Claims to Intermediated Financial Settlement Not Patent Eligible; Software Patents Still Alive

Alive_v_CLSBankThe U.S. Supreme Court issued its decision in Alice v. CLS Bank (June 19, 2014) touching on the patent-eligibility of software implemented inventions. The decision continues to allow patent protection for software innovation and technological solutions implemented with software. Therefore all software patents are not dead in the wake of this opinion, in fact, most survive under the USPTO’s interpretation.  However, this decision limits or eliminates, depending on the circumstances, the ability of one to obtain a patent covering a fundamental economic practice long prevalent in commerce that is merely implemented in a generic computer using generic functions without more.

The patents at issue in Alice were directed to using a computer to mitigate financial settlement risk, i.e., the risk that only one party to an agreed-upon financial exchange will satisfy its obligation. The court found that all of  the claims at issue are drawn to “the abstract idea of intermediated settlement, and that merely requiring generic computer implementation fail[ed] to transform the abstract idea into a patent-eligible invention.”

Mayo Framework
The court followed the two step framework presented in Mayo Collaborative Services v. Prometheus Laboratories, Inc., 566 U. S. ___ (2012), to determine whether the claims at issue were directed to an abstract idea. First, the Court determines whether the claims at issue are directed to a patent-ineligible concept (e.g. laws of nature, natural phenomena, or abstract ideas). If so, the second step is to determine whether there are additional elements in the claims that transform the nature of the claims into a patent-eligible application of the law of nature, natural phenomena, or abstract idea at issue.

Software is Still Patentable
The court asserted that the claims at issue in Alice did not improve the function of the computer itself, nor effect an improvement in any other technology or technical field. So at least software that does either of those things should be patent eligible. Further, the court acknowledged that it must tread carefully so that the rules about abstract ideas or other excluded subject matter do not swallow all of patent law. This is true because at some level all inventions use, reflect, rest upon, or apply laws of nature, natural phenomena, or abstract ideas. The court continued to recognize that the application of a law of nature, natural phenomena, or abstract idea to a new and useful end remains patent eligible.

Method Claims
The Court first looked at the method claims in Alice. In the first part of the Mayo framework, the court found that the method claims were directed to the abstract idea of intermediated settlement. The Court cited a 1896 publication discussing the use of a clearing-house as an intermediary to reduce settlement risk.

In the second part of the Mayo framework–whether there are additional elements–the court found that “the method claims, which merely require a generic computer implementation, fail to transform that abstract idea into a patent eligible invention.” The Court stated that the “wholly generic computer implementation is not generally the sort of ‘additional featur[e]’ that provides any practical assurance that the process is more than a drafting effort designed to monopolize the [abstract] idea itself.” The Court found that the method claims at issue did not do more than instruct the practitioner to implement the abstract idea of intermediated settlement on a generic computer.

The Court noted that the method claims did not improve the function of the computer itself, nor effect an improvement in any other technology or technical field. The Court found that all the computer steps were “well-understood, routine, conventional activities previously known to the industry.” Further, the Court found that the ordered combination of the steps using a generic computer added nothing that was not already present when the steps were considered separately.

The Court found that the computer readable medium claims failed for the same reasons as the method claims and because Alice conceded that the computer readable medium claims raise and fall with the method claims.

System Claims
The systems claims fared no better than the method and medium claims. The Court discounted arguments that the system claims were not abstract because they included specific hardware configured to perform specific computerized functions (“… a data storage unit having stored therein information….”, “a computer, coupled to said data storage unit, that is configured to…”, “a communications controller”, etc). The Court found these items were “purely functional and generic.” The Court stated that “Nearly every computer will include a ‘communications controller’ and a ‘data storage unit’ capable of performing the basic calculation, storage, and transmission functions…” Therefore, the Court found that the hardware recited by the system claims did not offer “a meaningful limitation beyond generally linking the use of the method to a particular technological environment.”

The Court found that the system claims were no different from the method claims in substance. The court stated, “The method claims recite the abstract idea implemented on a generic computer; the system claims recite a handful of generic computer components configured to implement the same idea.” The court stated that to interpret the method claims as patent-ineligible, but then interpret the system claims as patent-eligible would go against the Court’s previous precedent that section 101 should not be interpreted in ways that make patent eligibility depend on the patent draftsman’s art (e.g. the manner in which the claims are drafted).

Analysis: Generic Computer
Some have argued that  Court has imported the question of novelty, usually reserved for consideration under section 102 and 103, into the 101 patent-eligibility analysis under the second question: “whether there are additional elements in the claims that transform the nature of the claims into a patent-eligible application.”

The argument is that under the second step we must ask: are the additional components of the claim beyond the abstract idea old and well known? Whether something is old or well known if usually a questions determined under sections 102 and 103 by looking to see if the claimed subject matter is disclosed in the prior art or legally obvious in view of the prior art.

At least one commentator suggests that the court did not import a novelty test into the section 101 analysis, but instead presented a question of whether the claim does significantly more than state a fundamental principle connected with an instruction to “apply it.” The court said, “Stating an abstract idea while adding the words ‘apply it”’ is not enough for patent eligibility. Nor is limiting the use of an abstract idea “‘to a particular technological environment.” The Court said, “Stating an abstract idea while adding the words ‘apply it with a computer’ simply combines those two steps, with the same deficient result.” Therefore the reference to a generic computer not being enough is simply a recognition that Alice’s claim did no more than apply a old principle in a generic computer.

While the court provides that a generic computer is not enough to transform an abstract principle into patent-eligible subject matter, another commentator  notes that the court does not provide much guidance on what is enough.

