Guest Post on USPTO/EPO Joint Patent Classification System

PatentLawImage023Guest Post by Anthony Trippe, Chair of PIUG –  The International Organization for Patent Information Professionals.  I asked Mr. Trippe to provide his perspective as a professional search specialist on the USPTO/EPO joint classification announcement.

There are a lot of different ways to search for patents but arguably one of the most important ways is to find relevant patent classification codes for a technology and search for them in the country you are interested in. Ordinarily that would involve knowing the different classification systems of each of the individual countries and running the searches using different strategies for each country. Some of the drawbacks of a system like that involve inconsistencies between how classification systems are developed and what technologies they focus on, the likelihood that an individual doing the searching will do so in a country specific manner based on which system they are familiar with and the difficulty of comparing the technological content of patent documents from different countries when they are classified using different systems.

Today the USPTO and EPO took a big step forward in addressing these issues by issuing a joint announcement on a directive that they are going to start working together on a joint classification system which both companies will develop and support. The new system will be based on the European ECLA system which is itself an off-shoot of the International Patent Classification system from WIPO and will be co-developed by the two offices going forward.

The idea of a Common Hybrid Classification has been one of the ten Foundation Projects of the IP5 since the beginning so movement in this area was expected but the announcement of a joint classification system between USPTO and EPO is a distinct step forward and signals how serious the offices are about making this happen in a reasonable time frame.

This was an enormous decision on the part of the USPTO since the US has invested so heavily in its existing system which predated the IPC and ECLA systems. Director Kappos and his office should be congratulated for recognizing the enormous benefits that this change will have on efficiencies between the offices worldwide and the impact on overall patent quality.

Speaking as the Chair of the Patent Information Users Group I also can’t say enough about what this change will mean for the patent searchers of the world. Whether they work for a patent office, a law firm, are in private practice or represent a corporate entity, the ability to use a joint patent classification system will allow patent searchers to find relevant patents easier from around the world and will allow them to compare the technical content of these documents more efficiently since they will be working from a common framework. I spoke to a number of searchers about this development and to a person they greeted this decision with excitement and welcomed the opportunities that a joint classification system would represent.

While this is an exciting day there is still work that needs to be done. Tools will need to be developed to map the existing US system to the new joint system at least initially until users become accustomed to using the new system. There will also need to be work done on understanding where the existing US system can add value to ECLA and merge the two systems into the joint one. There will also need to be considerations made for incorporating other classification systems from around the world. In particular the Japanese F-Term system will have to addressed and hopefully incorporated into a joint system in the future. These issues can be addressed and it looks as if the USPTO and EPO are building momentum to make it happen as quickly as possible.

Both offices have also been working on automatic classification methods and hopefully the creation of a joint system will make this easier and will allow the groups to pool their efforts. Thinking ahead to the future the reader can imagine a situation where automated patent classification tools could be applied to other data sources besides patents. Imagine being able to use a single, specific code for a technology and be able to retrieve worldwide patent documents, non-patent literature and even perhaps documents on the web on that topic. A joint classification system could start us down this road and two of the largest patent offices in the world have taken the first step.

29 thoughts on “Guest Post on USPTO/EPO Joint Patent Classification System

  1. 26


    I have visited your blog and have found it helpful, mostly about the information about patent invalidation searches. Invalidation search is done to identify documents or prior use that may reduce the claim scope for one or more claims of a granted patent or to invalidate a granted patent.
    Thanks for the great post, very helpful information here.

  2. 25


    cheap jordans, a bot, is more on target than your posts. the classification system has zero effect on “patent law on a global scale.” The countries on that global scale have exactly the same patent law as before.

    Why would anyone track ya down and use your services when ya post so indiscriminately?

  3. 24

    Hello. this blog site is with amazing content material, extremely helpful for that operate I do. let me thank the feedback of other individuals who take part on this blog page for the reason that it serves to instruct more advantageous about it. wonderful employment. thank you incredibly much

  4. 23

    This new joint patent classification system will, indeed, be a significant step forward for patent law, on a global scale. Unless one is a staunch IP opponent, it’s hard to argue with a system that promises better patents, faster — and that’s what this joint effort has the potential to achieve.

  5. 22

    Public Searcher DIP, you are right about how, for example, being able to search US art by more than one classification system was an advantage that will be lost. OTOH, previous US administrations let the USPC deteriorate, and so ultimately that led us to where we are now. This was very short-sighted of them, and apparently born out of a mistaken assumption that keywords alone would be sufficient. Many people, both outside and inside the USPTO, pointed out their mistake at the time.

  6. 21

    Conceptually I like the idea of a unified classification system. It certainly helps when everyone speaks the same language. Assigning classes to documents under any particular system needs consistency. I see a big challenge coming when different organizations (USPTO, EPO and anyone else who adopts it) have different processes for doing the assignment. We may end up speaking the same language using different dialects.

    Even so, finding common ground will aid all patent researchers – at least to some degree. I wish all involved well in this endeavor.

