Skip to content

The Software Squeeze: The Treacherous Path for Software Patent Eligibility

The Software Squeeze: The Treacherous Path for Software Patent Eligibility published on No Comments on The Software Squeeze: The Treacherous Path for Software Patent Eligibility

The Alice decision has led to a heap of opinions invalidating patents for computer-related inventions – each holding that the reviewed claims provide (a) an idea that does not involve technology, and (b) an instruction to “apply it” using a generic computer. These decisions rebuke patentees’ failure to limit their claims with the details of a particular technique, and for overbroadly claiming any solution involving a simple concept.

However, a second line of cases, also citing Alice, have invalidated patents for the opposite reason: the claims recite an abundance of technical detail about a specific technique, yet the courts dismiss these details as either “purely conventional,” and/or “mathematical concepts” that are not eligible for patent protection.

These opposing lines of cases, driving against the body of software patents from opposite directions, require the inventors of computer-related inventions to walk a narrow path between qualitative description (claims rejected for overbreadth) and algorithmic description (claims rejected as mathematical or conventional). This path is being steadily narrowed, as the piles of invalid patents on each side continue to mount – threatening a categorical exclusion of computer-related techniques as patent-eligible inventions. This result is astonishing not only in its own right, but in contrast with the Supreme Court’s explicit and consistent rejection of such an exclusion over the 35 years since Diamond v. Diehr.

1. The Invalidity of High-Level Concept Claims

The Supreme Court’s efforts to reshape patent law are long on zeal, but short on specifics. The Alice decision pivotally turned on whether the claim recited an “abstract idea,” which the Court explicitly refused to define, choosing instead to drop hints that might mitigate for or against this determination.

The application of this principle by the lower courts has led to some lines of reasoning. Case (X) picks up on one of these “hints” from Alice and elaborates on its application to a particular patent, and following cases (Y) and (Z) cite the rationale in case (X), etc.

The first line of reasoning that has sprung from Alice involves cases that recite only an “idea,” without the details that provide for a limited set of embodiments or solutions. These cases invoke the following portion of Alice:

An instruction to apply the generic idea of using intermediated settlement using some unspecified, generic computer is not “enough” to transform the abstract idea into a patent-eligible invention.

Consider the following claims under review in this line of cases – specifically, the underlined portions:

  • IP Engine v. AOL:
      A search engine system comprising: a first system for receiving informons from a network on a continuing search basis, for filtering such informons for relevancy to a query from an individual user, and for storing a ranked list of relevant informons as a wire; a second system for receiving informons from a network on a current demand search basis and for filtering such informons for relevancy to the query from the individual user; and a third system for selecting at least one of the first and second systems to make a search for the query and to return the wire or demand search results to the individual user.
  • Thales Visionix v. Elbit Systems:
      A method comprising determining an orientation of an object relative to a moving reference frame based on signals from two inertial sensors mounted respectively on the object and on the moving reference frame.
  • buySAFE, Inc. v. Google:
      A method, comprising: providing a transaction performance guaranty service, wherein the computer offers the transaction performance guaranty service that binds a transaction performance guaranty to the online commercial transaction involving the first party to guarantee the performance of the first party following closing of the online commercial transaction.
  • Ex parte Cote:
      A method of using clusters in electronic design automation, the method comprising: … receiving data including a plurality of clusters… and preparing a phase shifting layout for the original layout by phase shifting each of the plurality of clusters independently of one another.
      An electronic design automation program comprising: a source code segment designed to receive data in a cluster format, each cluster having a predetermined spatial relationship to each other, and wherein the plurality of clusters are phase shifted independently of one another.
  • AllVoice v. Microsoft:
      A data processing method comprising: … inputting the recognised words into a processing application program which places the words in positions in the application, and which processes the recognised words such that positions of the recognised words are changed to form a processed word string; and using the interface application program to determine the positions of the recognised words in the processing application program, monitor changes in the positions of the recognised words, and to form link data linking the audio data to the recognised words, said link data comprising the audio identifiers and the determined positions of corresponding recognised words, said link data being updated in response to monitored changes in positions of the recognised words;
  • In re Fitt:
      A spatial index, comprising: a product structure represented by a network of nodes and edges; a local geometric bound set for each of said network nodes; a cumulative geometric bound of said local geometric bound set propagated along each of said network edges; and a match formed between a data model object and said cumulative geometric bound of each of said network nodes.
      A system for data management, comprising: a geometric model having a plurality of geometric model objects; a list of data model objects corresponding to said geometric model to define a product structure; and a geometric spatial index that supports geometric queries on said list of data model parts in a time-efficient manner.
  • Ex parte Villena:
      A system of improving the accuracy of analytic processing of real estate data, the improvement comprising, a means for data mapping, and a means for preconditioning data.
      A system for providing real estate information, comprising… a database of real estate property data hosted on the server computer, which database is populated with data that has been data mapped and preconditioned.
  • Ex parte Tie:
      A computer-implemented method, comprising: determining at least one element of a non-monotonic sequence, the non-monotonic sequence being one of a family of related non-monotonic sequences; using at least said element, determining at least one property of a number; and depending on said property, taking an action the effect of which is to enhance or degrade data security within a computer system or network.
  • Intellectual Ventures v. Capital One:
      A system for providing web pages accessed from a web site in a manner which presents the web pages tailored to an individual user, comprising: an interactive interface configured to provide dynamic web site navigation data to the user, the interactive interface comprising: a display depicting portions of the web site visited by the user as a function of the web site navigation data; and a display depicting portions of the web site visited by the user as a function of the user’s personal characteristics.