Not Abstract: Technological Solutions
As yet another commentator noted, the Court provided in Diamond v. Diehr, 450 U.S. 175 (1981) that a process of curing rubber using a well known mathematical equation was patent-eligible not because it involved a computer, but because the claimed process improved an existing technological process. Therefore, inventions directed at improving a technological problem should not be found to be abstract under section 101.

The question arises, what is a technological problem? At least the areas of applied economics and finance, such as intermediated settlement, which provide a non-technical method of organizing human activity are excluded from the meaning of technological.

USPTO Interprets Alice
On June 25, 2014, the USPTO issued a memo to its patent examiners instructing how to apply the ruling in Alice to pending patent applications. The memo instructs examiners to follow the two step Mayo analysis. The memo provides these examples of abstract ideas referenced in Alice under the first step of Mayo:

  • Fundamental economic practices
  • Certain methods of organizing human activities
  • An idea itself
  • Mathematical relationships/formulas

The memo instructs that inventions are not patent ineligible simply because the invention involves an abstract concept. The application of an abstract idea in a meaningful way is patent eligible.

If any of the above abstract ideas are found in the claims, the Examiner must proceed to the second step of Mayo to determine whether the claim is nonetheless patent eligible because it has additional elements that amount to significantly more than the abstract idea itself. Examples of limitations that may be enough to qualify as significantly more when recited in a claim with an abstract idea include:

  • Improvements to another technology or technical field
  • Improvements in the functioning of the computer itself
  • Meaningful limitations beyond generally linking the use of an abstract idea to a particular technological environment

The USPTO noted that limitations that are not enough to qualify as significantly more include:

  • Adding the words “apply it” (or an equivalent) with an abstract idea, or mere instructions to implement an abstract idea on a computer
  • Requiring no more than a generic computer to perform generic computer functions that are well-understood, routine, and conventional activities previously known to the industry.

Conclusion
While patent protection for software directed to fundamental principles of economics and finance without more is limited, patent eligibility for protecting software innovation and technological solutions implemented with software is still alive after Alice.

Start with a provisional or a non-provisional patent application?

Fork in the roadThis post is designed to explain the difference between a U.S. Provisional Patent Application and a U.S. Non-Provisional Patent Application. In addition this document will explain circumstances in which one might be preferred over the other.

You may think of the provisional or non-provisional patent application as two routes to obtaining a patent. The non-provisional route providing a one step start to the patent process and the provisional route providing a two step start to the patent process. Either route, when properly carried out, can result in a patent. The two-step provisional process will take about one year longer to receive a patent than the one step non-provisional route.

Non-Provisional Route
A U.S. non-provisional patent application is a patent application that when properly filed with the U.S. Patent and Trademark Office is placed in a queue, examined by a Patent Examiner. If the claims of the non-provisional patent application are ultimately determined by the Examiner to meet the legal requirements, the non-provisional patent application may issue into a U.S. Patent. Each non-provisional patent application should contain a title, a background of the invention, a summary of the invention, a detailed description of the invention, one or more claims, and drawings. Also an oath or declaration complying with the applicable rules is also required. A fee of about 530 dollars is also required (assuming you qualify as a small entity).

Provisional + Non-Provisional within 1yr Route
You can think of the Provisional route as providing a two-step start to the patent process whereas the non-provisional application route is a one-step start.

The first step in the Provisional Route is to file a provisional patent application. The inventor’s oath or declaration and claims are not required for a provisional application. The provisional application filing fees are lower (about $125 for a small entity). The provisional patent application must include a description of your invention.

To ensure that resulting patent claims receive the benefit of the provisional patent application filing date, we recommend that a provisional patent application be drafted in the same manner as a non-provisional application, except that claims need not be included in the provisional application. In addition we often draft at least some claims for a provisional application. If an application fails to provide an adequate description of the invention, this may negatively impact the inventors ability to obtain a valid patent.

The Patent Office will not examine a provisional application and will not issue a patent directly from a provisional application. Therefore, there is no such thing as a “provisional patent” only a “provisional patent application.”

So why would you proceed with a provisional application if it does not directly result in a patent? A patent can result from a non-provisional patent application filed within one year of the filing date of the provisional application, where the non-provisional claims the benefit of the provisional application. The filing of the non-provisional within one year is the second step of the two step process.

In order to retain the benefit of the filing of a provisional patent application, a non-provisional patent application that claims the benefit of the provisional application must be filed within one year of the filing date of the provisional application. If a non-provisional patent application claiming the benefit of the provisional application is not filed within one year of the filing date of the provisional application, then the benefit of the provisional application filing date will be lost.

The Provisional Route Takes Longer
The two-step provisional route takes longer than the 1 step non-provisional route. This is because the provisional application is never put in a queue at the patent office to be examined, only non-provisional applications are placed in the queue and examined.

Lets see how this works in the following example which compares the 1-step and the 2-step routes.

Non-Provisional – 1 Step Route
John’s attorney files a non-provisional patent application on January 1, 2012. The USPTO then places the application in the queue to be examined by a Patent Examiner. The average wait time until an Examiner examines a patent application is about 24 months. Therefore around January 1, 2014 an Examiner will being the examination process. Thereafter the Examiner will negotiate with John’s patent attorney, during what’s called patent prosecution. If the Examiner and John’s attorney come to an agreement on the scope of the claims in the patent application, the Examiner will allow the application to issue into a patent. The average wait time from the date of filing to the issuance of a patent is about 3 years. Therefore, if John’s case is average, he will be issued a patent around January 1, 2015 (3 years after the filing date). The resulting patent will have an effective filing date of January 1, 2012.

