0

I am working on drafting a provisional patent application (PPA) for an algorithm (which is novel, non-obvious and useful) and the presentation of the result of running of the algorithm on some data. The presentation is also novel, non-obvious and useful. I am wondering if I should file separate PPAs for each or if I can file one PPA and later (within 12 months) file for two separate non-provisional patents.

In case I am able to file two separate PPAs, for the PPA for the presentation part, since I need to enable it, do I need to disclose the algorithm or can I make the algorithm part generic saying something like "an algorithm that returns a score based on consideration of, but not limited to, factors X, Y, and Z" ?

3
  • Not an answer to your question, but be advised that an algorithm, by itself, is almost certainly not patentable. The application of an algorithm to perform a specific technical application might be patentable.
    – Eric S
    Sep 30, 2020 at 19:40
  • Thanks @eric-s I'm drafting the spec to include the technical application of the algorithm (not mentioned as such) in the main as well as other relevant embodiments.
    – vikx01
    Jun 25, 2021 at 16:21
  • Just including applications in the specification as examples isn’t good enough. The actual claims will need to be specific applications or else they’ll be rejected as abstract.
    – Eric S
    Jun 26, 2021 at 3:47

3 Answers 3

3

As @EricS mentioned, the word algorithm will trigger the word "abstract" which is hard to get around once it is invoked.

It doesn't really matter if it is one or two provisional applications since one or more non-provisionals can get the benefit of one or more provisionals. No one-to-one correspondence needed.

The only reason I can think of would be subject matter you might want to reserve the possibility to keep a trade secret. At the end of the year if no non-provisional is filed that refers back to a provisional, it never publishes and is not viewable by the public. Bear in might that whatever you do file a non=provisional on must be enabled - specifically the best mode you know of must be enabled. No hiding the ball.

3
  • I am guessing trade secret might be difficult to maintain. So I am considering going full in with the patent.
    – vikx01
    Jun 25, 2021 at 16:23
  • Thanks @george-white for the info regarding no need for a one-to-one correspondence between provisional and non-provisional. The only question I have regarding this is if the latest of the date of filing of the one-or-more provisionals on which a non-provisional depends is assigned as the priority date. I think I know the answer though, but just want to be sure. Maybe deserves another AskPatents question.
    – vikx01
    Jun 25, 2021 at 16:30
  • I’m not 100% sure which date will be listed as the priority date, most probably the earliest. But if any potential prior art comes up it will either be or not be fair game as prior art based in the dates and contents of the various provisionals.
    – George White
    Jun 26, 2021 at 5:31
2

As previously mentioned, this is a challenging area for patentability and it is more important that you draft the application in sufficient detail to describe the practical application of the algorithm and the inventive concept of your invention so you can argue against a potential rejection under Section 101. It may not be in your best interest to file a skinny provisional application. You may want to take a look at some of the USPTO materials that address the patentability of computer-implemented inventions. In particular, these examples may be helpful for identifying how the USPTO analyzes patent claims directed to certain computer-implemented inventions.

2
  • Please use the link tool to make your links actually work. Otherwise this is a nice answer.
    – Eric S
    Oct 1, 2020 at 22:51
  • @ipesqnyu Thanks for the SME links. I'll try to make the specification detailed enough.
    – vikx01
    Jun 25, 2021 at 16:19
2

Regarding the decision of filing one or two provisional applications, George White's answer is very on point. You should be aware though that presentations of information are not patentable in many countries, for example most if not all European countries. At the EPO, this is enshrined in Art. 52(d) of the European Patent Convention. When presentations of information are not claimed "as such", then they may be patentable, but pretty much presentations of information are only inventive (i.e. non-obvious) when the following occurs:

"A feature defining a presentation of information produces a technical effect if it credibly assists the user in performing a technical task by means of a continued and/or guided human-machine interaction process (T 336/14 and T 1802/13). Such a technical effect is considered credibly achieved if the assistance to the user in performing the technical task is objectively, reliably and causally linked to the feature. This would not be the case if the alleged effect depends on subjective interests or preferences of the user."

The above excerpt is from the following part of EPO's Guidelines for Examination, which I strongly advise you to read should you want to protect the presentation of information.

As for the algorithm itself, in line with ipesqnyu's answer, I suggest you to read the EPO's Index for Computer-Implemented Inventions of the Guidelines for Examination should you be willing to protect it in Europe.

2
  • Thanks @the-europeist for the relevant clause from EPO and the link to the index. That link doesn't work anymore. Here is the latest one: epo.org/law-practice/legal-texts/html/guidelines/e/j.htm I don't intend to file in europe but it's good to know.
    – vikx01
    Jun 25, 2021 at 16:33
  • @vikx01 Thank you for the heads up. I have updated the links. Jun 26, 2021 at 17:37

You must log in to answer this question.

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