3

I am confused by terminology, I feel.

Google "how to cite a patent", and every piece of instruction will refer to "the patent number". Google Patent, however, does not show a patent number, having only application numbers and publication numbers. I feel I have derived the following rules of thumb (ignoring the country prefixes and the kind codes):

  1. A patent application is filed - under an application number.
  2. The patent application is published - under a publication number.

That seemed easy - now:

  1. The patent application is granted. Is there always a "grant number"? (I feel not.) Is this the "patent number"? (I feel not.)

  2. The patent grant is published - under a publication number. Is this the same or a different one than the previous publication number (I feel this may vary: different with the PTO, same with the EPO and SIPO)? Is this the "patent number"? (I feel yes, for the PTO and the EPO; I have no idea whether to refer to a Chinese patent by application number of publication number. Part of that is that do not see a grant date/grant number/grant publication number on the english SIPO database.)

Improve this question

2 Answers 2

Reset to default
4

I defer to Eric Shain's excellent answer about how to obtain the numbers from Google Patents. I will therefore limit myself to a comparison of different countries.

Broadly, every country follows the same process:

Application → (Application) Publication → Grant (Publication)

Typically, "publication" refers to the application publication, and "grant" refers to the grant publication. "Patent number" always refers to the grant publication, since the application number and the application publication do not relate to a patent (that is, a granted patent), but only to a patent application.

Every country has a number to refer to each of these. Rarely the application publication is skipped (if the grant occurs quickly). And the grant publication may not be available (if it hasn't been granted).

Publication numbers (whether in the application stage or the grant stage) comprise three parts: a country code, a number, and a kind code. The country code and number are obvious. In the kind code, an "A" always denotes an application publication, and a "B" always denotes a grant publication. The relevance of any numbers in the kind code (such as the difference between A1 and A2) vary from country to country.

An application publication number and a grant publication number may be the same (such as with EP applications), since the kind code uniquely distinguishes them. They may also be different (such as with US applications). This is country-specific.

Applications don't have a kind code, since an application isn't a publication per se.

The formats of each of the numbers differs from country-to-country.

US

A US application, publication and grant each get different numbers.

Application: US 12/343233

Publication: US 2009/0246777 A1

Grant: US 8093020 B2

EPO

At the EPO, the publication number and grant number differ only by the kind code.

Application: EP 95936210.4

Publication: EP 0783694 A1

Grant: EP 0783694 B1

WIPO

For a PCT application, the application number includes the country of the receiving office of the application (because actually the number is assigned by the receiving office, so this avoids the need to sync numbers between receiving offices). But it is still a PCT application regardless.

Application: PCT/US2008/088220

Publication: WO 2009/086415 A1

Grant: None, PCT applications are never granted.

China

At SIPO, the publication number and grant number differ only by the kind code. China changed its numbering system in 2010. The EPO provides an overview.

Application: CN 201080059371.7

Publication: CN 102656458 A

Grant: CN 102656458 B

Japan

Japan unhelpfully uses the same format for application and publication numbers, but with different numbers. This leads to no end of confusion.

Japan also changed its format in 2000. Before then, the publication number actually changed after it was examined (but before it was granted). The EPO provides an overview.

Application: JP 2015-245934

Publication: JP 2016-119900 A

Grant: JP 6141396 B2

Improve this answer
4
  • This is what I was looking for, exactly. So to summarize that as answers to my questions 3 and 4: There is not always a new number associated with the grant of a patent, but there always is one. This is the patent number. This number can be the same as an earlier (application) publication number (EPO, SIPO) or it can be different (PTO, JPO).
    – bers
    Mar 1, 2018 at 22:28
  • @bers Exactly, though the kind code will always be different between a publication and a grant, even if the number is the same.
    – Maca
    Mar 1, 2018 at 22:40
  • So while "application" and "(application) publication" are two very separate processes identifiers, as I have correctly implied in my question, "grant (publication)" is actually only one process with a single identifier. And one could intuitively say that the patent office, by choosing the format of their patent numbers to be either new numbers or publication numbers (+ kind code), stresses either the "grant" or the "publication" character of that combined process.
    – bers
    Mar 1, 2018 at 22:40
  • Application, publication of an application, and a granted patent are the same three things with the same three meanings/importance from place to place. Do not over-read anything into the number formats.
    – George White ♦
    Mar 1, 2018 at 23:35
3
  1. A patent application is filed - under an application number.
  2. The patent application is published - under a publication number.

Both of those statements are true.

  1. The patent application is granted. Is there always a "grant number"? (I feel not.) Is this the "patent number"? (I feel not.)
  2. The patent grant is published - under a publication number. Is this the same or a different one than the previous publication number...

At least for US patents the number assigned to a granted patent is the "Patent Number". This number is different from the "Publication Number" which is what the application is published as. I think EP patents have the same number but follow that up with "A" for published applications and "B" for patents. For instance EP0783694 (A1) is the publication number for patent EP0783694 (B1). It gets a bit confusing for me because there is an earlier application number and a WO number (in this case WO9610747 (A1)). This question provides a good review of the process for patents.

Google Patents is not a government entity and can label things any way they want. If you look at an example patent you see they label the document number column as "App/Pub Number" which isn't at all specific. Depending on the Status column, it might be an application number or publication number.

enter image description here

If you look at the first page of the actual patent, you will see that the patent number is actually listed as "Patent No.". The application number is the original number used when filing the non-provisional application and what is sometimes wrongly referred to as an application number but what is really the publication number is listed as "Prior Publication Data".

enter image description here

If you go to the US Public Pair site. You will see "Application Number", "Patent Number" and "Publication Number" clearly delineated.

enter image description here

Looking in PAIR for this particular patent shows all three numbers listed appropriately.

enter image description here

Improve this answer
6
  • Including the graphics is very helpful. Thanks!
    – DukeZhou
    Mar 1, 2018 at 17:47
  • Wow, so much content to comment on :) This perfectly explains the PTO case, thank you! I will wait a bit with accepting this, because maybe someone will come up with a broader (although almost surely not as detailed) answer covering other patent offices or, if they exist, general rules.
    – bers
    Mar 1, 2018 at 17:57
  • Thanks for pointing out the new Google Patent interface. In the old one, publication number and application number are clearly assigned for each entry. Maybe the fact that the redesign shows the same data differently indicates that that was not perfect.
    – bers
    Mar 1, 2018 at 17:58
  • @bers I'm open to turning this into a community wiki and expanding it. Hope you don't mind me picking a couple of my own patents as examples. Pure vanity...
    – Eric S ♦
    Mar 1, 2018 at 19:39
  • No, of course not, I love that! Although you are not necessarily doing things that I could not do myself, I must admit that I would lack the experience and confidence to draw any general conclusions from it - hence my questions.
    – bers
    Mar 1, 2018 at 20:27

You must log in to answer this question.

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