56 thoughts on “Patent Searching Tools

  1. “USPTO examiners typically discover better prior art than that found by applicants, or other foreign patent offices”

    It is a widely held belief by most competent practitioners that the EPO does the best searching.

    Sorry.

  2. “USPTO examiners typically discover better prior art than that found by applicants”

    start with a faulty premise…

    fupe, did you think the obvious reason this might be may have more to do with applicants not submitting applications when they have found good prior art and less to do with search tools of the examiners?

  3. Since, in my experience, USPTO examiners typically discover better prior art than that found by applicants, or other foreign patent offices, I believe that the USPTO tools (primarily, EAST and WEST), which are used to search similar databases, are superior.

  4. Any EPO examiners out there, could you tell us about the incentive structure at the EPO? What’s the average amount of time you are given per application?

  5. The classification system at the USPTO is really messed up. Often it’s not organized logically to begin with, and hasn’t been updated for a long time so there may be thousands of documents in a generic subclass. And PG-PUBs may be classified by a computer that is wrong as often as it is right.

    A lot of NPL is technically available to examiners, but they are not given enough time to use it.

    Another reason EPO searches are often better is that EPO examiners get more time for examination. In my old art unit, we got 14.8 hours per case, while the equivalent European art unit got 32 hours per case.

  6. Actual,

    My point is that your posts look so much better when I am NOT sobered up.

    C’mon, you can figure out the point in the specific causal relation mentioned – as I said, even my pal 6 can.

    6, this round on me.

  7. Posted by: Einstein drinking with 6 | Feb 17, 2009 at 03:20 PM

    “Actual Inventor pay attention.

    There is a causal relationship between litigation and prosecution.

    Even my pal 6 knows this.

    6, your turn to buy the shots.”

    From a modern physics perspective there is a causal relation between everything. So what’s your point? Oh, and please sober up before responding.

  8. Actual Inventor pay attention.

    There is a causal relationship between litigation and prosecution.

    Even my pal 6 knows this.

    6, your turn to buy the shots.

  9. “Maybe not, but I’m willing to bet pennies to dollars that the sucess rate in litigating a patent is through the roof higher when using their services unless you’re a vet of the system already yourself.”

    Posted by: 6 | Feb 17, 2009 at 12:59 AM

    6 pay attention. The question was about writing patent applications not about litigating. As an “Actual Inventor” I would spare no expense in hiring a top patent litigator should I have to go to court to enforce or defend a patent.

  10. Falciola, could you please post again the link for your publication (WPI) ? The address is not working.

    Thanks

  11. But how about free online patent information search and analyzing tools? How much examiners and (professional) patent searchers use free tools? How do you compare the efficiency and quality of free searching tools and commercial (professional) searching tools?
    In addition to tools that Kyle and IPBloke cited there are tens of free patent and non-patent information searching tools, clustering, analyzing etc tools (I made a list about some tools, what I have used: link to ipestonia.ning.com). Are the searching tools used by examiners more powerful?
    Kyle asked very right question – why internal search tools of patent offices are not available to the public use?

  12. A: you have hit on the big issue, namely, how can a Statute define it, that the state of the art is bigger than what any one PHOSITA can possibly know. Is that fair? Is that practical? Is that reasonable? How can it Promote the Progress? I ask you, it’s mad, isn’t it?

  13. I have a question, how does one find prior art that is made available (to patent offices) by way of oral description and by way of use before or during patent prosecution. Does one search on these aspects as well? Is this something one may come across but only during litigation?

    These kinds of prior art may or may not be relevant in the US, but would like to find out on them from readers coming from other jurisdictions.

  14. “Especially since no patent attorney can make any guarantees or produce stats from an objective study showing the success rate is higher when using their services.

    Maybe not, but I’m willing to bet pennies to dollars that the sucess rate in litigating a patent is through the roof higher when using their services unless you’re a vet of the system already yourself.

  15. “A typical search on STN will be several hundred dollars.”

