How to read a patent in 60 seconds

Posted: September 6th, 2010 | Author: | Filed under: Startups | 43 Comments »

Someone’s claiming you’re infringing their patent!  You need to figure out how bad the situation is.

Or… your boss has asked you to take a look at an old patent you wrote, and see if someone else is infringing it.  And you have no recollection whatsoever, because it takes an average of nearly four years for a patent to be examined, so by the time anything interesting happens you’ve forgotten all about it.

Or… you’re just reading one of eighty hojillion Slashdot stories that use the title of the patent to write a “guess what obvious thing got patented” story, but you’re smart enough to know that the title isn’t actually the invention, and are curious what the real dirt’s about.

It can take hours or days to fully evaluate a patent.  When time’s short, here’s the quick and dirty way to figure out what the patent covers, usually in under a minute.

Step 1: Skip the title

The title of the patent can be just about as general as the author wants; for example, here is the the guy who patented the Tool.  It often describes the thing being improved on – not the new invention.  A patent titled “Virtual Desktop Manager” does not actually patent virtual desktops; it covers a particular set of features of a specific virtual desktop management implementation.

Step 2: Skip the drawings

Edward Van Halen PatentPatent drawings are mostly similar to high school notebook doodles except that they cost $5,000.  They’re generally impossible to read and only indirectly have a bearing on the enforceability of the patent.  The occasional exception exists: the incredibly edifying flowchart. The drawings that look like a giant gummy bear (because the invention is, actually, a giant gummy bear).  And sometimes… sometimes, a picture is simply worth a thousand words. Particularly when that picture depicts the inventor, one Mr. Edward L. Van Halen, demonstrating proper use of his invention.

Step 3: Skip the abstract

In other fields, the abstract is your best friend: a short, direct summary of the major points of a paper.  Patent abstracts are at best meandering and hard to read, and at worst deliberately misleading (so you think you’re in the clear, do whatever you planned to do, and then get sued anyway because the abstract has no bearing on the enforceability of the patent).

Step 4: Skip the specification

Now we’re getting to the meat of the patent!  And also skipping it.  You don’t care about the background, or the field. You don’t much care about the related art.  The brief summary of the invention doesn’t tell you what’s important; the description of the drawings is generally incomprehensible (unless it’s Edward L.). And the detailed description will send you catapulting in to catalepsy, while simultaneously not separating what’s actually novel and invented from the stuff that everyone knows already.

Step 5: Find the independent claims, and read them

The claims are the only part of the patent that have any actual legal enforceability.  While they’re still a pain to read, they’re forced to be one sentence so at least they’re relatively short (modulo the occasional run-on sentences half a page long).  They can be wicked difficult to parse in detail, but a skim will get you pointed in the right direction. This page also offers a decent primer.

Step 6: Back to skipping – toss the dependent claims

Any claim that starts with “The _____ of claim _____” is essentially a refinement or detail with narrower scope than the parent claim – if you infringe the baby, you’ll infringe the daddy too.  Skipadoodle.

And that’s it!

Getting sucked in to a patent dispute is no good for any entrepreneur.  By the time it’s done, you may be able to recite 40 pages of patenteese by memory, and have learned your Markush from your Jeppson.  But if all you need is a quick summary, just cut directly to the independent claims.  You’ll be done in a minute.

Big thanks to Adam Philipp at Aeon Law (who I use and heartily recommend) for giving this article a sanity check.  Also huge thanks to Tom Huseby, who introduced me to this clever trick.

Bonus information: how the patent office reads your patent

Pretty much the same way, most of the time.  They read the independent claims, then reference the drawings, and then move to the specification if a term or concept is unclear.  If you’ve got more than a minute, you won’t do wrong by following their example.

(You might want to subscribe or follow me on Twitter so you don’t miss new articles)


  • http://www.digitalquarters.net Ben Elowitz

    Love it. You make it fun.

    –Ben Elowitz

  • http://www.digitalquarters.net Ben Elowitz

    Love it. You make it fun.

  • http://twitter.com/MaxKennerly Max Kennerly, Esq.