Given those types of claims, and specifically the underlined portions, the courts’ decisions are understandable. These claims present a black box of functionality – where those of ordinary skill in the art wouldn’t be able to implement it easily. Rather, these disclosures exhibit the fatal flaw that key details are – as engineering texts often state – “an exercise for the reader.”

In some cases, the “black box” approach is perfectly fine – indeed, common and routine in engineering. An invention that utilizes data that is encrypted can be claimed as a method including the step: “encrypt the data.” How is that done? Rijndael, AES, RC4, DES… what algorithm do you use? In many such cases, the particular algorithm does not actually matter, so long as the data ends up encrypted – and a person of ordinary skill in the art can choose among many known cryptographic algorithms to carry out that step of the invention.

But in other cases (including the underlined portions above), the “black box” is a gaping hole of functionality that is undisclosed – and perhaps not even invented yet! How exactly does one use a computer to “determine that a non-monotonic sequence is one of a family of related non-monotonic sequences?” Or to “form matches between a data model object and cumulative geometric bounds of each of the network nodes?” These steps might be difficult or even impossible for a person of ordinary skill in the art to implement. In the worst of these cases, the claim does not even recite functionality, but simply the result, like: “encrypt the data in a manner that is unbreakable,” or “compress the data with a compression ratio of at least 95%.”

Attempts to claim inventions this way feel like “wishing” these capabilities of a computer into existence. However, in computer science, the line between “anyone can implement it easily” and “no one knows how to do this” can be subtle – as well as misunderstood. 1

The solution in these cases is clear: disclose, and better still claim, the missing functionality. Recite an algorithm that could be used to carry out the invention. Use the specification to teach how a software developer of ordinary skill would implement the novel technique using a computer and any specialized hardware. The resulting patent is certainly narrower, but it’s better to have a narrower patent than an invalid one.

Unfortunately, this solution runs directly counter to another line of emerging case law.

2. The Invalidity of Low-Level Claims As “Conventional” and/or “Mathematical”

The second line of cases involves the invalidation of patents that claim a specific technique, on the grounds that the claimed technique is an abstract “law of nature” or “scientific discovery,” rather than actual technology. These cases draw on the following citation of Alice (citing Mayo, in turn citing Gottschalk v. Benson):

Simply implementing a mathematical principle on a physical machine, namely a computer, [i]s not a patentable application of that principle.

Other cases invoke the following portion of Alice:

Taking the claim elements separately, the function performed by the computer at each step of the process is “[p]urely conventional.” Using a computer to create and maintain “shadow” accounts amounts to electronic recordkeeping—one of the most basic functions of a computer. The same is true with respect to the use of a computer to obtain data, adjust account balances, and issue automated instructions; all of these computer functions are “well-understood, routine, conventional activit[ies]” previously known to the industry. In short, each step does no more than require a generic computer to perform generic computer functions.

This line of reasoning suggests that if each disclosed step of a process is a capability that is easily within the functional powers of computers… then it can be dismissed!

