So just because a company has a patent on "X", say some very specific mRNA stuff, it has to disclose all its other non-patentable production secrets? That seems unreasonable.
A patent isn't "this stuff exists and it's ours now", it's protection of the process of creating it. What other production secrets are there? If a patent can't be replicated using only public information, it shouldn't be valid
That's if it's a process patent. There's also molecule patents, etc. I could be wrong but typically pharmas don't take out process patents because they are hard to prove and you need a warrant which has a nearly unattainable level of discovery to get on premises and determine if the process has been copied. With a molecule patent you can buy a sample and prove infringement with an analytical technique.
While the fda has disclosure requirements on a pharma manufacturing process, it does not have to be disclosed to the public.
Yes, and the idea that those could be stolen is the process working as intended. The fact that they could be stolen and little recourse would be possible is the incentive for companies to apply for patents, which are supposed to be better for the public good that trade secrets.
you're not understanding. If I patent a process, I am required to publish the process. Then, a competitor can copy the process (because it is published). How do I as, the process patentor defend against infringement?
Because it is a process, I have to go on their premises and prove that they are stealing my process. You must obtain a warrant to do that (because otherwise it is trespassing), and because competitors could just willy nilly claim "you're stealing my process" in a spurious fashion to gain access to the premises (and learn about what competitors are doing), judges typically put up a very high burden of proof to issuing those warrants.
You'll effectively never be able to enforce a process patent, so you're better off never patenting it and keeping it a trade secret instead, that way you have a bit of a moat.
By contrast, for a molecule or plant patent, the existence of the product is proof of infringement, so it's easier to enforce infringement.
For a molecule patent, you do not need to disclose the method you use to make it. IANA(P)L, You might be required to disclose one way of making the molecule, but it almost certainly doesn't have to be the way that you do it "in prod".
For big companies, you need one whistleblower. You hire one employee over. They say "Oh gosh. We were doing the same thing." It's hard to keep that sort of thing secret.
Warrants are granted to law enforcement, not in civil suits between corporations. The word is subpoena, and if you knew even the basics of what you're talking about, you'd use correct terminology.
Legal discovery usually doesn't have a "very high burden of proof," because there are very reasonable ways to compartmentalize information. Legal discovery usually leads to a ton of transparency, at least to the legal teams on both sides (and a firewall, so that can't go over to the engineers).
MOST executives are diligent about responding to subpoenas. Judges have broad discretion for contempt-of-court, up to and including prison. If you're the CEO of BioCorp, and you can:
1) Respond promptly, completely, and accurately. Risk: You might wipe $200 million off of your books and cost each investor $2
OR
2) Stonewall the court or lie. Risk: You might get thrown in prison.
Which do you do? Most executives pick responding promptly. Not all, but most.
Most companies are super-careful not to infringe patents that they know about, because it's not fun. Unless it's a case of mutually-assured destruction. There are whole industries where everyone infringes, and no one sues anyone.
If you are talking purely about a process patent, then yes, you will have such an issue. But process patents aren't the only way you can protect your production process, you can also patent various necessary equipment.
Ultimately, if you rely on keeping a process secret, you will most likely fail. Processes are trivial to copy, patent or not, and if you don't file a patent then it's not even illegal to copy your process when, not if, it becomes known.
Anything can be called 'business related information'
EU patent office says a patent must provide enough information to be reproduceable by a competent proffeshional - that's the whole point of a patent, to be usefull after it expires. It is not there to list things the company would lile to protect.
US patents must provide enough information for one of ordinary skill in the art to practice the inventions recited in the Claims of the patent.
Thus, depending on what the claims cover, plenty of detail about complete systems can be omitted from the patent specification. For example, if the claims of a patent are directed to cache management for a messaging platform, most details about how the messaging platform works can be omitted as long as the cache management system is sufficiently described.
If there are unique features of the messaging platform that enable the new cache management system to work, detailed description of those features would need to be included in the patent specification. But unrelated stuff, such as, routing mechanisms, protocol optimizations, security, and so on, that may be part of the messaging platform can be omitted from the cache management patent specification.
Thus, the messaging platform org could keep plenty of trade secrets about other parts of their platform while still obtaining a patent on their cache management system.
