Kaavo Inc. v. Amazon.com, Inc. (D. Del. 2018)

Cloud Computing Claims Deemed Patent-Ineligible

By James Korenchan --

This month, in an infringement case against Defendant Amazon, Judge Stark of the U.S. District Court for the District of Delaware ruled that Plaintiff Kaavo Inc.'s cloud computing claims are patent-ineligible under 35 U.S.C. § 101.  In related cases dating back to 2016, the Court ordered that the asserted independent claims be found patent-ineligible, as well as one of the dependent claims.  The Court later ordered limited discovery, claim construction, and summary judgement briefing with respect to the eligibility of the remaining dependent claims.  Kaavo then moved for reconsideration of the Court's Order invalidating all of the asserted independent claims and the one dependent claim, whereas Amazon moved for summary judgement.  The Court denied both motions without prejudice and instead ordered new briefing to allow for consideration of the § 101 issues in light of Federal Circuit patent-eligibility decisions since early 2016.  Renewals of both motions were at issue in this latest decision, in which the Court granted Amazon's renewed motion for summary judgement of invalidity of the remaining dependent claims and denied Kaavo's renewed motion to reconsider.

The patent at issue, U.S. Patent No. 8,271,974 (the '974 patent), entitled "Cloud Computing Lifecycle Management for N-tier Applications," is directed to managing a cloud computing environment, where the cloud environment is used for running a software application (e.g., a web-portal, a word processing program, an inventory management program, a database management service, etc.).  The patent describes a cloud environment as an environment in which cloud providers can provide users with access to resources such as datacenters and/or other IT-related capabilities.  More particularly, the patent describes that a cloud environment can be an "N-tier environment," which refers to logical groupings of components based on their functionality, examples of which include a "presentation tier," an "application tier (e.g., a logic or business logic tier)," and a "database tier."  Notably absent from the patent is a description of the improvements and advantages that the claimed invention provides -- an absence that worked heavily against Kaavo in this case.

The Independent Claims and Dependent Claim 12

First, the Court addressed Kaavo's motion for reconsideration of the asserted independent claims and dependent claim 12.  A representative independent claim of the '974 patent is as follows:

1.  A method for managing a cloud computing environment for use by a software application comprising:
    determining a requested initial cloud environment based on user-defined provisioning information, where the requested initial cloud environment is not yet instantiated and is an N-tier computing environment;
    sending an initialization event based on the requested initial cloud environment, where the initialization event is configured to cause an initial cloud environment configuration to be made available to an application;
    sending application data that is configured to cause the application to begin execution in the initial cloud environment configuration;
    receiving monitoring environment data that represents a current cloud environment state;
    determining a requested adjusted cloud environment based on the monitoring environment data, where the requested adjusted cloud environment is an N-tier computing environment; and
    sending a cloud environment adjustment event based on the requested adjusted cloud environment, where the cloud environment adjustment event is configured to cause an adjusted cloud environment configuration to be made available to the application.

And dependent claim 12 is as follows:

12.  The method of claim 1, where the determining a requested initial cloud environment is performed using a processor, a hardware circuit, or an integrated circuit.

In its motion, Kaavo argued that the Federal Circuit decisions in Enfish, McRO, BASCOM, Visual Memory, Core Wireless Licensing, Berkheimer, and Aatrix -- all of which had not yet been issued when the Court deemed the independent claims invalid under § 101 -- support a finding that the independent claims are directed to an improvement of the functioning of computers and improve existing technology and are thus patent-eligible.  In part, Kaavo argued that cloud computing is a "fairly recent" development and differs from conventional computing technology.

But the Court was not convinced, briefly highlighting the improvements provided by "the self-referential table in Enfish, the computer memory system in Visual Memory, and the user interface Core Wireless," and noting that neither the claims nor the specification in this case indicate the presence of any improvement sufficient to support a finding of validity.  Thus, the Court maintained that the independent claims are directed to the abstract idea of setting up and managing a cloud computing environment.

The Court did not specifically address dependent claim 12 in prong one, nor did the Court specifically address McRO, BASCOM, Aatrix, and Berkheimer.  (Though the Court cited Berkheimer elsewhere in the decision: "[w]hen there is no genuine issue of material fact regarding whether the claim element or claimed combination is well-understood, routine, [and] conventional to a skilled artisan in the relevant field, this issue can be decided on summary judgment as a matter of law.")

The Remaining Dependent Claims

The Court identified the following additional limitations as representative of all the remaining dependent claims:

• forecasting an optimal cloud environment for future use with the application based on monitoring environment data (claim 5)

• forecasting a future cost associated with executing the application based on monitoring environment data (claim 6)

• the user-defined provisioning information is determined using a needs analysis algorithm and a user input received from a user interface (claim 8)

• the user-defined provisioning information comprises geographic data (claim 9)

• receiving security information, determining a requested security action based on the security information, and sending a security event based on the requested security action (claim 11).