Provisional — 2 Step Route
Now lets look at a scenario where John first files a provisional patent application and then files a non-provisional application within one year. John’s attorney files a provisional patent application on January 1, 2012. The USPTO sends John’s attorney a filing receipt for the filing of the provisional but the USPTO does not place the provisional application in the queue to be examined by a Examiner. Remember provisional patent applications are not examined. John’s attorney places a reminder on his calendar to file a non-provisional patent application within one year of the filing of the provisional application. Before January 1, 2013, John’s attorney asks John if John wants to continue to pursue a patent on John’s invention. If John says Yes, John’s attorney prepares and files a non-provisional patent application claiming the benefit of the provisional application on or before January 1, 2013. Now the non-provisional application will be handled as described above in the one-step, non-provisional route, except the process starts one year later.

The USPTO now puts John’s non-provisional application in the queue to be examined by a Patent Examiner. In about two years, around January 1, 2015, an Examiner will being the examination process. Thereafter the Examiner will negotiate with John’s patent attorney during patent prosecution. If the Examiner and John’s attorney come to an agreement on the scope of the claims in the patent application, the Examiner will allow the application to issue into a patent. If John’s case is average, he will be issued a patent around January 1, 2016 (3 years after the filing date of the non-provisional application). The resulting patent will be giving the benefit of the filing date of the provisional application and therefore will have an effective filing date of January 1, 2012.

Therefore, as you can see each route results in a patent that has an effective filing date of January 1, 2012. However, the one step route results in a patent issuing a year earlier than the two step route.

Why Go the Provisional, 2 Step Route?
There are at least two reasons why one might choose the two-step provisional route.

Disclosure Needed Before Complete Development
Another reason the two step approach might be chosen is when disclosure is needed before the invention is completely developed. Often new products are developed on a deadline so that the product can be shown to a customer, presented at a trade show, or otherwise made available to the public. U.S. Patent law encourages the filing of a patent before certain events, such as a public showing or an offer for sale of invention. Therefore, if you are on a deadline to make an offer for sale or a public showing, but you believe that you will further develop the invention or you will add features to the invention within one year, a provisional application can be drafted and filed on the invention before any showing so as to provide the most protected approach.

Then if further developments are generated within the year of the first provisional application, additional provisional applications can be filed, and then the non-Provisional application filed within one year of the first provisional can claim the benefit of each of the provisional applications. Also, the new developments can be added to the non-provisional application filed on or before the one year date. The new developments will only be accorded a filing date when they are provided in a patent application. This approach reduces overall patent costs when products are developed over time when intervening offers of the invention or showing are required.

Cost Spreading/Shifting
The first reason is to reduce the up-front cost of filing a patent application and to spread the patent application cost over a year period. Lets take an example of a simple invention. A very simple invention might generally cost $3500 in attorney’s fees to prepare a non-provisional application. The filing fee for a non-provisional application for a small entity is about $530. Therefore the total cost of filing the application in a 1-step non-provisional route is $4030.

Now lets look that the two step provisional application route for the same invention. Lets assume that it cost $1500 in attorney fees to prepare claims for the invention. Since claims are not required in a provisional application the total attorney’s fees for preparing a provisional application on that invention would be $2000. The filing fee for a provisional application for a small entity is about $125. Therefore the total cost to file the provisional application on that same invention would be $2,125, which is $1905 less than the cost to file the non-provisional application. Therefore the up-front cost to file a provisional is less than the up-front cost to file a non-provisional.

However, a non-provisional application claiming the benefit of the provisional application must be filed within one year of the filing date of the provisional. Assuming there is no change in the invention and therefore no change in the application content at the time of preparing the non-provisional. Claims would be drafted and added to the content of the provisional application. The cost of drafting the claims would be $1500 (the cost not incurred above when writing the provisional application). The government filing fee for the non-provisional is $530. Therefore, the cost of the second step, filing the non-provisional application, is $2030.

Under the two step process, the cost of the application is split over a year’s time: $2,125 for the provisional application + $2030 for the non-provisional application, which totals $4,155. As you can see the two step process cost an additional $125–the cost of the provisional application– which is not incurred in the one step non-provisional process.

Therefore, in this example for a very simple invention, you can see that for the addition of $125 to the overall cost of the application, $1,905 can be reduced from the initial application cost and shifted about one year down the road.

Furthermore, if events develop during the time between the filing of a provisional application and the deadline for filing the non-provisional application, such that you do not want to proceed with patenting your invention, then $1905 is saved. This might happen if you determine the cost of commercializing your invention is more than expected or the market did not respond as expected to your invention. The two-step route gives you a year to determine whether a patent on your invention is worth further pursuing from a business perspective while keeping the initial application cost to a minimum. The foregoing cost represent a hypothetical example. Actual cost will vary depending on the invention and may not conform to the above example.

Should I Write My Own Patent Application (DIY Patents)?

DIY_PatentsYou may wonder whether you should write and file your own patent application (e.g. a DIY (do-it-yourself) patent application) or whether you should have a patent attorney file directly a patent application that you wrote. This posted explains why we generally don’t just file the documents you give us as a provisional patent application without review. However, client written materials are usually a good place for us to start.

Since there are few requirements that must be met for the USPTO to accept a document as a provisional patent application, there are a large variety of disclosures in provisional patent applications. Since the USPTO will not examine a provisional patent application, they will not scrutinize the level of detail provided in any document that is filed as a provisional application. Given this, there is a wide range in the level of detail among provisional applications that are filed. This can range from a back-of-the napkin sketch of an invention filed as a provisional to a full application that you would see filed as a non-provisional application.