These snippets of Court wisdom have been used to reject claims that recite actual computer-implemented technology – irrespective of the clear application of techniques to a specific technology context; the novelty or non-obviousness of the combination; the technical merit of the invention, and the sufficiency of the disclosure. Examples include:

  • Google v. Unwired Planet, LLC:
    A method of controlling access to location information generated for wireless communications devices operating in a wireless communications network, comprising:

    receiving a request associated with at least one wireless communications device over a data network from a requesting client application;
    accessing a subscriber profile associated with the at least one wireless communications device in memory, wherein the subscriber profile includes: a list of authorized client applications; and a permission set for each of the authorized client applications, wherein the permission set includes at least one of a spatial limitation on access to the location information or a temporal limitation on access to the location information;
    determining whether the requesting client application is an authorized client application;
    determining whether the request is in accordance with the permission set for the authorized client application; and
    if the requesting client application is an authorized client application and the request is in accordance with the permission set for the authorized client application, providing access to the location information for the at least one wireless communications device.
  • McRO v. Activision:
    A method for automatically animating lip synchronization and facial expression of three-dimensional characters comprising:

    obtaining a first set of rules that define output morph weight set stream as a function of phoneme sequence and time of said phoneme sequence;
    obtaining a timed data file of phonemes having a plurality of sub-sequences;
    generating an intermediate stream of output morph weight sets and a plurality of transition parameters between two adjacent morph weight sets by evaluating said plurality of sub-sequences against said first set of rules;
    generating a final stream of output morph weight sets at a desired frame rate from said intermediate stream of output morph weight sets and said plurality of transition parameters; and
    applying said final stream of output morph weight sets to a sequence of animated characters to produce lip synchronization and facial expression control of said animated characters.
  • Ex parte Kun Liu:
    A computer-based method for generating an anonymous graph of a network while preserving individual privacy and the basic structure of the network, said method comprising the steps of:

    receiving an input graph G(V,E), wherein V is the set of nodes in said input graph and E is the set of edges in the input graph;
    determining a degree sequence d of the input graph G(V,E), wherein d is a vector of size n=|V|, such that d(i) represents a degree of the ith node of the input graph G(V,E);
    applying a programming algorithm to the degree sequence d to construct a new degree sequence {circumflex over (d)}, wherein the new degree sequence {circumflex over (d)} has an integer k degree of anonymity wherein, for every element v in sequence {circumflex over (d)}, there are at least (k-1) other elements taking the same value as v, and wherein said programming algorithm minimizing distance between the degree sequence d and the new degree sequence {circumflex over (d)};
    constructing an output graph G(V,E) based on the new degree sequence {circumflex over (d)}; and
    outputting the constructed output graph G(V,E), wherein E is the new set of edges in the output graph, and such that E .andgate. E=E or E .andgate. E.apprxeq.E (relaxed version).
  • Ex parte Gross:
    A computer implemented method for structure evaluation of time-dependent acyclic graphs, the method comprising the following steps performed by at least one hardware processor:

    determining a validity of at least one node of an acyclic graph structure, with the at least one hardware processor, by comparing an identified validity period to the at least one node of the graph structure, each of at least a subset of the at least one node associated with a particular time period;
    identifying at least one invalid node, the at least one invalid node associated with a time period outside the identified validity period;
    removing the at least one identified invalid node;
    substituting, for each of the at least one removed invalid nodes, a corresponding skip node, where each skip node corresponds to a single removed invalid node, each skip node including a skip function, each skip function defining a skip relation that defines a relationship for valid nodes adjacent each skip node, the relationship pointing from a last valid node located before the skip node to a next valid node located after the skip node;
    generating a validity graph for the identified validity period including:

    at least one of the nodes of the graph structure that is valid for the identified validity period based on the time period associated with each of the at least one nodes, and
    at least one skip relation for each skip node, where data in the at least one invalid node is absent from the validity graph; and

    storing the validity graph in a repository.

These claims do not provide a high-level description of a technique. These claims do not only describe the result. Rather, each claim recites a specific set of steps that are to be followed to perform the disclosed technique. A person of ordinary skill in the art could implement each claim by taking each step, and writing a straightforward set of instructions that carries it out.

When the courts that invalidated the first set of cases issued their opinions, they might well have referred to these claims as examples of how claims ought to be written.