    I had the stic send a search back to me because they didn’t want to spend 13k on the search. I resubmitted it to the chemical arts STIC as a spe told me to. They kicked it back to my STIC (and told them to STFU about the costs no doubt lol). One of the searcher’s spes came and talked to me about my exp with the STIC and wanted to see if I was satisfied (i.e. wanted me to rat out the other guy who kicked the search back to me). I liked the guy so I didn’t rat him out, but I feel like it would have been pretty easy to find out who it was since there’s like 3 chemical experts in our STIC.

    One application searched = ~14k+the regular 4kish.

    So, did you guys hear about the budget crisis we’ll be facing soon since nobody is submitting hardly any apps? This must be murder on prosecution firms. Apparently the budget crisis will hit way before the backlog takes a substantial hit. But who knows, this recession could bring the backlog under control :) I forsee an increase in fees in your future. The cost of examination is simply way higher than is being paid.

  16. Posted by: A | Feb 16, 2009 at 08:29 PM wrote:

    “Just wondering how many actual inventors actually follow that route, and more importantly – what is their success rate in obtaining patent grants?”

    One would have to reason an “Actual Inventors” success rate would be just as good if not better than hiring a patent writer/patent attorney.

    Especially since no patent attorney can make any guarantees or produce stats from an objective study showing the success rate is higher when using their services.

  17. I’m a USPTO examiner in the chemical arts and can’t comment on our search tools as compared to professional searchers but I’ll just list of some of our more unique databases available:

    About 90% of a typical search for me runs through EAST. I US, IPC, and Derwent’s CPI classification schemes along with word searches.

    For applications where I feel that there is significantly more non-patent literature, I search STN (Registry, CAPlus, USPatfull, SciSearch). STN is a lifesaver for composition claims because STN has indexed many of its holdings by components and their percentages. A typical search on STN will be several hundred dollars.

    Again for composition claims, there is a german database that has indexed US, EPO, DE, and JP patents by composition and is free of charge to USPTO searchers.

    From time to time, I’ll log onto Dialog to use INSPEC’s database.

    For most author searches I go through ScienceDirect, Scirus, Google Scholar, SpringerLink, and Wiley.

    For general reference materials, we have access to the entire ASM Handbook series (20-something volumes), several online libraries, and our very fast inter-library loans and document delivery through our internal library for things like an ASTM or MIL standard.

  18. Just wondering how many actual inventors actually follow that route, and more importantly – what is their success rate in obtaining patent grants?

  19. Mr. MaxDrei Wrote:

    So, we have a pioneer who files in a technical field where “Prior art is seldom an issue”. Whatever field could that be then?

    Dear Mr. Max:

    I can’t answer your question because its based on a statement I did not make. Please allow me to correct you. I said, “as an “Actual Inventor” that only files applications on pioneering inventions ” Prior Art” is seldom an issue.”

    Perhaps your true issue is with the definition of pioneering invention. If you have a definition you use in the EPO I would be interested in hearing that.

    Mr Max also wrote:

    “Might I ask you two things, Actual Inventor? The number of parties who recognise the value in your work, and beg you for a licence; disappointingly small is it?”

    Actually I prefer to keep my inventions trade secret and mostly use them for my own commercial interest. So again your question is largely irrelevant.

    Mr. Max further asked:

    “Second question: do I detect that, although you don’t run into art during prosecution, you do face other problems. What would they be then? ”

    As an “Actual Inventor” in the USA I am very fortunate. Our Patent Statute is very clear and easy to understand. So easy in fact that “Actual Inventors” do not even need Patent Attorneys to write and prosecute a patent application.

    Our MPEP is on the web as is any court case decisions that are relevant to the invention.

    Plus we have something called an Inventions Assistance Center, staffed by actual Patent Examiners that are available toll free to answer any questions an “Actual Inventor” may have about the administrative procedures of the USPTO.

    I have found this service to be particularly helpful and the staff always courtesous and friendly and willing to do whatever they can to help, “Actual Inventors.”