As a general rule we only file provisional applications that look like a non-provisional application, except that the provisional might have fewer claims. We generally will not take client-drafted documents and file them directly without review and revision to meet our non-provisional application standard for the reasons explained below. We think this approach best serves our client’s patent needs.

Priority Date
First, the reason for filing a patent application is to seek a resulting patent or at least to hold open the opportunity to obtain a patent on an invention. Patents are only granted on inventions that are new. Therefore, one function of the filing of an application is to establish a priority date (filing date) of the application. The priority date generally establishes your rights in the invention as of the priority date (with some exceptions). Therefore someone inventing or filing a patent application after your priority date should have inferior rights to you. In sum, the priority date (the filing date) of your provisional application gives you rights in your invention superior to later inventors of the same invention.

However, in order for your invention (or subfeatures or functions thereof) to be granted the priority date of your provisional application the invention must be enabled and adequately described. If a court later finds that the disclosure of your provisional application was inadequate, you could loose the benefit of the provisional application filing date. If priority is lost with respect to all or a subset of the features of your invention, it would be as if the provisional application was never filed on those features. In other words, to gain a benefit from the provisional application, the provisional application must be drafted properly and completely.

In order to obtain a patent, a non-provisional application claiming priority to the provisional must be filed within one year of the filing date of the provisional application. If information must be added to the subsequently filed non-provisional application because the provisional application did not contain sufficient detail, a question will arise as to whether that new information will receive the benefit of the provisional filing date.

Example
Why does this matter? Lets look at a hypothetical scenario. Assume that you file your provisional application on A+B on January 1, 2012. Then you consult with an attorney and the attorney notes that you left out some important details in the description of your idea. Lets call those left-out ideas “C.” So the attorney prepares a non-provisional application having A+B+C and claims priority to the provisional application. The non-provisional is filed on December 12, 2012. If +C would not be known to one skilled in the art by reading your provisional application, then A+B will get the benefit of the January 1, 2012 but A+B+C will not.

Assume that on October 1, 2012, another inventor “John” files an application on A+B+C (John did not learn of the invention from you, but independently invented). Therefore, it is possible that inventor john will have rights to the invention A+B+C over you because he was the first to disclose the A+B+C combination in the patent application (certain exceptions to this might apply). This may be true even if you intend your provisional application to cover A+B+C, but did not provide sufficient detail. What is the likelihood that “John” the inventor would file an intervening application? It is impossible to know. However, if you are proceeding forward with business plans on your invention in reliance on the provisional application, you want to be sure that you can actually rely on it.

Important Details
The wording in a patent application, provisional or non-provisional, is extremely important. Often times in patent disputes after a patent is granted, attorneys are wishing that just one additional word had been added to the corresponding patent application at the time of drafting that would make the case or thankful that that word is not there, depending on which client (the patent owner or the accused infringer) the attorney represents.

Too much and Not Enough
While it is not impossible for a client to draft a provisional patent application that meets the description and enablement requirements–and we have on occasion seen a well drafted application prepared by a client–most times we find that after reading a client-drafted application we would not feel comfortable opining that the client-drafted application would meet the requirements. Often the client-drafted application says both too much and not enough at the same time.

The client filed application might say too much by making unnecessary admissions that could limit the scope of protection afforded over the invention in a resulting patent. It might say too little by not describing the invention, its various parts, and its operation in sufficient detail. Further, it might say too little by not describing the invention broadly enough and thereby leaving on the table subject matter that could have been claimed.

Will a one page write up and a drawing or two be sufficient, if filed as a non-provisional, to secure your rights? It depends on the circumstances, but usually more detail is needed. There are online service providers that will assist you in filing your documents as a provisional application, however, we think this do-it-yourself process may provide clients a false sense of security, or at least creates a such a risk when client documents are not reviewed to sure they meet the legal description and enablement requirements.

A Good Starting Point
This is not to say that the client should not draft a detailed description of the invention and provide drawings. Such client-written materials are usually a great place for us to start drafting a patent application. Often times a detailed description of the invention provided by the client can speed up the patent application drafting process on our end and can reduce the overall costs of preparing a patent application. Therefore we encourage clients to provide us with detailed information about the invention. However, for the reasons explained above we will generally not file that information directly as a provisional application, but instead we will use it as a starting point to draft an application that broadly protects the client’s invention and meets our standards.

What is Patent Pending?

When a patent application is filed with the United States Patent and Trademark Office (USPTO) covering a product, service, or portion of a product or service, the owner has patent pending status in the United States regarding that invention. Basically patent pending status means that a patent application has been filed in the United States. Patent pending status continues until the corresponding patent application is abandoned or until a patent is granted.

Patent pending status does not mean that the patent applicant will necessarily obtain a patent. It only means that a patent is requested through the filing of a patent application. It is possible that the USPTO will find that the subject matter of the patent application does not meet the requirements to be granted a patent. On the other hand, a patent application corresponding to a claim of patent pending may indeed ultimately received a patent.

The patent pending notice is designed to put the public on notice that an application on some aspect of the corresponding product or service has been filed and might result in a patent. Therefore the patent pending notice may be effective in deterring competition during the pendency of the patent application. This may be particularly true if the start-up and production costs of producing and or selling the invention are high. As a competitor might not want to heavily invest in producing and selling a product when there is a chance the competitor will need to shutdown that production when a patent is granted. However, it is also possible that competitors will try to compete during the patent pending period.

How to Stop Copiers During Patent Pending?

copycopyA patent applicant does not receive full patent rights until the patent office grants a patent. This article explores your options when a competitor starts copying your invention while your patent application is pending but before your patent is granted.