And yet, these claims were also invalidated – on the following grounds:

  • Google v. Unwired Planet, LLC:
    The claimed invention is directed to an abstract idea, specifically, the abstract idea of controlling access to location information using a subscriber profile. The preambles and all claim limitations of both claims support this abstract idea, and the ‘752 patent repeatedly discloses that the invention is directed to controlling access to location information. We are persuaded that the identified abstract idea is similar to the kind of “organizing human activity” at issue in Bilski and Alice. 2
  • McRO v. Activision:
    While the patents do not preempt the field of automatic lip synchronization for computer-generated 3D animation, they do preempt the field of such lip synchronization using a rules-based morph target approach. And if, as Plaintiff suggests, the inventive step is the use of timing rules, given the state of the prior art, that still leaves an abstract idea at the point of novelty, and preventing the development of any additional ways to use that abstract idea in the relevant field. 3
  • Ex parte Kun Liu:
    Applying the first step [of Alice], we find that claim 1 is directed to a mathematical model for producing a mathematical based graph. Each step involves the use of mathematics and/or algorithms. Accordingly, we find that the claim is directed to a patent-ineligible concept. The preamble includes features such as computer-based method, processor, computer storage medium, but these features are not set forth in the method steps. Moreover, these features are the type of generic element that have been determined to be insufficient by the Supreme Court to transform a patent-ineligible claim into one that is patent-eligible. 4
  • Ex parte Gross:
    The claimed invention involves the application of a filter to identify an invalid node (i.e., mathematical application in a computer) and producing a graph indicating a skipped or invalid node as it relates to the valid nodes before and after the skipped node. The claim is directed to a patent-ineligible concept (i.e., abstract mathematical application and creation of a graph). Furthermore, the creation of a validity graph and storing it in a repository does not constitute an inventive concept that is sufficient to ensure that the claims amount to significantly more than a patent-ineligible concept. 5

The disturbing tone of these cases is a refusal to see the forest for the trees – or, in the language of Diehr, to consider the claim “as a whole.” It is equivalent to looking at a machine and stating: “This is merely a combination of well-known parts: nuts, bolts, brackets, power sources, and motors. The individual components are entirely conventional, and so add nothing to the abstract, inventive concept of the machine.” 6

3. The Intersection

When reciting an invention reciting software, a claim can either describe those portions in one of three ways:

  1. Systematically – i.e., the functionality of each component of the technique (such as “an encryption module that encrypts the data”);
  2. Logically – i.e., the conceptual model, structure, or logic of the technique; or
  3. Algorithmically – i.e., the activity performed by the computer to implement each step or element of the technique.

However, the two lines of cases above – one pressing from the top down, demanding more detail; and one pressing from the bottom up, denigrating the information providing these demanded details – cast a pall over all three claim-drafting approaches. The intersection leaves patents for computer-related inventions in a precarious state, trapped in a vise between two treacherous conditions. This trap manifests in two different ways:

Trap #1: Functionality vs. Logic

  • Claims to computer-related inventions that recite the details in a qualitative way – such as the objective or functionality of each step – are at risk of being rejected under 101, under overbroad, results-oriented claiming. (Alternatively or additionally, such elements are increasingly likely to be dragged into a 112(f) means-plus-function interpretation, and then invalidated under 112(b) because the specification does not support it.)

  • Claims to computer-related inventions that recite the details in a quantitative way – the specific, algorithmic logic that is applied at each step – are at risk of being rejected under 101, as reciting an abstract mathematical model or data structure.

Trap #2: Overview vs. Algorithm
  • Claims to computer-related inventions that recite the details as a high-level overview – such that elements or steps are presented as black boxes – are at risk of being rejected under 101, as failing to be sufficiently tied to technology.

  • Claims to computer-related inventions that recite the details as a low-level algorithm – such that the elements or steps describe exactly what a computer is doing at each step – are at risk of being rejected under 101, as reciting “conventional” computing functions that are generalized out of context into: “storing data,” “reading data,” “performing calculations,” “searching,” “providing output,” etc.

How might we describe the path between each of these zones of patent-ineligibility? Elusive? Speculative? Volatile? Imaginary? Treacherous?

We can certainly point to examples of claims that should careen into each chasm of ineligibility. However, the converse is not true: any patent claim targeting the narrow path between these perils may be held to fall into either of them – based solely on the reviewer’s philosophical bent about technology, and personal definitions of empty terms like “abstract” and “algorithm.” The sum of these cases is an in-fact categorical exclusion of patent eligibility computer-related inventions, particularly those utilizing software as a significant element.