  7. 20


    Godd posting and great job in leading PIUG from obscurity to relevance.

    A single patent classification system will be a boon to the patent offices for training and prior art searching. Instead of trying to teach and utilize three (USPC, ECLA, IPC) incomplete systems, they should be able to cobble together one complete system. Hopefully the new found love of classification exhibited by the Europeans continues and the US can resume their contributions which have been on hold for ten years.

    There will be a couple of downsides for professional searchers that do not really apply to the examining corps. The redundancies inherent in multiple systems allowed us a double check on our work. And knowing when to use each system was one of our secret weapons, particularly for invalidity work!

    Alun: Keep beating the drum about US Examiners inability to access ECLA codes. It is embarassing that they can not. Especially with the DWPI file!

    David Lewis: I mean no disrespect, but you should definitely learn about ECLA if you’re searching LCD art. The US classifications for LCD display structure pales in comparison in both fine detail and completeness. Take a look at this to see what you’ve been missing:
    link to

  8. 19

    It is not clear to me that using a joint classificaiton system is good. I have not made a comparison lately, but I know that for example, there used ot be 52 US subclasses for liquid crystal displays (most of which were short and easy to search) and only 2 very long nearly impossible to search subclasses in the international classification system. At times various people have suggested that fewer subclasses are better, which is a major falicy. The fewer subclasses, the fewer categories there are for categorizing things, and the longer each subclass is and the more difficult it is to do a search. The point being that if we substitute our current classification system with one that has far fewer subclasses categorizing references, it may be far harder to find anything. Of course, more and more examiners and professional searchers rely on keyword searches. So the less anyone actually uses the classification systems, the less important this issue becomes.

  9. 18

    Classification is the key. At some level, it boils down to correctly cataloging each of these things. I suggest you contact Olik at PTO classification USA and get him involved. There are now decades of neglect, and growing every day.

    I’d like to help, but I am merely an orange monkey on the outside looking in.

  10. 17

    Yet another caveat. ECLA has a backfile for 1920-1974, classified by the old Dutch classification. The latter was only ever published in Dutch and French, so I used the French edition, because I never learnt Dutch. If you don’t speak French either, you would be SOL. Then again, this backfile is not covered by EAST/WEST anyway.

  11. 16

    Just to note that nothing I said has any bearing on keyword searching or searching by fields other than classification, which is, as you might say, a whole other subject.

  12. 15

    I’ve been a patent agent in the US for a decade, doing only prosecution for at least the last five, but I was a patent searcher for 16 years, both in the UK and the US, and I have also searched in Munich.

    This is indeed a takeover disguised as a merger. The USPTO had failed to keep up with classification for some time before I quit searching. OTOH, the EPC is a good system, and is updated reasonably frequently, perhaps not as often as the USPC was at its peak, but far more often than the latter is nowadays.

    The EPC has the advantage over the US system of using the top-down hierarchy of the IPC, which makes it much easier to locate the right classification, but with a further layer that the IPC doesn’t have, making the number of references to be reviewed more reasonable. I see the USPTO as contributing to this bottom layer.

    The last I knew, EAST and WEST in the USPTO were already searchable by USPC, IPC and EPC, **but** with the EPC truncated to the same level as the IPC, i.e. omitting the vital lowest layer of the EPC. This issue is generally not understood by searchers who have never used ECLA, which is the English language of an EPO database that classifies patents from 10 countries by the full length EPC (FCLA is the French version and GCLA is the German version). Everything in ECLA is in the EPO file on EAST/WEST already, but **only** searchable by IPC or **truncated** EPC.

    Whatever we do, we musn’t let the USPTO get away with ditching the USPC unless they first add the **full length** EPC to EAST/WEST, in both the US file and the EPO file, and for both the Examiners and the public. That is likely to be the real issue of importance here.

    BTW, the US is one of the countries in ECLA (which is why US refs also come up when you search the EPO file in EAST/WEST), so US patents/published apps already have a full length EPC, which means the USPTO have no excuse. It’s not as if they would have to reclassify the old art.

    PS: Just to be clear EAST/WEST and ECLA are completely different from the corresponding databases available on the Web.

  13. 14

    I don’t know about this. People say epo classification is better, and in some respects it is (it is updated fairly well I hear, and it seems to be). But that is not the end of the story. The way they break things down is as bad or worse than the US system. And that is a close 2nd concern to whether or not it is updated.

    I also note that still there are cases that come in with no classification, even under the EPO classification system.

  14. 13

    I like the U2 song “One”, thanks Ping for sharing with us.

  15. 11

    Did B just get punished for crossing the RED?
    Did MM leave because of this, and he had no control? How sad is that?
    I know now that Texas is the key to my nightmare.
    And that is why the Rabbit ran.
    Is tht what you want?
    Otherwise if you can’t tell me straight out, I don’t understand. And you got me going in circles.

  16. 10

    No because I don’t understand what he’s trying to say, or ask? I’m tired of guessing.