    Also its quite easy to pick up a Patented Yourself Book for a general overview of USA patent concepts. Although books cant be relied on to keep pace with changes brought about from court decisions it’s still helpful.

    So as you can see Mr. Max, there is a wealth of resources here in the US for the “Actual Inventor” to write and prosecute their own patent application without any problems at all!

  20. So, we have a pioneer who files in a technical field where “Prior art is seldom an issue”. Whatever field could that be then? My experience is that the commercial value of technology goes in inverse proportion to the distance between it and the most relevant “prior art”. Might I ask you two things, Actual Inventor? The number of parties who recognise the value in your work, and beg you for a licence; disappointingly small is it? Second question: do I detect that, although you don’t run into art during prosecution, you do face other problems. What would they be then? Just interested.

  21. “I’m an examiner.

    I’m not sure what public searchers use.

    Posted by: Hecky’s: It’s the sauce | Feb 12, 2009 at 11:45 AM”

    Dear Mr./Ms. Hecky’s: Its the sauce

    As an “Actual Inventor” I use the USPTO web site, Google, Google Patents, Yahoo, MSN, Ask.com, and trade journals or publications in the field of us, and finally the EPO web site.

    Fortunately, as an “Actual Inventor” that only files applications on pioneering inventions ” Prior Art” is seldom an issue.

  22. In my opinion, the real question is not “what search tool are you using?” but what data are you searching? There are an ever-growing number of open source, semi-free, commercial and official patent databases on the web, and federated search systems like Scitopia, Scirus and Science.gov that search both patents and NPL collections. Many of the non-official patent search sites do not disclose their data covearage or update frequency. This is unhelpful for a patent searcher; I can’t imagine searching a patent database and not knowing what I was searching. This is my main complaint about Google Patents/Google Scholar. Google Patents is fine for quickly retrieving a known patent, researching your great-great-grampa’s patent, or finding a few representative patents in a field of technology, but for doing a comprehensive patentability or validity search? No way! Google Patent’s USPC/IPC data is terrible. It’s not current and their are serious formatting problems, e.g. 16/22 displays as 1/622 or 162/2.

    The USPTO’s WEST internal patent search system and its public web-based databases are very similar. First, they’re both based on the BRS/Search full-text search and retrieval system, which was developed at SUNY Albany in the late 1960s, spun-off as a commercial product in 1976 and is now owned by Open Text. Their data is identical; both are updated weekly on Tuesdays and Thursdays when new patents issue and applications are published. The type and number of searchable indexes are almost identical; I believe that WEST has a few fields (e.g. kind code) that are not available to the public. The main different between WEST and the public databases is that WEST contains OCRed full-text from 1920-1975. In the public database, patents from 1790 through 1975 can only be retrieved by number, current USPC, date, and application type.

    I have heard rumors that the USPTO may be working on a project to make WEST available to the public via the USPTO website. It is already available in designated patent depository libraries. Apparently, the USPTO obtained Congressional approval last year. I’ve heard similar rumors regarding the EPO’s EPOQUE system.

  23. I was a public searcher for 16 years, first in the UK and then in the US, becoming a US patent agent towards the end of that time, and I now only do prosecution.

    IME, I agree with those who say that EAST is the best tool for searching US patents. EAST is available to the public in the PTO and AFAIK also in some Patent Depository Libraries (PDLs) elsewhere in the US (I think?). Online systems, including the PTO’s own, are very much inferior to EAST.

    EAST does include foreign patents, but you have to deliberately select two additional databases before you search, which I always did. However, unlike the US art, the foreign art is not full text, just abstracts, and the classification searchable for the foreign art is limited to IPC and a neutered form of ECLA that is truncated to leave off the last part that would have made it narrower than IPC and thus actually useful! For these reasons, retrieval of foreign art is much worse than for US art. The truncated ECLA doesn’t even work for JP art atall. BTW, most US patents are in the wrong IPC, so only use USPC for US art.

