Anticipated by the Internet

Doesn’t anyone think that debates over “combinations of references” are outdated? I usually only use one reference — Google.  Occasionally I combine Google with a directed PTO search — although even that will likely become more rare as Google continues to digest and index the history of written word. We are information rich, and arguments that exclude certain bits of knowledge from “prior art” based on inaccessibility do not make sense when that information was freely available online and indexed by Google. 

On another note, in the current regime, where the Internet is not a single reference, couldn’t a showing that two references would have both been retrieved with a single Google search serve as evidence of a motivation to combine the two?

Of course, these teasers suggest a new business model: Archiving of Google search results for later use during litigation.

23 thoughts on “Anticipated by the Internet

  1. This discussion is bringing up some very interesting points.
    I disagree with the statement that “The information disclosure regime at the USPTO is increasingly becoming obsolete in view of Google, the USPTO search facilities, esp@cenet, Delphion and Derwent, to name just a few.”
    In fact, I would say the opposite is true. With the exponential proliferation of information, including the use of such tools, it is even MORE important for applicants to disclose the (closest) prior art.

    With respect to using Google search terms to demonstrate a motivation to combine, I do not see why this should be possible. Google acts as a pseudo-document classification and retrieval tool but, in my opinion, does not serve, in and of itself, as a source document or disclosure. Whereas a web page to which it links/points may be.

  2. mr. crouch:
    very good post. i can see no rational reason that “the prior art” for purposes of a 102 rejection has to be limited to a single reference. But I had never thought of this until your post.


  3. Even assuming that a Google search could provide a motivation to combine, you’d have to provide clear and convincing evidence that the particular search would have produced the same result at the time the invention was made. Good luck trying to get Google to help you out with that task! Moreover, wouldn’t you also need clear and convincing evidence of a teaching, suggestion or motivation to perform the particular search?

    A better place to look for a teaching, suggestion or motivation to combine is the vast piles of office actions at the USPTO. If you’re having trouble finding a teaching, suggestion or motivation to combine, why not see if an examiner has already combined the references in question in an office action that predates the priority date?

  4. Gee, I’m really glad to see Dennis validate my preferred method of prior art searching!
    When I get very few or zero results, I do print these out with a date to document.

    Also, is indexed by Google, which usually hits on issued patents, thus leaving the PTO site useful only for published applications.

    I also agree that learning to formulate search terms to get at the core subject matter in a few seconds is a key skill that many have not acquired.

  5. “At some point soon (if not already), we have to be able to say that being of ordinary skill in almost any art includes at least rudimentary computer and web-surfing skills. ”

    One would think!

    I still see at least an email/month here at the firm from partners asking (usually poorly worded) questions which are easily addressed by 15 seconds of Google searching.

  6. Dennis, My question was really (rhetorical) in response to Pat’s hypothetical. My opinion is that the scientists google search itself is not a public disclosure. As to the question whether a phosita could have done a particular search, I think the impermissible hindsight arguments are strong.

  7. The Google algorithm is a trade secret so even if two technologies appear in a search result, the motivation remains unknown.

  8. The idea that a Google search can return a single reference that already combines two elements of a claim (such as the fan spoon example above) is helpful to locate possible prior art. However, the fact that Google will find two separate documents in a single search is not, IMHO, evidence in itself of a motivation to combine.

    In science, often 90% or more of the challenge is asking the right question. (In theory, every experiment in a lab is designed around trying to find an answer to a question. My background is biochemistry, so my perspective is biased.) It is a source of great frustration to me when an examiner starts from the point of asking the right questions and states that the last 10% is “obvious.” In many cases, once the correct question is asked, the rest is easy. The really difficult and inventive part is figuring out what question to ask. Polymerase chain reaction (PCR) is rather obvious, once it has been explained, but was worth a Nobel prize as perhaps the biggest advance in biotech in the past 50 years.

    Thus, if you start a Google search from a point after the correct question has been asked, the last little step may appear obvious. As said above, this is using hindsight.

  9. How can I find whether someone has used a particular google search? Is this even possible (assuming I’m a plan old member of the public and not a government actor, etc.)???


  10. Dennis:

    What do you mean by “Of course, these teasers suggest a new business model: Archiving of Google search results for later use during litigation.”?

    E.g., if a scientist does a google search to see if drug “x” to treat disease “y” is novel and he finds it is, are you suggesting his google search counts as a public disclosure against his later filed patent application claiming drug “x” to treat disease “y”?