A person or entity that files a patent application on an invention has patent pending status until a patent issues or the application is abandoned. However, patent pending means that the patent applicant requested that the government grant a patent. Patent pending does not mean that a patent will necessarily be granted. The Patent Office might determine that a patent should not be granted because the application does not meet the requirements for obtaining a patent, for example, the application may be denied if the invention is not new.

The rights that are provided for patent holders are not the same as those provided for patent applicants during the patent pending phase. This is because not every application will become a patent. Therefore, it would not be appropriate to allow patent applicants to stop others from making, using, selling, and importing the invention (e.g. patent rights) described in their patent application because it is possible the Patent Office might later determine a patent should not be granted.

What can be done about copiers during the patent pending stage?

Provisional Rights
There is a section of the law that provides that a patent applicant can get a reasonable royalty for sales of products or services (e.g. copiers or would be infringes) that are covered by a resulting patent. These rights are known as provisional rights (35 U.S.C. 154 – the use of the word “provisional” is unrelated to “provisional” as used to reference a provisional patent application and does not depend on whether or not a provisional application was filed).

The problem with provisional rights as that there are many requirements that must exists in order for a patent applicant to have provisional rights cover a copier or would be infringer. These requirements include (1) the royalties only begin after your application is published and do not begin on the date the application was filed, (2) the copier must have actual notice of your application, which means they must receive a copy of the published patent application or have otherwise read it, (3) the claims in your application must be substantially identical when the patent application is granted as they were when the application was published, and (4) you cannot seek back royalties under the provisional rights section until you actually obtain a patent.

The most problematic requirement is the substantial identical requirement. Changes are often made to the claims during the patent prosecution stage where your patent attorney negotiates with the patent examiner regarding the scope of the patent protection that will be allowed in a resulting patent. Due to the common nature of such changes, generally recovery of royalties under the provisional rights section is unusual.

Notice letter
Even though provisional rights might be rare, you may choose to have a letter sent to a copier to put them on actual notice of your published patent application by attaching a copy to the letter. The letter might request that the copier stop or might invite the copier to a licensing discussion. This may produce results, especially if the startup and design costs are significant for the invention at issue. The copier might not want to invest in those costs if it would be forced to shutdown when you obtain a patent. However, the copier may choose to ignore your letter. There is no requirement that the copier respond. The copier may conclude that you will not obtain a patent or if you do obtain a patent it will not cover the copiers goods/services. Further the copier might try to redesign the product/service to avoid the subject matter of the patent application after receiving your notice.

Speeding Up Application Processing
Once you notice an infringer you might decide to request that the Patent Office speed up the consideration of your patent application. However, sometimes you can benefit from a delay at the patent office where a competitor builds a market and then you are able to take over the market or obtain licensing fees once a patent is granted.

Other IP Claims
It is possible that you might be able to make other intellectual property claims depending on your circumstances, such as claims of breach of contract, copyright infringement, trade dress infringement, trade secret misappropriation, or other infringement infringement.

Conclusion
It can be frustrating to see competitors selling copies of your product in the marketplace while a patent application is pending but before an application is granted. Occasionally a notice letter to the copier will be effective. In other instances you may need to wait until a patent is granted to take action.

Photo credit to flickr user David Goehring under this creative commons license.

Inventor of Pay-Per-Click and Ad Auction Fails to Patent, Google Profits

InThePlex

We were ready to go public and we’re on fire, revenues going through the roof and all that, and were getting our IP portfolio together for the bankers, and everybody was like, ‘What patents do we have?’ And we didn’t have too many. … All we could do was patent everything else we could think of, a bunch of obscure things like the way we accept bids. These were silly patents, but the real patents would have been worth billions. … we learned our lessons about patent protection.”

said Bill Gross, founder of GoTo, who developed the concept of pay-per-click online advertising and real time auctions for pay-per-click ads before Google. These two concepts were used by Google in its search engine and made Google one of the most profitable businesses in history. GoTo failed to patent its inventions related to these concepts and potentially lost a lot of money as is explained below. This is the second post about Google based on information from Steven Levy’s  book In The Plex: How Google Thinks, Works, and Shapes Our Lives. The first post is here.

Google’s Auction Based on Ad System from GoTo
In 2000, Google’s venture capitalists were pressuring Google to make more money. So, Google founders, Larry Page and Sergey Brin, instructed Google employee Salar Kamangar to look into ways to make more money with the ad system. Eric Veach and Kamangar decided to use auctions to sell ads next to search results. And they got this concept from a search competitor called GoTo.

GoTo was founded by Bill Gross. Gross implemented an auction to determine the price that an advertiser would pay placement when a given keyword was searched. The higher the bid, compared to other bidders, the higher on the page your advertisement would appear in the search results.

The difference between GoTo’s implementation and how Google implemented the auction was ad placement. Gross’ implementation intermingled ads with organic search results. Many people disliked the intermingling, including Page and Brin. Google provided ads on the right and at a block at the top of the page, but not intermingled in the list of organic search results.

In addition, Gross implemented pay-per-click advertising. Before pay-per-click, most ads were sold on a pay-per-impression basis. An impression is when a person saw the ad. So regardless of whether a person clicked on the ad, the advertiser was charged when a person saw the ad.

GoTo Waits Too Long to Apply for a Patent & Is Barred
In 2000, “We were ready to go public and were on fire, revenues going through the roof and all that, and were getting our IP portfolio together for the bankers, and everybody was like, ‘What patents do we have?’ And we didn’t have too many,” said Gross. GoTo missed the one-year bar.

