Software Litigation Consulting

Andrew Schulman
Consulting Technical Expert & Attorney

Patent Litigation book

Patent Litigation: An Introduction to Using Patent Claims and Limitations to Show Infringement and Invalidity

[Title possibly changing to: Patent Litigation in the Age of AI: A Somewhat Boring Guide to a Surprisingly Important Topic]

TL;DR — NotebookLM’s summary of a version of this page: This excerpt details the structure and content of Andrew Schulman’s forthcoming book, Patent Litigation: An Introduction to Using Patent Claims and Limitations to Show Infringement and Invalidity. The book, still in draft form with a working title referencing the Age of AI, takes a detailed, bottom-up approach to the practicalities of patent disputes, focusing heavily on the minutiae of patent claims and their limitations. Schulman explicitly notes his use of AI tools like Google’s NotebookLM and various chatbots to help organize his extensive notes, even creating an AI “chatbook” version for readers to query about specific topics like claim charts and discovery. The comprehensive 24-chapter outline covers everything from the initial investigation and the role of claim construction, to detailed infringement analysis, invalidity defenses (anticipation, obviousness, and enablement), and concludes with remedies, damages, and future trends.

by Andrew Schulman

I resumed serious work on this book in August 2025, using Google’s NotebookLM to try to whip my hundreds of pages of rough notes and drafts into a coherent outline from which I can finish writing the book. See page on using NotebookLM to finish writing the Patent Litigation book. I won’t be delegating much actual writing to NotebookLM or other AI chatbots, but I’ll be using NotebookLM, the Gemini chatbot embedded in Google Drive, Anthropic Claude, and ChatGPT to help organize my material.

There is an AI “chatbook” version of this Patent Litigation book; you can query it about the topics noted in this outline (though the chatbot’s answers are not always an accurate reflection of my notes). Try asking it questions such as:

  • “What are underlying purposes for claim charts?”
  • “What’s the book say about the discovery phase of patent litigation?”
  • “What are key cases on claim chart acceptability?”
  • “How does collateral estoppel apply to patent invalidity?”
  • “The book seems to have a lot from economics scholars like Romer, Scotchmer, Mokyr, Jones, Simon. What’s all that doing in a book on claim charts, source code, reverse engineering, etc.?”

Some topics below are covered with detail disproportionate to their role in the entire book; for example, there’s a lot more below on enablement than on obviousness. This is because these topics were only sparsely covered in earlier versions of the book’s outline, so I’m over-compensating for them here. Coverage in the final book should be more proportionate.

Summary outline

  • Preface
  • Introduction
  • Chapter 1: “$1 Billion … for That?!” (Caltech v. Apple/Broadcom) and “Why Not Sue Over This?” (Google AI Transformer Patents Not Asserted Against ChatGPT)
    Chapter 2: “Promoting Progress of Science and the Useful Arts”: How Does Patent Litigation (and the Patent System Generally) Fit In?
    Chapter 3: Patent Infringement, Claims, and Limitations
    Chapter 4: A Small Example: Printer Toner
    Chapter 5: An Extended Example: “Facade Server” Patent
    Chapter 6: The Role of Claim Construction in Patent Litigation
    Chapter 7: Literal Infringement Analysis: Comparison of a Patent Claim with an Accused Product or Service
    Chapter 8: Equivalence, Function/Way/Result, and Interchangeable Parts Analysis
    Chapter 9: “Means for…”: So-Called “Functional Claiming”
    Chapter 10: Defending a Patent Infringement Case
    Chapter 11: Patent Invalidity: Pulling the Rug Out from Under Plaintiff’s Property
    Chapter 12: Obviousness — Both “Crown Jewel” and Mere Fallback
    Chapter 13: Enablement and the Underlying Purpose of the Patent System: Do Patents Disclose New Information?
    Chapter 14: Remedies, Damages, and Patent Valuation “Show Me the $”)
    Chapter 15: Pre-Filing Investigation, “Informal Discovery,” and Monitoring the Marketplace
    Chapter 16: Bringing (and Defending) a Patent Infringement Suit: Pros, Cons, and Plausibility
    Chapter 17: Claim Charts and the Local Patent Rules (LPRs)
    Chapter 18: Discovery in Patent Litigation: Getting Information from the Other Side
    Chapter 19: Expert Witnesses and Expert Evidence in Patent Litigation
    Chapter 20: Strategy, Tactics, and Alternatives to Patent Litigation
    Chapter 21: Multiplicity Issues
    Chapter 22: International Issues
    Chapter 23: Trends
    Chapter 24: Conclusion
  • Glossary & Abbreviations
  • Checklists
  • Bibliography
  • Cases Cited