    IMHO the EPO find better art because they use the full ECLA, not the truncated version searchable on EAST, and because they rely less on keywords.

    As for non-patent art, most database hosts like DIALOG or STN are too expensive for public searchers to use except on validity searches, whereas I understand that the USPTO and some large corporations get unlimited use for a (high) flat rate, but that only works for large organisations, and most search firms on either side of the Atlantic are small.

    BTW, DIALOG and STN are hosts, not databases. A lot of the same databases are available on either one, but the search language is different. For those who used the old APS at the PTO, the search engine was licenced from STN, so APS search skills are transferrable to STN, but for DIALOG you have to learn different operators, etc.

  24. I agree with Public Searcher DIP (above) — there’s no one tool.

    A few things to add:

    1) Here’s a site a buddy and I created which allows you to search many of the resources mentioned here from a single interface: http://www.priorsmart.com

    2) Wouldn’t it be nice if the patent offices made their internal search tools available to the public? [Minus the private stuff...] What’s preventing this?

  25. There is no magic bullet for searching. We need good classification of patents (both US and ECLA/IPC), good text searching of patents, and better text searching of NPL (full text, proximity operators, search history, and citation links). I wish we could search NPL in the orderly manner we search patents with one database, but that just doesn’t exist yet to my knowledge and probably won’t for a while.

    Every technology and search is different, so the examiner or searcher has to use the tool most useful to that specific case. And no tool is inherently better unless you work in a narrow field or have a very specific need. (I’m biased to EAST since I get to use it for free, but will definitely acknowledge it’s US orientation and limitations.)

    For example: If you want to review 10 references that use specific terms, google patent search it. If you want to review the claims of 2000 US patents with images at the same time, use a different tool like East or it’s equivalents (Qpat, Delphion, Derwent, Thompson, etc.). If you want to review NPL to find an applicants own disclosures, use yet another tool like Dialog, STN, IEEE, ACM Portal, Sciencedirect, Web of Science, Google Scholar.

    I do wish the USPTO would put forth more effort to keep their classifications relevant as Europe seems to be doing. US Publications in particular are being very poorly classified which reduces the usefulness of classification based searching. Europe though, really needs to reclassify all the old art that still resides in the outdents before they can claim the upper hand!

    DIP

  26. I work as an EPO examiner. Although,I cannot compare our tools to that of my counterparts at the USPTO, I do however agree with the comments of Mikk and especially with Sarg. Our classification scheme is the backbone to our success. We get time and money to manage, create and tweak it – this is why it produces such exact results.

    Mark makes the comment about EPO searches being of a higher quality that that of the USPTO. This is a subjective opinion and not one that I fully agree with. I often cite prior art discovered by the USPTO. True, that sometimes there is perhaps better prior art to be found but this is more to do with the individual than the organisation.

  27. Hey, thanks Mikk. That’s a striking set of 30 ppt frames, showing how much the EPO is already doing to search non-patent literature. Chinese and Indian traditional knowledge, Google Scholar etc. It’s all there. I’m impressed.

    But, for EPO President Brimelow, that’s totally inadequate. She’s currently making a big song and dance about the importance of searching stuff published in Asian languages, Chinese, Korean, Japanese, that sort of stuff. Lucky that her EPO is adequately funded. They get it from claims fees you know. So, keep filing loads of claims at the EPO please, all readers.

  28. When I visited patent searching conference search matter (organized by EPO) and had a opportunity to spend a day with examiner, then he introduced me examiner’s of EPO search tool EPOQUE – this is probably most powerful search tool I’ve seen.

  29. For anyone interested in this technical issues, I would strongly recommend the reading of the journals such as “World Patent Information” and “Scientometrics” where you can have an idea of the avalible tools and what you can actually do with them. I wrote an article in WPI on the search and the analysis patent information in the field of biotechnology that can be downloaded for free at link to sciencedirect.com.