A patent application generally must be filed within one year of an offer for sale, a sale, or a public disclosure of the invention. This is referred to in the U.S. as the one year bar. At the time GoTo realized it needed to file patent applications on its core features of real-time auctions and its pay-per-click ad system, the system had already been disclosed for more than one year. Therefore, GoTo was barred from filing patent application(s) on those features and lost the ability to protect those core features with a patent(s). All GoTo could do “was patent everything else we could think of, a bunch of obscure things like the way we accept bids. These were silly patents, but the real patents would have been worth billions” Gross said.

The one-year bar in the U.S. can be problematic for independent inventors, startups, and small businesses because often you will want to determine whether there is a market for the invention before spending money on the patent process, but patent law encourages you to file a patent application before you make your invention public and not to rely on the one-year grace period (if you are interested in foreign protection, many foreign countries provide no grace period).  Here, we don’t know the reason why GoTo missed the one year bar, maybe they didn’t think about patenting, maybe they didn’t realize the value of the core technology within one year, or maybe they didn’t have the resources, etc. Whatever the reason, the one year bar is a common problem, not only for GoTo, but for inventors and companies generally. Patent law encourages early patenting. This post provides ideas for patent protection based on the available financial resources.

Google Turns Down Goto’s Requests for Merger; Goes Alone
Later in 2001 Gross suggested to Page and Brin that GoTo and Google merge. But according to In The Plex, “Kamangar successfully argued against any kind of acquisition, saying that Google could do it alone.”

However, when Google implemented its own auction it implemented it as a generalized second-price auction, which is different from the direct auction the GoTo used.

Further the Goto’s (now named Overture) method required bidding for each slot on the page. So if you wanted to bid for the 1st position and the second position you had to place two bids, one for the first and one for the second. In Google’s system, you could place one bid and receive the  position that corresponded to that bid at the end of the auction. Also, the Google system factored not only the bid price but quality of the ad to determine the location of the ad on the page.

Improving on Other’s Work
So while the GoTo implementation was different in some ways from the system that Google built, it is possible that if GoTo had filed patent applications on the core features, that the resulting patents could have covered Google’s system and given GoTo better leverage in merger negotiations with Google. Who knows how much they would actually be worth, but Gross said, as quoted above, they would have been worth billions.

Some might argue that Google implemented a better system, with the second price auction, and therefore Goto should not be rewarded. But, Google started with GoTo’s invention and then improved upon it. The patent system does reward inventors when others use the invention regardless of whether its the original or an improvement. See the basic wheel v. spoked wheel example for more on this concept.

Lessons about Patent Protection
Later Goto (then called Overture) sued Google on the “silly patents.” And Yahoo eventually acquired Overture. Google settled those suits, but for much less than if Goto had patents on its core technologies.

According to In The Plex, Bill Gross was positive when addressing the fact that his ideas involving pay per click and ad auctions had made billionaires at Google, but not at his company. Gross said, “I feel we won…There was a satisfaction in braking the code. We originally invested $200,000 in GoTo, and when we sold Overture, we made $200 million. That’s a pretty good return. And we learned our lessons about patent protection.”

What happens after the patent application is filed?

PatentApplicationFiledWhat happens after a patent application is filed depends on whether a provisional patent application or a non-provisional patent application was filed. If a non-provisional patent application is filed, once the patent office receives the complete patent application they will place your application in a queue to be examined by a Patent Examiner.

Non-Provisional Applications – Timing
Depending on the backlog of patent applications to be examined, it could take about 21 months, on average, for an Examiner to begin the examination process on your patent application. This time frame can be shortened if a priority examination is requested and granted. When the Patent Examiner picks up your application for examination, the examiner will do a prior art search (a search of the prior patents and patent applications and other sources of prior art) to determine whether your invention, as set forth in the claims of your patent application, is new and otherwise meets the requirements for patentability.

The Patent Prosecution (Negotiation) Stage
At least some claims are rejected (or another objection is raised) as a result of the first substantive examination in about 77 to 95 percent of all patent applications. When claim rejections or other objections are found by the examiner, the examiner will issue an office action explaining the reasons for the rejections and objections.

Many patents result from patent applications that received at least one rejection in an office action after the first substantive examination by the examiner. In most cases, the office action is not the end of the road, but rather the beginning of a negotiation (known as patent prosecution) between the patent attorney, on your behalf, and the Examiner at the USPTO as to the scope of the protection that will be granted over your invention in a resulting patent.

If you decide you would like to respond to the office action and proceed forward with the patent application, we will review the office action and advise you of the options for responding and the strategy corresponding to each option. We will endeavor, where possible, to explain the relative likelihood of success and cost with each option.

There are a number of ways to respond to an office action. One way is for the attorney to draft a written response to the office action which amends the wording of the claims and/or makes arguments on behalf of the patent applicant on why the application should be allowed to issue into a patent. Another means that can be employed is for the patent attorney to setup a telephone interview with the examiner were the attorney can discuss the issues raised by the Examiner in the Office Action and seek to place the patent application in condition for allowance.

After the first written response is filed, the Examiner, if convinced, may allow the patent to be granted. However, the Examiner may issue a second office action responding to the attorney’s written response. Where again, the attorney will advise the client regarding the options for responding in a manner similar to the first office action. The attorney may then prepare a written response or hold a telephone interview or both. The office action issuance and response can occur any number of times, if government fees are payed to continue the prosecution, depending on the client goals.

Sometimes patents are allowed at the first office action, after the first office action, after a one or more subsequent office actions, or after appeal. In certain circumstances, an appeal can be pursued to the Patent Trial and Appeals Board (PTAB), if the Examiner maintains a position which is incorrect under the law. Further appeals to the courts can be pursued, although relatively rare.