This is a surprising result, given the consistent refusal of the Court to recognize any such exemption, dating back to Diehr. The Court in Bilski v. Kappos strongly denied this proposition:

It was once forcefully argued that until recent times, “well-established principles of patent law probably would have prevented the issuance of a valid patent on almost any conceivable computer program.” But this fact does not mean that unforeseen innovations such as computer programs are always unpatentable. Section 101 is a “dynamic provision designed to encompass new and unforeseen inventions.” A categorical rule denying patent protection for “inventions in areas not contemplated by Congress… would frustrate the purposes of patent law.”

The machine-or-transformation test may well provide a sufficient basis for evaluating processes similar to those in the Industrial Age – for example, inventions grounded in a physical or other tangible form. But there are reasons to doubt whether the test should be the sole criterion for determining the patentability of inventions in the Information Age. As numerous amicus briefs argue, the machine-or-transformation test would create uncertainty as to the patentability of software, advanced diagnostic medicine techniques, and inventions based on linear programming, data compression, and the manipulation of digital signals.

The Alice Court similarly refused to create broad exclusionary rules:

We tread carefully in construing [the] exclusionary principle lest it swallow all of patent law. At some level, “all inventions . . . embody, use, reflect, rest upon, or apply laws of nature, natural phenomena, or abstract ideas.” Thus, an invention is not rendered ineligible for patent simply because it involves an abstract concept. “[A]pplication[s]” of such concepts “to a new and useful end,” we have said, remain eligible for patent protection.

These platitudes about the value of software, the breadth of patent law, and the narrowness of the 101 “useful” exclusion squarely conflict with the application of patent law to specific claims.

4. The Future

The open question of the patent-eligibility of computer-related inventions raises the critical question: Where is patent law heading?

This question requires us to step back and ask: Is this actually what the Court intended? In both Bilski and Alice – as well as Chakrabarty and Diehr – the Court has held consistently that 35 USC 101 prohibits the formation of broad, categorical exclusions for particular classes of technology. It would be disingenuous for the Court to mold its case law to reach this result as a policy-based end-run around the statute. The more likely explanation is that the Court did not intend Alice to be so applied: its sentiment that it “need not labor to delimit the precise contours of the ‘abstract ideas’ category” was coupled with a presumption that the application of this “non-delimited” test would not end up “swallowing all of patent law” – and yet, here we are.

As I have noted, even Diehr’s claims would be fail miserably under today’s Alice rationale. The unqualified use of Diehr as the high-water mark in every 101-based Court decision for 35 years demonstrates that the ground truth of Alice does not comport with the Court’s intentions.

As Judge Moore noted in the Alice Federal Circuit en banc dissent-in-part:

Today, several of my colleagues would take that precedent significantly further, lumping together the asserted method, media, and system claims, and holding that they are all patent-ineligible under § 101. Holding that all of these claims are directed to no more than an abstract idea gives staggering breadth to what is meant to be a narrow judicial exception. And let’s be clear: if all of these claims, including the system claims, are not patent-eligible, this case is the death of hundreds of thousands of patents, including all business method, financial system, and software patents as well as many computer implemented and telecommunications patents.

If all of the claims of these four patents are ineligible, so too are the 320,799 patents which were granted from 1998-2011 in the technology area “Electrical Computers, Digital Processing Systems, Information Security, Error/Fault Handling.” Every patent in this technology category covers inventions directed to computer software or to hardware that implements software. In 2011 alone, 42,235 patents were granted in this area. This would render ineligible nearly 20% of all the patents that actually issued in 2011. If the reasoning of Judge Lourie’s opinion were adopted, it would decimate the electronics and software industries. There are, of course, software, financial system, business method and telecom patents in other technology classes which would also be at risk. So this is quite frankly a low estimate. There has never been a case which could do more damage to the patent system than this one.

That this dire consequence was explicitly recognized at the time of Alice, when calls for patent reform to address perceived abuses were at maximum fervor, is remarkable. The recognition of its truth and implications will continue to grow, and the emergence of problematic cases such as noted above will push it into center stage. And although the score card is lopsided, there have been some decisions (notably DDR Holdings) upholding claims to technology that satisfy 101.