Addressing prong one of Alice, the Court stated that each of these dependent claims broadly recite functional goals of the claimed cloud environment, but that the patent does not describe the technical details of how to achieve the functional goals in a non-abstract way.

Claims 5 and 6, for instance, each recite a type of forecasting.  The patent generally describes forecasting as being performed using techniques such as "neural networks, time-series algorithms, and regression analysis," but the patent does not provide an algorithm or any additional details as to how forecasting is performed.  With regard to claims 8 and 9, the patent does not describe any algorithm or other manner in which to determine "user-defined provisioning information."  (Note: The Court found "need analysis algorithm" to be indefinite in claim construction.)  And for claim 11, the patent does not provide details as to how security information is received, how a security action is determined, or how a security event is sent.  The Court also noted that the patent mentions various modules for performing operations related to the dependent claims, but the claims do not require use of the modules.

The Court concluded its prong one analysis by reiterating that the dependent claims do not provide improvements to computer functionality and that the patent does not describe such improvements.  Kaavo's arguments to the contrary relied on expert opinion, but the expert opinion in turn relied on the cloud computing context of the invalid independent claims and was thus insufficient to create a genuine issue of material fact.

The Court's prong two analysis was much the same as prong one in that the Court highlighted the lack of specificity in the claims and specification:

For instance, Claims 5 and 6 do not specify how the forecasting is performed, what monitoring data is used, or how it is used; any generic algorithm, neural network, or regression analysis could be used.  Claim 8 does not limit the method to any specific algorithm or user interface; any generic algorithm or user interface could be used.  Claim 9 does not teach how the geographic data is created or determined, or what type of data is required; any generic method or data could be used.  Claim 11 does not limit how the security information is collected, how the security action is determined, or how the security event is used; any generic method could be used.  Plaintiff has not cited to any portion of the specification that fills any of these gaps.

The post-Alice eligibility landscape can be tumultuous, but it is consistently important that a patent recites and describes sufficient details as to how the invention provides non-abstract, innovative technological improvements.

Kaavo Inc. v. Amazon.com, Inc. (D. Del. 2018)
Memorandum Opinion by District Judge Stark

Comments

[Paul Cole]

Another decision from which it is apparent that detail in the application at the time of filing, and claims that are not over-broad and over-speculative, are critical to success.

[Skeptical]

Ping-pong-ping-...

[Skeptical]

...and on a different note (more towards the elephant in the room),

"...ordered new briefing to allow for consideration of the § 101 issues in light of Federal Circuit patent-eligibility decisions since early 2016."

As a CAFC judge noted, the use of such cases to WRITE the law is known as "common law" writing.

Of course, the problem with common law writing is that patent law is a particular set of laws that the Constitution has set out so as to have only ONE authorized branch of the government that can be the writer of that law. And that branch is NOT the judicial branch.

This is not to say that Congress cannot share its authority. Such sharing has occurred on a number of occasions throughout this nation's history.

But it is to say that such sharing is very much constrained and requires (among other things) clear avowal and limits pronounced.

As has been (dangerously?) pointed out, the Supreme Court has NOT been granted such authority since Congress created section 35 USC 101 (and carved out from the previous single paragraph, section 35 USC 103).

The use of common law then for writing (in the common law style) the patent law of 35 USC 101 is thus ultra vires.

Will Congress (or for that point, pretty much any attorney hewing to their state ethics oaths) wake up to this FACT?

Sadly, I remain:

[Karl P. Dresdner, Jr. PhD, US Patent agent 63,319]

Patentable are a paper clip, chemicals, machines, methods of manufacture, the list goes on and on. These kinds of invention were patented because a growing commerce needed them. The patent office is essentially as old as the Constitution but failing because it is not evolving in a progressive manner.
35 USC 101 is illogical in denying all kinds of computing technology and worst of all is harming the US economy and the harm is growing as the legal system blindly invalidates valuable patents and hurts established companies. It is time for change. 35 USC 101 suffers the invertebrate mentality of discrimination which plagued [plagues]sexes and races. The purpose for patents is lost upon the high & mighty court gods who do not budge from playing it safe.
All computing technology must be used with computer hardware that is located somewhere. It is a farce to confuse mental acts with software that acts faster, has more memory, and can do things a human mind never can. Our minds are not driving like software at a super-high speed that operates within a manufactured computer. Biological mental acts and non-biological mental acts are not the same. The later should be patentable.
The patent office is doing unneeded harm to a major fast growing major profit sector of the US economy. It is anti-business and falling in its fundamental purpose. 35 USC 101 needs a big reality check.