Whether your patent application is granted relatively quickly depends on many factors including: the goals of the client, the scope of protection sought, the degree to which the subject matter area of the invention is crowded (e.g. many similar inventions disclosed in the previously), how close your invention is to other inventions in the field, the disposition of the Examiner, and whether a patent search was performed before that application was filed, among other factors.

Provisional Patent Applications
Provisional patent application are not examined by the U.S. Patent and Trademark Office. Instead, they act to provide you with a priority filing date with respect to the subject matter disclosed in the provisional patent application. In order to receive a patent based on the subject matter disclosed in a provisional patent application, a non-provisional patent application must be filed within one year of the filing of the provisional application and make a claim of priority to the provisional application. Then that subsequently filed non-provisional application will proceed as described above. For more information on whether to start with a provisional or non-provisional click here.

Risks of Waiting: Independent Invention & Google PageRank

InThePlexQuestions often arise about the risks of waiting. Waiting to get started, waiting to file a patent application, waiting to launch a product or service, etc. The risk is that someone else is independently working on the same problem or the same idea and beat you to the market or the patent office or both. What’s the chance of someone independently inventing in your space?  It is impossible to know for a particular circumstance, but we do know that independent invention around the same time happens. According to In The Plex: How Google Thinks, Works, and Shapes Our Lives, Larry Page of Google was not the only person in 1996 to recognized that the link structure of the internet was the basis for a powerful way to find information on the web. The resulting algorithm based on link information allowed Google’s search results to be much better than other search engines at the time it was introduced. Two others were working on this idea, but Google was the first to make it to market.

Adjacent Possible and Independent Invention
As Steve Johnson noted in his book Where Good Ideas Come From sometimes when a scientist or inventor comes up with a new idea or invention, they learn that one or more others have independently come up with the same invention at about the same time period, such as within a year. Johnson notes that this is because the invention becomes “an adjacent possible” once founding or necessary elements or parts are created, discovered, or otherwise available.

To demonstrate the idea of the adjacent possible, Johnson notes that if Youtube was created 10 years earlier in 1995 it would have failed. This is because in 1995 most web users were on slow dial-up connections and it could take an hour to download a standard Youtube clip. In 1995, Youtube’s innovation was not within the adjacent possible, but ten years later, with broadband internet and Adobe’s Flash technology, it was.

PageRank
As In the Plex explains, Larry Page realized you can estimate the importance of a web page by considering the web pages that link to it. Page called his resulting algorithm based on linking “PageRank.” Page said, “In a way, how good you are is determined by who links to you and who you link to you determine how good you are…”Both the number of links and the quality of the links mattered. If a respected web page linked to a page, that link would be worth more than a link from a less well-known or respected website.

In addition to considering the number and nature of links, the anchor text used for the link is considered. Early search engines would provide results based on the content of a particular page. Therefore the results to a search for “newspaper” might not include the most relevant results, such as the New York Times website, if the New York Times did not say “I’m a newspaper” or otherwise use the word “newspaper” on its website. This caused search engine results to be poor and allowed the best result to be buried under relevant results. If a web page used the word “newspaper” to link to the the New York Times web site, then newspaper is the anchor text. And if enough and/or important websites link using that anchor text, then search results for “newspaper” would include the New York Times website. Page’s algorithm ranked each web page based on the incoming links, anchor text, and other factors. The resulting ranking of each page was used to determine which search results were returned for a particular web query, which vastly improved the relevance of search results as compared with other search engines at the time.

Others Independently Discovered Linked Based Ranking
Jon Kleinberg, a postdoctroal fellow at IBM’s research center, discovered that the link text, e.g. what a linking page said when linking to anther site and how many pages linked to a particular site, was useful in determining the relevance of a particular page. While having the same core idea, Kleinberg and Page each approached it differently. Kleinberg wanted to understand network behavior, Page wanted to build something that helped people in searching the web. Kleinberg said “all sorts of IBM vice presidents were trooping through Almaden to look at demos of this thing and trying to think about what they could do with it.” But they couldn’t figure out what to do with it.

Another person,Yanhong (Robin) Li, in 1996 saw the link between the ranking of scientific papers based on the number of other papers that cited them and ranking web pages in the same way. He came up with a search method that calculated relevance from both the frequency of links and the content of anchor text, which he called RankDex. Li described this to his bosses at Dow Jones, but they didn’t do anything with it. Li said, “I tried to convince  them [Dow Jones] it was important, but their business had nothing to do with Internet search, so they didn’t care.”

Li first filed a patent application himself in June 1996 after reading a self-help patent book. But then Dow Jones hired a patent attorney and filed another application on the same invention in February 1997, which resulted in US Patent No. 5,920,859. A patent application for Page’s Page Rank system was filed January 1997, which resulted in U.S. Patent 6,285,999. As you can see, the patent application were filed within about a month of each other. Li left Down Jones and eventually started Baidu in 2000, which is now the largest Chinese search engine.

Conclusion
These three individuals, Page, Kleinberg, and Li, were each independently working, without knowledge of the others, on one of the most important developments in improving internet search results, and in turn, improving the utility of the web. While it may be possible that you can sit on your ideas for years without consequence, waiting carries the risk that someone else is independently working on the same problem or invention as you. The patent office, and sometimes the market, rewards first movers.

When does my patent expire?

