I'm a math-nerd, not a lawyer, so I'm going to get everything here wrong, I'm sure.

USPTO, in supporting its purpose in encouraging innovation, says patents require "enablement" which allegedly is "teach a phosita to make and use the invention" outside of the claims. [1] Businesses do not EVER want to document their "magic smoke" so when they make patents they want it as obtuse as they can get away with, but allegedly they like to leave parts out so they aren't "instructing the competition".
Academia, as in academic publishing, is like the opposite of both of those in that the goal is to provide the "how to" very completely and disruptively, and without being rewarded a limited monopoly for the work.

Game theory says that I should do what the market does, tit-for-tat. There are multiple strategies and many players. If some folks make patents with holes, or unusably obtuse, then they aren't necessarily disclosing but they still get some form of limited monopoly. They also can make follow-on patents that cover the same actual material but disclose the other elements, and allow them to retain the monopoly longer than the 20 years allotted.

What is the correct approach, and can you back up or justify why it is correct?

Is there an indication in the 5% of patents that get licensed or 1% that get litigated on whether the hand-holding and complete teaching disclosure has an advantage in those areas? Is the relationship between enablement and claims a meaningful element when patents are tried? Do patent applications without the obtuse enablement get processed more quickly or are they more likely to be approved?

I personally prefer the more complete and simple enablement strategy. What I do want to do is to be able to put the assignee of a patent into the best possible position.

Thanks in advance, and I hope its okay if this question is a little philosophical. I originally asked in law but they voted to close it. Unlike the (kinder) stats admins (in cross-validated) who move it, the law-folks suggest I delete it there and then write it here. Also, they jump hard on the delete but I can only post once every 40 minutes here so that is a bit of a catch-22.

  • Are you trying to draft a patent without the use of a patent attorney? You do realize that there is a gate keeper called an examiner who decides whether or not the invention is enabled.
    – Eric S
    Commented Jun 8, 2023 at 0:06
  • 1
    The examiner does judge completeness and clarity of disclosure and written description under section 112 and in an egregious case might reject based on enablement. But the requirement that it be enabled to the point that a POSITA does not need to perform “undue experimentation” is more likely to be argued in court not with the examiner,
    – George White
    Commented Jun 9, 2023 at 2:28
  • @EricS - nope. A patent attorney brings many things to the table, one of which is "legacy". In computers its a euphemism for paperweight, in NASA it means worth its weight in gold (or better), but in patents it means if you use a credible firm that your patents get more credibility in examination, and are more likely to be successful if ever used in a court of law. I have some very tough things that I understand very well, so I am trying to stage for success patent folks who do NOT want to get into optimal control systems engineering or advanced machine learning. Commented Jun 9, 2023 at 13:59
  • 1
    With 99 issued patent worldwide, I am well aware of the dynamic between inventor and attorney. Especially those attorneys who are not well versed in the patent's field. I generally try to convey all I know about the invention to the patent attorney, but this does not mean I draft the patent application. That is the job of the attorney.
    – Eric S
    Commented Jun 9, 2023 at 16:14
  • 1
    Interesting question if a bit opinion based that I would be happy to try to answer.
    – Eric S
    Commented Jun 9, 2023 at 18:20

2 Answers 2


In the US it is not the enablement requirement that addresses the secret sauce issue, it is the “best mode” requirement. Inventors and applicants are required to show the best way of implementing the invention they know of, as of the date of filing. Without that requirement one could enable a version that did not work as well as the version with the secret sauce.

The 2012 AIA law changed things in a strange way; it kept best mode as requirement but removed it from the list of things that could get your patent torn up by a court.

In practice, your thoughts about someone deciding if they see a payoff in hiding the ball is not usually an issue because one wants to file early, which is before a product based on the invention is refined and optimized. To be ready for patenting does not require a working design that is practical, reliable or cost-effective. The famed “page rank” that people think is secret technology of google is patented and therefore not secret. But the parameters used in actual implementation have been tuned over time and can be held as a trade secret.

Any optimization after the filing date doesn’t affect the enablement or best mode requirements unless it turns out the invention, as filed, is impossible.

Two other points. You mention patenting part of a product and then filing on other parts of the product. That does happen for many reasons but note that once the product is sold or otherwise disclosed to the public there is a one year limit (in the US) to file on anything in it. In the rest of the world there is zero grace period so all filings need to be done before public disclosure.

Second, you mention your preference for a clear enabling disclosure to benefit your future licensee. If by that you mean to have a solid patent that is not vulnerable to attack for lack of enablement, great, I agree. However an obscure specification doesn’t prevent you for later giving all know-how you have to your licensee.

Patent documents can seem obscure due to the standard (for patents) language used, especially in claims.

  • Where can I read more about the 2012 AIA law? How does one apply for global patent; or are there at least regional aggregations? Does one need to apply for 220 patents to have global protection (assuming all are granted)? Commented Jun 8, 2023 at 17:42
  • 1
    @EngrStudent This is a question and answer site, not a forum. If you have a question you should post it as such. You may well get useful information by reviewing existing questions.
    – Eric S
    Commented Jun 9, 2023 at 0:27
  • @EricS - it is a learning site. Mileage and methods vary by groups. I can ask it as a new question though. Commented Jun 9, 2023 at 13:55
  • 1
    @EngrStudent Yes, this is a “learning site” whose format is questions and answers. Questions/Answers are far more searchable than comments. Stack Exchange specifically discourages the use of comments for questions and answers and so push moderators to try to enforce the rules. I’m sorry if this seems arbitrary, but I don’t make the rules.
    – Eric S
    Commented Jun 9, 2023 at 15:55

I am not a lawyer so I'm not sure I can answer your question authoritatively. However you state the following as though it were a fact:

Businesses do not EVER want to document their "magic smoke" so when they make patents they want it as obtuse as they can get away with, but allegedly they like to leave parts out so they aren't "instructing the competition".

It is my experience that companies do not generally try to make make patent specifications obtuse and confusing. They do try to make patents broad and defensible. In fact, since the claims can be amended while prosecuting the application, many times inventors are inclined to provide an excess of information in the specification to support the widest range of claims. Since not providing sufficient enablement risks the granting of the patent and since you can't edit the specification outside the claims, it is a risky strategy to try to hide important information.

You must log in to answer this question.

Not the answer you're looking for? Browse other questions tagged .