    Or are you referring to something else? Either way, I am curious as to what people think the answer is to my hypothetical.


  11. At some point soon (if not already), we have to be able to say that being of ordinary skill in almost any art includes at least rudimentary computer and web-surfing skills.

  12. I would think that the results of a Google search do nothing to indicate a motivation to combine. Some of the results may even teach away from the combination, thus bolstering non-obviousness.

    I don’t know how Google does its indexing, but wouldn’t the algorithm that it uses need to use the same logic as one of skill in the art? Or is it enough that one of skill in the art can simply type something into Google to make assocaitions between certain technologies?

  13. Google is an advertising medium. It serves up the results that, often, someone paid to get posted at the top of the page.
    If you’re thinking of using a search engine for something as important as 102 or 103 information, then that deserves use of a technical or fee-to-business search service.
    Also, a Google, or like, result is hard to date-stamp – what’s the date of EACH part of the information listed vs. any date referred to in the result, or the date of the posting.

  14. Google is great for uncovering already existing business methods. Helps point out that throwing together a quick web site that does something that is old and well known is not patentable subject matter.

  15. Dennis – this is an interesting thought, but likely no more valid than saying (in the old days) that because two patents were both sitting in the shoes in the public search room and could both be found that this would make it obvious to combine them. Although this never stopped many examiners from trying.

  16. Dennis, I agree with most all your points. Keep up the good work, you are a huge public benefactor & great educator-at-large.

  17. The information disclosure regime at the USPTO is increasingly becoming obsolete in view of Google, the USPTO search facilities, esp@cenet, Delphion and Derwent, to name just a few. Whenever you have a new application to file (except perhaps in the chemical arts) information retrieval and rapid structured searching are becoming increasingly available.

    But I don’t think that makes references inherently obvious to combine. The key question is whether the skilled person would know in advance that there is any advantage that would flow from combining them.

    In truth there are three questions:

    (1) Was it obvious to the skilled person that the references can be combined?
    (2) Was it obvious how to combine them?
    (3) Was it obvious that when they were combined there would be an advantage flowing from the combination (which might be the advantage discovered by the patentee or might be some other advantage predictable from the references)?

  18. **Doesn’t anyone think that debates over “combinations of references” are outdated? I usually only use one reference — Google.**

    I think you have mischaracterized Google as a reference. Google (or at least its search results) is not a reference. Google is a just a fancy version of a card catalogue (As a side note: I wonder if card catalogue can still be found in libraries?) — this assumes, of course, one is not citing Google as art against related products, methods, etc. (e.g., search engines).

    You want anticipatory art from a single reference?… just find a good technical dictionary for the particular field of interest. I have several versions of McGraw-Hill’s “Dictionary of Scientific and Technical Terms.” I’m willing to bet that you could probably find many (if not all) of the recited limitations for most claims in that dictionary. However, 3M v. J&J Orthopedics, 976 F.2d 1559, 24 USPQ2d 1321 (Fed. Cir. 1992) addresses that issue. The argument is basically that guessing, testing, speculation, or “picking and choosing” from a broad disclosure does not constitute an identical description of a claimed invention within the meaning of 35 U.S.C. § 102.

    I agree with the other poster that the number of results that can be returned from Google (thousands, hundreds of thousands, millions, etc.), depending upon the search terms, would tend to make the TSM argument less persuasive.

    I’ll go back to the dictionary definition of obvious that I really like, which is “easily discovered.” If one having ordinary skill in the art at the time of the invention would have to look through more than 20? 50? 200? search results, then perhaps the invention would not have been “easily discovered.”

  19. Novice and bad patent examiners already do “keyword” rejections on a daily basis, albeit using east and west instead of google.

  20. It is quite common for Google searches to produce in excess of 1 million results. Mr. Wright’s “fan spoon” search results in 4,620,000 pages. Are you saying there is a motivation to combine any two of these 4.6 million pages?

  21. Dennis, I don’t think that two references being found by a Google search qualifies as a “motivation to combine.” It all depends on what the PTO’s search string is. It may be improper hindsight.

    For example, if I were to invent a spoon with a built in mini-fan to cool my soup automatically, what would the search string be? If it was “fan spoon,” then I would argue that it’s a hindsight motivation problem. If it was “spoon cool,” then that’s moving a step away from hindsight. And if the search was “cool soup,” that would be a much better search string, and arguably could produce a valid motivation to combine.

    Am I making sense? -Colin

Comments are closed.