    If the boss says, “you’re not reading the whole thing!,” point him or her to this Federal Circuit (i.e., the court that hears all patent case appeals) opinion:

    http://scholar.google.com/scholar_case?case=2207195741320793153

    Section II(A) — the entirety of which should be read by anyone charged with reading a patent — begins:

    “It is a “bedrock principle” of patent law that “the claims of a patent define the invention to which the patentee is entitled the right to exclude.” Innova, 381 F.3d at 1115; see also Vitronics, 90 F.3d at 1582 (“we look to the words of the claims themselves … to define the scope of the patented invention”); Markman, 52 F.3d at 980 (“The written description part of the specification itself does not delimit the right to exclude. That is the function and purpose of claims.”).”

  • http://twitter.com/donpark Don Park

    This is great for reading patents but can the same can be done for writing? I mean skip all but the independent claims.

  • http://pulse.yahoo.com/_IDMRHPQZKV5RIOLYQ2ESGUD4QE Kevin

    You’ve described the approach I took when undertaking a number of infringement analyses a few years back. You start with a stack of paper 18″ high, and gradually work through the independent claims of each one. And eventually, you may end up with a handful that need further analysis.

    Thankfully, I don’t do this stuff in my current line of work

  • http://twitter.com/BacheTechnical Bache Technical

    Love it, although our Patent drawings do not look like high school doodle and cost a hell of a lot less than $5000!

  • http://twitter.com/BacheTechnical Bache Technical

    Love it. Although our Patent Drawings do not look like high school doodles and cost hell of a lot less than $5000!

  • http://twitter.com/bjh_ip Ben Hoyle

    An excellent guide.

    It is unbelievable how many times I repeat this advice and still the “eighty hojillion Slashdot stories” keep coming.

    If you have read the independent claims and still have no idea what’s going on I’d recommend going back to the drawings, it’s sometimes less of a slog to flick through these if time is short.

  • Anonymous

    If the patent office gives it roughly the same amount of attention, it’s no surprise there are so many infringement claims.

  • http://www.linkedin.com/in/johndmccarthy johnmccarthy

    Cool, you just singlehandedly justified a decrease in billable hours by all patent attorneys I will ever use in the future. “6 hours to read the whole thing? Really?” Thanks!

  • Anonymous

    LOL, that really does make a LOT of sense dude.

    http://www.anon-surf.at.tc

  • Anonima

    Not really. The patent is a deal between society and inventor. The inventor agrees to disclose his invention so that a skilled person can understand and reproduce it. That’s what the description and drawings are for. The inventor gets his protection (a kind of monopoly) for what it defines in the claims.

    For inventions easy to undestand, the claims can be self-explanatory but normally you will need to go to the descriptiona and the drawings if you want to _reproduce_ the invention.

    To know if you are infringing the patent, the claims are the place to look at, though, as the post explains in such a great way :-)

  • http://twitter.com/Marotte Emmanuel Marot

    Great Post: both useful and fun!

  • Charlie Sgc

    Do I need a patent lawyer to research current patents? Is there a website I can go to and practice this method and save the money on a lawyer?

  • Bryan

    Usually, it’s not at all possible to reproduce the invention from the patent. Because, usually an invention relies on other patents from the same and different inventors, and generally any patent worth having also relies on trade secrets which are far less obvious, far more innovative, and which will never be voluntarily disclosed. It is in this way that patent holders get the protection of patent law without every giving society the true benefit of disclosing the invention.

  • JJ

    “Someone’s claiming you’re infringing their patent! You need to figure out how bad the situation is.” Unfortunately (or fortunately) our lawyers will often tell you explicitly NOT to read the patent that is claiming infringement in case it goes further than that. That way when they ask you if you have knowledge of patent X you can say a straight up “No”, not “Yes, but…”.

  • http://twitter.com/Obdurodon Jeff Darcy

    Having been involved in the patent process a few times, I’ve often had to point out that it’s the claims rather than the description that really matter. The description often includes tons of boilerplate that’s shared between several related patent, and much of it might have practically nothing to do with the current one. The drawings and particularly flowcharts are often there because the patent office insists on them for certain classes of inventions (this is from a former patent examiner and current practicing patent attorney less than a month ago) and not because of any actual informative value.