  30. I think the EPO uses a system called Epoque. A commercial version of the system is available called PatAnalyst. (www.patanalyst.com). Personally think it is a very good tool because you have access to all the figures + full text in a single view. Also a very powerful macro based search language and a good stuctured browsing interface. The site seems to be dead though. In the EPO version, I think it integrates all the available full text patent databases + Derwent + non patent literature.

    Thomson has a product called thomson innovation which is good due to the data sources. It combines NPL+DWPI+DOCDB+FULLTEXT in a single tool. But the searching interface is pretty basic and the browsing is not so good.

    Patbase from minesoft is very good for searching and browsing, but the data available is limited to DOCDB+FULLTEXT

    /Marc

  31. Hecky’s–thanks for the response. Medline is a nice alternative to Science Citation Index, but it doesn’t seem there is an equivalent for software/communication technologies. Sounds like the USPTO needs the web interface.

  32. I think there is some way to access Science Citation Index/Web of Science, but it is pretty clumsy — not nearly as nice as the interface we had at NIH. Basically you have to know what you are looking for in the Dialog/STN versions. With the web version it was much easier to stumble from one interesting link to another.

    However when we search medline or biosis (or similar databases) with Dialog/STN, we do uncover a lot of stuff. I am still amazed by the esoteric journals i find cited whose names i have never even heard of before. And STIC (our library) can get just about anything through inter-library loan. Even the esoteric stuff appears in my inbox within 2-3 days.

  33. For examiners: What access to non-patent literature (NPL) do you have?

    For instance, for cost-intensive searching in litigation, I’ve found Web of Science and Science Citation Index to be very valuable, especially for software/communications patents. These DBs show backward and forward citations for articles, which–like Google Patents–can allow for rapid searching within a chain of related art.

    Are those or similar tools available at the USPTO? EPO? Other agencies?

    Professional searchers: Do you use these tools?

  34. People seriously flip front pages? I’ve seen gems on FIG. 12 that would never have been apparent from the front page.

  35. For mechanical examiners needing drawings EAST with its fast flip rate is by far the best.

    EAST can not text search back past 1920 but Google Patents can.

    According to MPEP 904.02(c)US Patent examiner’s need to be careful about giving out too much information in an Internet search if the application is unpublished.

    “The Patent Librarian’s Notebook” link to patentlibrarian.blogspot.com
    discussed a new search site:
    link to boliven.com
    which has an OK flip rate of the front page when using its “Quick Flip” feature.

  36. East is great for all of the reasons listed above, and the more experience you have with it, the better your searches will be. I agree very much with Public Searcher DIP. I also use many proximity operators, such as distance in a sentence, same paragraph, etc. I also try to use as many angles as possible, meaning various combinations/subcombinations of terms.

    Using STN, USPATFULL which is supposed to be full text searchable does not seem nearly as easy or reliable as East, IMO. Same for PCTFULL and EPFULL (?), hence the reason I do not get a good deal of patent lit from the ROTW. EAST only allows for a search of EP and PCT abstracts.

    I do use STN for the NPL, and think that the databases available and interface are fairly good. Bottom line, the more searching you do, the better you get.

  37. EAST ranks results in the browser based pretty much on Date.

    Google Patent ranks results based on their magical algorithm that ranks based upon word occurrence or what not.

    In my experience, i have had much more luck using google patents to get a search going than EAST. However, once i find a few close pieces of art, EAST becomes a much more powerful tool to go into in-depth searching by class, etc.

    I think the PTO seriously needs to look into licensing an in-house Google Patents type database for initial searching.

  38. DIP: I’m pushing the ECLA. Obviously not everything has an ECLA class, but what does is well sorted. It’s great if you’re looking for a quick novelty destroying reference or fishing for synonyms. If you don’t find knockout art, then of course you have to go up to IPC, but that’s a snap because ECLA uses the IPC tree as a skeleton.