In the interim – our mission, as patent practitioners and owners and examiners, is to continue making a strong case for the utility and value of software, such as:

  • For practitioners – submitting well-written patent applications for computer-related inventions, and arguing zealously on their behalf.

  • For examiners – giving meaningful computer-related patent applications a fair review, and, when the claims are well-distinguished from the claims, issuing a notice of allowance that explicitly supports their sufficiency under 101.

  • Publicly discussing the unforeseen consequences of Alice, such as during public roundtables.

  • Responding to opportunities for input: filing amicus briefs in notable cases, and submitting responses to the USPTO’s requests for public input. 7

The next several decades of patent law – and the technology industry! – will be defined by the courts’ application of the Alice decision (possibly soon to be supplemented with functional claim language as another crucial issue). The gravity of this issue is difficult to overstate, and we must apply ourselves to this area with all of the energy that we can muster.


  1. In a rare moment of introspection, the Mayo opinion acknowledged that judges are not good decision-makers on subtle technical issues:

    In any event, our cases have not distinguished among different laws of nature according to whether or not the principles they embody are sufficiently narrow. And this is understandable. Courts and judges are not institutionally well suited to making the kinds of judgments needed to distinguish among different laws of nature.

    This quote raises questions of how adequately courts and judges are “well-suited” to formulating economic policy regarding the industry of technology, and then manufacturing a lot of brand-new legal complexity, as the reinterpretation of a single word (“useful”) of a statute that has not changed since 1952.

  2. The court’s technical error, in this case, is conflating the broad concept of user privacy – i.e., whether Service Provider #1 can accept private information about a user, and hand it over to Service Provider #2 – with computer security – i.e., whether an application executing on a device is or is not permitted to access stored data.

    This precedent bodes poorly for patent eligibility within the entire field of computer security – which, when discussing issues like trust and authentication, routinely uses Alice and Bob analogies. While the security experts in these discussions implicitly understand this as an anthropomorphic description of the interaction of devices, the courts do not seem interested in these minor details in their abstract-by-analogy analysis.

  3. This decision appears to assert that a patent can be ruled invalid if it would “preempt” the use of a claimed technique, no matter how specifically defined or recited – which is the sole legal right granted by every patent! This portion of the McRO holding is especially bizarre, given this explicit warning in the same decision:

    We must be wary of facile arguments that a patent preempts all applications of an idea. It may often be easier for an infringer to argue that a patent fails § 101 than to figure out a different way to implement an idea, especially a way that is “less complicated – less liable to get out of order – less expensive in construction, and in its operation.” But the patent law does not privilege the leisure of an infringer over the labors of an inventor. Patents should not be casually discarded as failing § 101 just because the infringer would prefer to avoid the work required to develop non-infringing uses of the abstract idea at the heart of an appropriately circumscribed invention.

  4. Although the PTAB’s decisions are often succinct, the terseness of this rejection is a plain categorical assertion that claim elements that are “mathematical” or “algorithmic” are per se “abstract.” This is plain error by the PTAB, and one that the courts should address with due haste – we can only hope that the applicants recognize this opportunity and take this step.
  5. The problem in this case is that PTAB judges do not seem to be aware of the widespread use of directed acyclic graphs (DAGs) throughout computer science. In fact, a Google search for the term “directed acyclic graph” produces search results in the following areas of software architecture:

    That is literally the first page of results from Google. Yet, to the PTAB, DAGs are disembodied mathematical concepts that may or may not have any practical use within computer science…

  6. Of course, this argument has never been applied to machines – where every element is given due weight, and the combination is respected – only with inventions in the field of software does this dismissive approach seem to manifest, as further evidence for the peculiar, and increasingly malicious, disrespect of software by the courts as a legitimate form of technology.
  7. The USPTO’s requests for input should make note of the proper context of these requests – specifically: the USPTO has no authority to make substantive changes to patent law. Organizations like the Electronic Frontier Foundation routinely misunderstand this facet, and submit embarrassingly uninformed proposals that the USPTO cannot legally implement.

    Rather, these opportunities can be used for two types of ideas: first, ideas for improvements to patent examination and administrative processes that are within the USPTO’s authority, and that could raise tangible benefits for the patent system; and second, data that the USPTO can bring to Congress to inform patent reform initiatives.

Leave a Reply

Your email address will not be published. Required fields are marked *