    The one quibble I have is that sometimes the secondary claims do matter. Many patents are only novel in the sense of applying an old technique to a new domain, and that’s in the secondaries. The primary is often deliberately worded to be as broad as possible, and that will get your attention, but it’s almost like a decoy. You can still be liable for infringing the “baby” even if the “daddy” is considered too broad to be novel, and these “babies” are often the real goal of the patent so you’d better be familiar with their scope.

  • Bryan_reddick

    yeah, but where can I pick up one of these Eddie Van Halen Musical instrument supports?

  • http://iplawyerintraining.blogspot.com Yuliya Mathis

    Great post Dan. The other tip I find that really helpful is if you’re really having trouble reading the claims, it sometimes helps to read the Summary section of the specification. Because let’s be honest sometimes the claims read like they are written in Klingon. Patent drafters often including the “the embodiments,” i.e. the claims written in general plain English, in the Summary section. They usually start with “Another embodiment..” or “In one example…” or “in accordance with the present invention…” etc.

  • Alan

    Excellent guide. As a practicing US patent agent (patent prosecution (filing applications and working to get them issued), *not* litigation), I just want to point out that this approach is useful for some, but not all, circumstances (e.g., those identified in the introduction, particularly when considering infringement). I would not recommend this approach when reviewing patents in view of application drafting or responding to office actions. In those cases, for example, it is important to review and comprehend more than just the claims.

  • Pingback: How to Read a Patent In 60 Seconds | Dan Shapiro | Voices | AllThingsD

  • Peter Zatloukal

    Hilarious!! That’s exactly how I do it :)

  • Bartmans

    Nice reading, but it only applies to granted patents. If you are reading a patent application (which there are by definition more of) you should also remember that such an application does not convey any – or only very limited – rights. Thus, why bother reading patent applications at all (unless you are interested in the technology)?

    Argument behind this: one of the most made mistakes is that patent applications are seen as granted patents by the public. In general the patent as garnted will have a (much) smaller scope than the application.

  • Pingback: links for 2010-09-16 « AB's reflections

  • Anonymous

    I hope people don’t take this too seriously. The words in the claims are often defined by the specification. For example the words computer, network, and message can mean entirely different things based on what was stated in the specification. Likewise, some of the words in the claims are entirely made-up, which requires one to read the specification. Don’t go firing your patent attorney yet!! : )

  • SJ

    If an invention relies on other patents from other inventors, isn’t that an example of the society deriving a benefit from patent law? Without patents, those other inventors would have had no reason to disclose their inventions.

    In my limited experience, patents are focused on the ingenious minutiae of an invention; you can make a new toilet seat mechanism but you’re only going to get a patent on the tiny spring-loaded piece at the center of the device that makes it all work because only that part meets the criteria for a patent (novel, inventive step forward, unobvious… there are others)

    Also, my experience is Canadian so make of it what you will!

  • http://www.danshapiro.com/blog Dan Shapiro

    Congratulations on entering practice! To your second paragraph: I realize that’s a common tactic in applications. But if the patent has been granted, and the primary claim is “as broad as possible”, I believe that you’re bound by the primary claim, no?

  • http://www.danshapiro.com/blog Dan Shapiro

    Agreed, and hopefully I made it clear that this is a shortcut that is useful but by no means exhaustive.

  • http://www.danshapiro.com/blog Dan Shapiro

    I like patents.google.com.

  • StevenP

    Not sure I concur with this. “Usually” it is not possible to reproduce the invention? Really? I’ve drafted and reviewed innumerable patents, and I don’t see this as being true at all.

    First, if you are referring to the issue that practicing a claimed invention may infringe another patent, that is an entirely different issue. (Reminder: patents do NOT give an owner/licensee the right to practice an invention, but only to EXCLUDE others from practicing it.)

    Second, if you are further suggesting that an undisclosed trade secret would be necessary to practice the claimed invention, then the written description is inadequate and the patent should not issue for lack of enablement. If the trade secret is needed to make the invention work in the best way, then the patentee failed to disclose the best mode and again the patent should not issue. (This may also raise inequitable conduct issues that could render the patent unenforceable.)

    [I'm sure others may likely add to this or correct a few things in my statements above!]

  • StevenP

    Not sure I concur with this. “Usually” it is not possible to reproduce the invention? Really? I’ve drafted and reviewed innumerable patents, and I don’t see this as being true at all.

