’ve found experimentals that were just plain wrong. It’s very common for papers to publish only the very best yield they ever had, and the 19 other experiments that gave 2% yield are not reported.
So, it’s not a walk in the park. How long it takes is highly variable.
I think many patents are written to gain as much protection as possible while providing the competition with as little useful information as possible. I never found any very useful information in a patent when I worked as a chemist years ago.
Yea. Getting useful information from patents is a lot like staring at those magic 3D images. It looks like a lot of useless nonsense until you really stare at it until your tired. Those things are actually written by lawyers, so to us it looks like a bank statement. Over the past year, I have had to actually get information out of one or two, and the information is usually there. It’s just buried.
The drawings are all next to useless. Patents are limited to black and white images unless there are extraordinary reasons for color. Most of the images I’ve seen looked like they were faxed twice before being crumpled up and thrown in the trash and then scanned to be inserted in the word document at a 7 degree angle off of whatever layout would be unnatural to look at.
Bolding above is mine. I am a chemical engineer with 10+ patents. Most patents I have are about processes or methods or a system. Products (assuming you mean a gadget or something equivalent) should be fairy easy to reproduce. Ping me - if you have a particular patent in mind and have the license or rights.
It could be quite hard to reproduce a specific product, because patents cover inventions, not specific products. Commercializing an invention often requires significant additional investment, research, and development, and the results obtained during this commercialization effort do not need to be published in order to maintain protection of the ultimate commercial embodiment under the original patent.
It is possible to invent something, and obtain a patent covering hundreds of possible implementations / variations / embodiments of your invention. You may then proceed to spend years researching, optimizing and testing to see which specific implementation of your invention is the most commercially viable, has the best performance, is cheapest, or whatever, and then sell that without disclosing the details to anybody.
Regardless, the original patent is still enforceable against anyone using or selling a product covered by the original broad claims, even if their product is significantly inferior to your optimized product.
Legally, it should be possible for a person “skilled in the art” to read the patent, and reproduce the claimed invention with a “reasonable” amount of time, experimentation, etc. However, there is no guarantee that his reproduction of the claimed invention will perform as well as the commercial product.
For instance, let’s say I get a patent on a new class of lubricant, totally different from what has been done before. I specify a wide variety of processes, ingredients, etc. that can be used in order to produce lubricants of this class. Some of these processes, etc. are costlier than others, and some of the specific lubricant variations produced lubricate better than others, but they are all novel and different from any other lubricant people have produced before. I can get a very broad patent that covers all these possible variations.
I can now retreat to my laboratory and spend several years optimizing my processes, ingredient selection, etc. in order to produce a specific lubricant of this new class that costs the least and lubricates the most, and then sell it commercially. I do not have to disclose the specific process/composition/etc. of this product to anyone, and my original broad patent will remain in effect.
If you wanted to replicate my commercial product using only the information in the patent, it is likely that you would have to spend the same several years I did in determining the optimal version of the invention. You should be able to relatively easily produce something that falls into the class covered by my original patent, but there is absolutely no requirement that the patent enable you to easily replicate my ultimate commercial embodiment.
EDIT: On the other hand, some patents are not very broad. A patent covering an incremental improvement in a well-established field will generally be much narrower and more specific than a patent covering a revolutionary improvement / massive step ahead of the rest of the field. A narrow patent like this may be much easier to replicate, simply because the range of possible variations is limited. Of course, people always try to seek the broadest patents possible.
I will point out another wrinkle that may work in the favor of someone seeking to replicate a product covered by an existing patent.
Let’s say I have invented a whole new class of lubricants, which we’ll call type A lubricants, and then as discussed, spend years perfecting an optimized commercial version, and choose to keep the details of this version secret.
If the optimized commercial version did not involve any additional “inventions” (as defined by the patent office, a separate topic unto itself), just a whole lot of engineering grunt work, I am in the clear. However, if any of my improvements to the general “type A lubricant” from my original patent are in fact patentable inventions themselves, and I choose to keep them secret as opposed to filing new patents, I am actually vulnerable to a third party independently inventing the same improvements and filing a patent themselves. In that case, even though they would not be able to sell their product (since it would infringe on my original, broad patent), they could come after me for patent infringement, since my optimized version would be infringing on their new, likely more narrow patent.