  39. Sarg, Question for you. Are you pushing the ECLA classification (brand new), the EPC classifications or the IPC classifications? I can’t tell. US Examiners can search by IPC and EPC, but often don’t for whatever reason. This is not a problem with the tools they have, but the training they have received and continue to receive. I’ve worked with several dozen ex-examiners and 95% were terrible patent searchers. Great at narrowing a search to 4 references (see any file wrapper), but not so great at reviewing several hundred, using multiple strategies, or adjusting on the fly.

    6, To solve the good pats in your big sub prob do this:

    L1 = Good subclass
    L2 = All your great terms and synonyms
    L3 = L1 and (L2 or L3)

    Put the images in your right window and the KWIC (key word in context) on your left window. Now you get your cake and eat it too.

    Dennis, If you want a good comparison of third party vs East vs EPO databases, ping the http://www.PIUG.net community. They are much more search oriented than this blog’s community. They’re a little chemical focused and mostly in-house searchers using 3rd party software users, but they can answer any of your questions.

    Anybody, when you visit the PTO for an examiner interview, go check out the East search engine. It’s free on the first floor of the Madison building in the Public Search Room. The Search Room of Crystal City (scary people) it is not. Very sterile and quiet. For doing clearance or full validites, it is fantastic. For doing quicky patentability searches, use whatever you’re using.

    DIP

  40. I have used EAST for years, but have recently moved from Alexandria and was forced to look into other resources. As a result of many months of evaluating the different tools available online (both free and fee-based), I can say that EAST is not the most inclusive tool, but its speed is what makes it the best resource for patentability searches or any patent searches that only consider US art. Nothing comes even close to it.

    As soon as you wish to incorporate foreign patent art, other databases will provide a much better coverage. The same databases will also include a lot of fancy features (exporting of documents, claim charts, graphs, image mosaics, etc), but their most evident drawbacks are slowness, lots of server problems and computer crashes, extended family structures lethal to searching, and lack of good operators and commands, in some cases.

    Nowadays, not having EAST at hand, I rely on a range of tools. EP classification is oftentimes a great targeted approach. Non-patent literature has also become more relevant.

  41. “In most cases I end up completely disregarding the EPO references.”

    An examiner disregarding references from another? Well, I never!

    The other benefit of using the EPO search is that you can find stuff from around the world vs. just the US. Given most of the comments here, you’d think that only the US generated prior art.

    Not saying the EP class is the silver bullet for all art areas, but it’s a very good place to start because of its high resolution (vs IPC) and logical hierarchy (vs. US).

  42. “As a speed example, a front page picture search of 3 or 4 thousand patents can easily be done in a few hours. You can flip at 0.5 seconds per patent with no delay.”

    That is a major benefit, as well as a major shortcoming of East. What I would like to see is front page + the next 5 pages in mini format off to the side in a different window. I have two whole screens to work with here! I can’t count the number of times I’ve zipped through a 2000+ sub, then word searched it after, and read more than the front page + 1 or so, and then found what I was looking for in Fig. 3. Then again, zipping through also nets a lot of art …

  43. “To search all those databases for real was a multi-thousand dollar endeavor.”

    Welcome to the PTO’s world. Except we simply consider that a cost of doing business and spread it out to everyone who files for and maintains a patent. If applicants actually had to pay for the examination of their patent then they might think twice before filing.

    Obviously I’m a google man, so if public searches and EPO guys don’t have access to that then yes, my search tools are better.

    I get a lot of good stuff from the EPO, but many times it is useful in a way that is different than how the EPO considers it useful.

    “***The European classification system*** = one of the best”

    Sad but probably true.

    There are other good search sites out there as well. One of which is:

    link to emerjent.com

    That one helped me bust a few cases that google/East failed me on.

  44. As an examiner in the electronic/computer arts, I primarily rely on text searches using EAST. As the first commenter noted, EAST has a lot of useful options to narrow down search queries such as contextual searches (words in the same sentence or paragraph), field specific searches, etc. These options, in my opinion, give you a lot more to work with than, say, Google’s patent search feature, which I assume is one of the tools searchers outside the USPTO are using.