    First, if you are referring to the issue that practicing a claimed invention may infringe another patent, that is an entirely different issue. (Reminder: patents do NOT give an owner/licensee the right to practice an invention, but only to EXCLUDE others from practicing it.)

    Second, if you are further suggesting that an undisclosed trade secret would be necessary to practice the claimed invention, then the written description is inadequate and the patent should not issue for lack of enablement. If the trade secret is needed to make the invention work in the best way, then the patentee failed to disclose the best mode and again the patent should not issue. (This may also raise inequitable conduct issues that could render the patent unenforceable.)

    [I'm sure others may likely add to this or correct a few things in my statements above!]

  • http://twitter.com/mypatents Adam Philipp

    What Jeff was pointing out is that the broader claims might fail more easily, but you could still end up infringing the narrower dependent claims.

    So a corollary might be, treat a dependent claims like and independent claim if you know its parent independent claim is invalid (e.g., not new, obvious, or the like).

  • Graf

    Can u help me please? “Notation and Nomenclature” what does this section means?

  • Freud Flintstone

    As a Patent Examiner, I can say that this is true.

    But, other than that, I find little to criticize in the OP. Dependent claims further narrow the independent claim, so if you’re looking for infringement you need look no further than the independent claims.

    Just remember, applicant can act as their own lexicographer, so unless you’re certain that the terms in the independent claim(s) are absolutely clear to anyone, check the spec to be sure they’re not saying one thing but meaning another.

  • http://www.hesido.com/ hesido

    1) A method of reading patents

    2) The method of reading patents in Claim 1, wherein the reader skips certain parts to understand it under a minute.

  • publiclobbyist

    It’s nothing but a priesthood of patent attorneys coupled with big industry and
    gov’t out to make a lot of money. So, the USPTO wants help in
    invalidating patents? They need US to protect their integrity? Pay me, effers.

    I say, issue automatic patents ($100 fee) that confirm the filing date
    and let them duke it out in court when an infringement comes
    up–something that has to be done regardless of the so-called monopoly
    guarantee. This is a win-win A) the patent ppl still have a job B) the
    small guy can go ahead with “infringement” confident that he can
    invalidate the big guys’s claim (using this article’s premise). The only
    loser is the Fed who won’t be collecting the 1,000’s of dollars they
    now suck out of the poor inventor trying to make a go of it.

  • Pingback: Patents and You | Re-Learning Efficiency

  • Pingback: Dan Shapiro » How to read a patent in 60 seconds | Hornet Dear Bernard

  • Michael Sander

    This is definitely a good primer on reading a patent. All good patent lawyers skip right to the claims when doing an initial review. However, there is one other piece of information that is almost (but not quite) as important as the claims: the priority date of the patent. The priority date of the patent is what determines what constitutes prior art, anything published after the priority date cannot be used to invalidate the patent. Determining the priority date is not always straightforward, but it is often the same as the filing date of the patent. The filing date of the patent is printed on the front of the patent in the left column about 1/3 down the page. Do not confuse it with the patent date, which is the big date on the top right. Even though the filing date is in smaller font, for finding invalidating prior art, the filing date is much more important than the patent date.

  • Denis Frolov

    Dear Dan,

    I am using the cooments to contact with you because I haven’t succeded in finding your email.

    I am writing regarding this article of yours — http://www.danshapiro.com/blog/2010/09/how-to-read-a-patent-in-60-second/.

    Do you think we could translate this article to Russian and publish on our corporate blog (http://habrahabr.ru/company/abbyy/) — of course, with the name of the author clearly indicated and a link to the original text?

    We are a language software development company and our developers will certainly appreciate this great article.

    Thank you for your time.

    Looking forward to hearing from you.

    Best regards,
    Denis Frolov
    ABBYY Language Services

  • Pingback: Dan Shapiro » How to read a patent in 60 seconds | HTML5 Based Design

  • BoulderPatentGuy

    Hmm. IMO, you need to read the spec to understand the claims. But hey, what do I know. I’m just a patent attorney. Instead, you should just follow the advice of a blogger.

  • Luck Of Fire

    patents are such a pain to read