As such, the patent system does motivate companies to publish publicly / patent as much as they can, and not to maintain trade secrets unless they are absolutely certain these secrets cannot be derived by third parties.
Doing this sort of activity totally defeats the spirit of patents. Patents are there to disperse knowledge / advance products and the author gets protection in return for a certain amount of time. Doing what you are suggesting just defeats the purpose - and it maybe so broad that you may not be able to get the patent and b - even if you get the patent and someone copies it - it maybe hard for you to prove patent infringement. It would be far better for you to hold the composition of your lubricant as a business secret - just like Coke does since. It has the added benefit of never expiring.
See my second post. It is perfectly valid, legal, and within the spirit of the patent law to invent something new, patent it, and then choose to keep subsequent derivative improvements as trade secrets (whether these represent novel, non-obvious “inventions” or not).
As the lawyers are telling me, the minute you start selling it, that constitutes public disclosure. That would mean that no one, not you or a competitor could patent those improvements. Also, assuming you kept a good notebook, your improvements would be documented as prior art.
Patenting a whole product is very rare. What is patented now is some aspect to a product that is novel. I have a patent about a way of efficiently implementing a type of digital filter. This is not a product. It can be used in a phone to make some parts of the signal processing cheaper and use less power but it doesn’t tell you how to design a phone.
I don’t think you are. You are supposed to teach your preferred embodiments at the time of filing. To be in the clear you must complete your filing before you start the optimization you discuss.
Optimization you do after filing, if you never teach it, is not a contribution to the patent system, but it isn’t a cheat either. But holding out on preferred embodiments you know about when you file is cheating, I think.
One issue is that patent lawyers use a peculiar dialect in order to make the patent as broad as possible. Example: A transistor is used in a patented circuit. The patent won’t refer to it a transistor. It will be a “electrically controlled switching device” or some such so that the patent still applies to the circuit built with vacuum tubes or relays. When you extend this to dozens of parts within the patent, it can become really difficult to follow.
That said, I have used patents to learn how a number of interesting widgets work. The signal to noise ratio for patents in general is pretty low, but if you look at the patents for products actually being sold it becomes much better. Also a lot of older patents are far easier to read…the patent lawyer dialect has evolved mostly in the last 50 years or so. A lot of the older patents are only a page or two in nearly plain english.
See, the ‘Best Mode’ requirement of 35 U.S.C. 112.
Recent examples of patent claims being held invalid for failure to disclose the best mode include cases like Ajinomoto v. ITC. Brief summary from the winning side’s attorneys here. An interesting discussion on best mode disclosure in pharmaceutical research can be found in Bayer v. Schein, 301 F.3d 1306, 1314-23.. It’s lengthy, but the CAFC really goes into what makes the “best mode” in the case, and compares it to several other best mode cases in different art areas.
Well, some points.
First, as the inventor, you have a 1-year grace period before your own published work / public disclosure / offer for sale can be counted against you. And your own research notes or records will never be counted as prior art against you unless you publish them.
Second, the sale of a product produced through an industrial process does not necesarily represent a public disclosure of the process itself.
I did not mean to imply otherwise in my post - in order to avoid violating the best mode requirement, my hypothetical inventor would historically have had to file the first application before beginning his optimization / commercialization work.
However, Congress nullified the best mode requirement with the America Invents Act in 2011, which has been signed into law. Failure to disclose the best mode is no longer a basis for invalidating a patent.
From the lab chemists’ point of view, there is huge variation in the reproducibility of procedures. I’ve worked on many syntheses from all the big name pharmaceutical companies and I would say 50% work well, 30% don’t work but can be made to if you can work out what crucial details has been left out/changed and 20% are just fiction and I suspect were never really made (but that’s just my opinion). No idea why, but I never found a single good Pfizer prep, but old ICI preps were bulletproof, GSK middling, Merck good…
A process chemists fom AstraZeneca told me they were always told to supply their worst procedures for making a compound, rather than any optimised one, which makes sense I guess…