DSC01048Generally for newly issued patents, the patent will expire 20 years from the earliest filing date. However, the patent will expire earlier if maintenance fees are not paid. Maintenance fees must be paid to the United States Patent and Trademark Office (USPTO) at 3.5, 7.5, and 11.5 years after the date the utility patent issues. Maintenance fees are not required for U.S. design patents; they are required for utility patents. The USPTO provides a six month grace period after the 3.5, 7.5, and 11.5 year due dates, within which the maintenance fee can be paid with an additional surcharge.

Some patents receive a patent term extension from the USPTO, if under certain circumstances the USTPO takes too long to examine the patent application and grant the patent. Patent term extension periods are generally shown on the first page of a patent. You should seek advice of a patent attorney when trying to determine whether a particular patent is expired.

The maintenance fees are set by the USPTO and change periodically. The maintenance fees increase progressively so that the 7.5 and 11.5 year fees are greater than the 3.5 year fee and the 11.5 year fee is greater than the 7.5 year fee. This progressive fee structure incentivizes patent owners to let their patent expire unless they are generating a certain financial benefit to the patent owner at the time each respective maintenance fee becomes due.

Photo credit to flickr user kermitfrosch under this creative commons license.

Patentability v. Non-Infringement: If I Have A Patent How Can My Product Infringe?

Patentability_v_Non-InfringementHow can a patent be granted over an invention and yet the commercialization (manufacture or sale) of the invention infringe a prior patent? A common misconception about a patent is the belief that a patent is a permission slip that allows you to make or sell your invention. But this is not the case. A patent is not a grant of permission for you to make your invention. Instead a patent provides a negative right. This means the patent allows you to stop others from making, using, selling, offering for sale, and importing your invention in the United States. Therefore in certain situations the commercialization of a patent invention can infringe a prior patent as explained below.

Basic Example: The Wheel
Here’s an example with a very basic invention to illustrate the concept. John invents the wheel. John files a patent application on the wheel and the patent claims “a circular surface 10 that orbits a hub 12.” The patent office grants John a patent because no one else had previously disclosed such a wheel in the prior art.

The Improvement: Spokes
Bob sees John’s wheel and decides to improve on it. Bob decides rather than having a solid wheel, which is how John’s wheel is made, Bob is going to take all the weight and material out of the wheel and connect the outer surface 20 to the hub 22 by a number of spokes 24. Bob applies for a patent on his spoked wheel. The Patent Office grants Bob a patent on his spoked wheel having a claim of: a circular surface that orbits a hub, where the circular surface is connected to the hub by a plurality of spokes. The patent office grants this patent because the prior art did not disclose a circular surface that is connected to the hub by a plurality of spokes. This is the new component. Since there is something sufficiently new in Bob’s invention, the Patent Office grants him a patent on it.

Infringement Analysis
Can Bob make his spoked wheel without infringing on John’s wheel patent? As explained above, a patent is not a grant of permission to create the corresponding product of the invention. So if John claims infringement, Bob can not win in defense by pointing to Bob’s patent. To determine whether Bob’s spoked wheel would infringe John’s wheel patent, we have to compare Bob’s spoked wheel to the claims of John’s wheel patent. As explained above John’s wheel patent claims “a circular surface 10 that orbits a hub 12.”

Does the spoked wheel have a circular surface that orbits a hub? Yes. The spoked wheel has a circular surface 20 that orbits a hub 22. A product infringes a patent if the product has each and every element of at least one claim of the patent (or an equivalent element, under the doctrine of equivalents, but that is outside the scope of our discussion here). It doesn’t matter that Bob’s spoked wheel has more elements than John’s patent claim, e.g. the spokes. Infringement is avoided in most cases by removing elements that are claimed, and not by adding elements that are not in the claims. The patent claim(s) set out the minimum that is necessary for a product/process to be covered by that claim. So John’s wheel patent may be avoided by removing the circular surface and making it square, for example, but that won’t function very well as a wheel. Further John’s wheel patent might be avoided by eliminating the hub, but then it would be difficult to connect the wheel to other things, e.g. wagons.

You can see that a patent having broad claims can cover variations of the invention that might not have been known at the time of the invention. The claim of John’s wheel patent does not say a circular surface that orbits a hub, where the area between the circular surface and the hub is solid or filled. Such a “solid” limitation would narrow John’s patent. The solid limitation would not cover Bob’s spoked wheel because the area between the circular surface and the hub is not solid, but instead is open with intervening spokes. However, John’s Patent claims were well drafted to not include too many unnecessary limitations. This also demonstrates how critically important claim drafting is in writing a patent application. Sometimes one word in the claim can make the difference between your patent covering another parties product or not.

In this example case, no one can make a spoked wheel since the commercialization of Bob spoked wheel would infringe John’s patent. Also, Bob has a patent on the spoked wheel so that John and others can’t make a spoked wheel. If a third party, Adam, started making a spoked wheel, Adam’s spoked wheel would infringe both Bob’s spoked wheel patent and John’s wheel patent. The only way that spoked wheels can be made and sold during the life of John’s patent, is if Bob and John come to a licensing agreement. Such a licensing agreement could provide that Bob will pay John a royalty on each spoked wheel sold until John’s patent expires. When John’s patent expires, then Bob can sell his spoked wheel without John interfering.

While this example provides that Bob invented an improvement to John’s wheel after seeing it, the results would be the same for the case where Bob invented the spoked wheel without seeing or knowing of John’s wheel.

Conclusion
As the above example shows, the Patent Office does not care if your invention would infringe a prior patent. They only ask whether the invention provides something that is sufficiently new. Therefore, your patent will not protect you from claims of infringement from others. A patent is an offensive weapon not generally defensive (except in certain strategies not relevant to this example). The only way to know whether your product will infringe a prior patent is to have a clearance or non-infringement search completed.