    Having said that, I do still occasionally perform patent searches using Google for two reasons. The first is that it gives me different results, which may sometimes be helpful if I get stuck. The second is that it highlights search terms in the actual scanned image of the patent, something EAST doesn’t do (EAST only highlights the text version of the patent, not the image).

    Regarding EPO searches, based on my experience I would disagree with the previous two commenters. Whenever I get a case that includes an EPO search report, I usually find the cited references to not be as good as something I could find on my own. In most cases I end up completely disregarding the EPO references.

  45. “What tools are the EPO exmainers using that gives them better results?”

    ***The European classification system***

    End of story. I do searches all the time, and it is the first thing I try, always. It is by far the least pain for the most gain.

  46. Hecky,

    No ripping from me regarding your search tools. I’m in the Madison building’s Public Search Room using the PTO’s East and it’s head and shoulders better than the Thompson and Questel equivalents. Those products are getting better though. A telling fact is that the Thompson and Questel in-house searchers (we all know each other in our small world) use East for their patent based searching even with free access to their own products.

    For NPL, the examiners access is phenomonal for one reason: unlimited access at zero cost. On the outside we constantly have to balance cost vs. effectiveness. That is one of the reasons searchers freaked out at the NPL requirements of the Accelerated Examinations. To search all those databases for real was a multi-thousand dollar endeavor.

    Back to East, the speed of access and effectiveness of this engine is phenomonal in the hands of an experienced user. But Hecky can tell you, it’s not user friendly, even for examiners! As a speed example, a front page picture search of 3 or 4 thousand patents can easily be done in a few hours. You can flip at 0.5 seconds per patent with no delay.

    DIP

  47. Mark, this art that the EPO is finding; is it stuff which the USPTO ought reasonably to have found. For example, is it a WO, or an EP-A with a US family member? Or (worse) is it even a publication of the USPTO, which the EPO found and the UPSPTO didn’t. No wonder so many of those using the USPTO as PCT RO opt for the EPO as ISA.

  48. This is an excellent question. As a patent practitionner who handles both US and foreign sides of the application process for clients, my overall impression is that EPO searchers tend more often than not to find more pertinent prior art than the USPTO. It is frustrating to have a US patent granted, then to have dead-on prior art cited in the corresponding European case. This happens even though prior to writing our applications we conduct our own searches as well, and cite the results to the PTO. What tools are the EPO exmainers using that gives them better results?

  49. Is there anybody out there with a foot in both camps. How can EPO Exrs with no experience of US search tools say that their tools are better? And vice versa. Outside the patent literature, all use the same tools, but within the patent literature, who’s experienced in BOTH US and IPC/EPO?

  50. I’m an examiner.

    I’m not sure what public searchers use. We have EAST (which I use myself) and WEST, both of which allow text searching of US patents, US Pre-grant publications, WO documents, and other foreign patents. These are pretty good, especially as they can be field-specific (i.e., search for keywords X and Y, within Z letters of each other, limited to the claims).

    I am in the biotech area and the sequence search tools we have are great. A single query will search all major public databases (GenBank, EMBL, UniProt), as well as all patent databases (issued US patents, pre-grant publications, and GENESEQ, which is a database that includes WO publications). We can also query unpublished US sequences (every sequence ever submitted since about 1990 that is over 10 nucleotides or 3 amino acids is in that database).

    Personally, I also use Dialog (some examiners prefer STN). This allows for search of multiple non-patent databases (Medline, conference abstracts, news reports) simultaneously. I often find my best prior art, including the inventors own disclosures which for some reason don’t appear on an IDS, with these searches.

    I’ll leave it to others to either rip me to shreds on the inadequecy of my own searches, the office’s failure to use database X, Y, or Z. However in my opinion we have great search tools. I realize this is technology-specific — what works for biotech may well be entirely inadequate for computer chips.

Comments are closed.