  17. 9

    Ping are you trying to compete with Sarah?

  18. 8

    To those not familiar with them, perhaps I have to explain the Japanese F-terms mentioned both in Mr. Trippe’s post and in my comment. The JPO uses, apart from the IPC-based FI classification, a very fine-grained indexing system called F-terms. Whereas the FI classification relates to the general field of the application, each F-term refers to an individual feature of the document. Each document can therefore receive quite a handful of F-terms, which can then be used in searches in place of keywords. This enormously improves both the speed and accuracy of searches, and also helps overcome language barriers. The downside is, of course, that such a supplemental feature indexing system requires a lot more effort to maintain than simple classification. Even the JPO hasn’t applied it over the board.

  19. 7

    Two for Tuesday:

    You say
    Love is a temple
    Love a higher law
    Love is a temple
    Love the higher law
    You ask me to enter
    But then you make me crawl
    And I can’t be holding on
    To what you got
    When all you got is hurt

  20. 5

    All of this, all of this can be yours
    Just give me what I want
    And no one gets hurt

  21. 4

    I’m not a searcher and know nothing of Japanese F terms. But it is plain for all to see, that Rodrigo’s post is full of good stuff. Two points particularly struck me, namely

    1) his plea to take the best from the Japanese system into the ECLA system. If somebody else does it better, copy them of course (unless they have got a patent on it)

    2) his plea to resist embarrassing political gimmicks. Hear him, hear him.

  22. 3

    As in many more aspects of its patent system, the US was an outlier with respect to patent classification: both the European classification ECLA and the Japanese FI classification (not to be mistaken with the F-terms) are IPC-based, and most other countries also work with the IPC. The completely different US classification system was indeed a serious complication for searchers. It also lead to significant duplication of work.
    This said, I hope that, if the USPTO has committed itself to this change, it will also dedicate enough resources to the transition, because it will require a lot of effort to train its examiners in the new classification system, never mind reclassifying all the old patent literature. Cooperation with the EPO will help, but the EPO has already its hands full maintaining the EPC for its own purposes, so the scope will be limited. The EPO and the JPO have been working together for a while, yet the EPC and the Japanese FI classification are still far from completely harmonized.
    I also hope that, in the future, the IPC will be kept free of further political gimmickry like the creation of the Y02 class (“Technologies for mitigation or adaptation against climate change”). A patent classification system is a work tool, not a means for showing the patent offices’ commitment to the cause du jour.
    Finally, I also agree with Mr. Tripp’s desire to see the Japanese F-terms more widely adopted, since they are possibly the cleverest patent classification system around, and fully compatible with any IPC-based system.

  23. 2

    I see it differently, Plurality.

    Think how often in industry and private practice a straightforward take-over is presented (to reassure the stake-holders) as a “merger of equals”. With that thought in mind, go back and read the above guest post, and the press releases.

    You said it yourself: USPC is not keeping up with the onward march of technology. You suppose the same is true of the ECLA. Wishful thinking, perhaps? The EPO has enough funds to keep its subject matter classification system up to date, and enough motivation to do it right.

    In future, how will the USPTO search business methods, programs for computers, methods of surgery, all the things that under the EPC are not eligible? Mr Trippe doesn’t say.

    Meanwhile, Mr Trippe, already I see EPO search reports citing ever more non-patent literature. For Europeans, this is nothing new. Others of your readers might have to “imagine” this happening. We not.

    I welcome this initiative. Nobody gets hurt. Everybody wins.

  24. 1

    I’m extremely skeptical about the benefits from switching to such a system. Setting aside all the concerns about examiners transitioning to a new system, I still see two big pitfalls that could emerge from this project.

    One, the political inertia of USPC and ECLA is likely to create a hybrid beast that is the result of disagreement and compromise, rather than an attempt to design a system that fulfills the primary design goal of helping examiners more easily find prior art.

    Secondly, a system that spans multiple patent offices will be mired in even more red tape than the current systems are. The USPC (and, I’m guessing, ECLA as well) suffers from class listings that fail to keep up with advances in the art. The result is subclasses with thousands of references, thus requiring the use of keywords to narrow down results rather than permitting a full search of an individual concept. The reason why USPC has not received the attention and maintenance is a lack of funding for such projects.

    But with multiple agencies affected by reclassification projects, the biggest roadblock will shift to be perpetual red tape in terms of getting the agencies to repeatedly agree on the proper classification schedules for hundreds of different areas of technology. The perpetual lack of funding will only serve to compound the problem.

    We sorely need an update to the USPC, particularly in rapidly-moving areas of technology. There are applications coming through all the time that can’t be classed correctly because there are no subclasses for those technologies. The money could instead be spent on updating the individual class systems, probably for a fraction of the cost. But I understand that this is where the political winds are blowing at the moment… so we can only hope that the offices do this right instead of just making the problem worse.

Leave a Reply

Your email address will not be published. Required fields are marked *

You can click here to Subscribe without commenting

Add a picture