 

Detailed (sometimes overly-detailed) rough outline

Preface

  • “In the Age of AI” in title: AI & machine learning in patent litigation
    • Only a little bit on 101 patentability of LLMs and ML, or on AI as inventor
    • More on how to use (and not use) AI chatbots in litigation; how modern AI changes discovery, claim charts, expert work, source-code examination (even for cases not directly related to AI or ML); confidentiality issues with AI use (often need a local model)
    • And on litigation of AI/ML inventions
    • Book will include some back-and-forth dialogs with AI chatbots, to illustrate where they help or hinder
  • About the reader (intended audience: associate attorneys; technical consultants; and patent litigation students)
  • About the author (attorney and computer programmer; has worked for 25+ years on factual side of software/network patent litigation; about equally for defendants & plaintiffs, including some NPEs)
  • Plaintiff/defendant neutral
  • Largely USA-centric (though see section on international aspects, including import, export, ITC, China, UPC)
  • Largely utility patents; some design patents
  • The book might have been subtitled “How to get down in the weeds”; the book is definitely for those who want to do that, rather than stay at a 30,000 foot level
    • Though also see lengthy sections on patent litigation strategy; and Chapter 2 on the role of the patent system, and patent litigation and patent claims in particular, in innovation and economic growth.
    • “Somewhat Boring” in subtitle: not that author actually sees it that way, but much of the work is tedious, and book “leans into” tedium. The book emphasizes low-level details. Something that looks Big and Important at the 30,000 foot level is really a collection of lots of little details.
    • Book basically takes a reductionist, bottom-up approach, but tries (especially in Chapter 2) to tie this to “the big picture”.
  • Not a standard treatise or casebook; reader won’t see much in-depth analysis of case opinions (there are many excellent books for that); instead, the reader will see many topics based on problems/solutions the author encountered in 25+ years working on the tech side of patent litigation, and hasn’t seen recorded elsewhere.
  • Overwhelming amount of detail in some areas is not intended to contradict the general principle of “Less Is More”
  • Acknowledgements

Introduction

  • Patent claims are tools to do something, not passive descriptions of technology
  • Really an entire book focusing on the minutiae of patent claims?; yes, even billion-dollar patent judgments end up being based on a few words in a patent claim
  • The book’s focus is on litigation (how to actually use a patent), rather than on patent prosecution (getting a patent in the first place)
  • Focus on claims, and on the limitations (elements and steps) that make up claims
    • A working subtitle said the book was a “Claims-Based Guide”, and some readers will think “Well heck, what else would it be, if not based on the claims?”, but unlike books on patent drafting and patent prosecution, material on patent litigation generally deals with nitty-gritty of the claims than one might expect
  • Focus on facts (factual investigation and marshaling facts), more than on patent law as such
    • A bit of a false distinction, since law itself emerges from the facts of individual cases, at least under Common Law: “ex facto jus oritur“, roughly translated as “The law springs from the fact; law is born of fact” or “The law arises out of the fact. A rule of law continues in abstraction and theory, until an act is done on which it can attach and assume as it were a body and shape”.
  • Focus on technical facts, more than on economic/accounting (though see sections on remedies, and on interaction of technical & economics experts)
  • Using claim construction as a tool, not simply providing the usual rehash of the “canons” of claim construction
  • Who cares? Why aren’t we talking about weighty issues like “patent reform” and subject-matter eligibility, and instead focusing on seemingly-little things like claim charts?; in part because there are simpler procedural fixes to perceived patent problems (such as unpredictability, or beliefs that patents are “too strong” or “too weak”); Local Patent Rules (LPRs) are more tractable tools for controlling “patent abuse” than is 101.
  • All these minor-looking details — such as how evidence of infringement or non-infringement are marshalled, how expert reports should be structured, how patent valuation relates to individual patent claims — might be viewed as: “this is how the lofty-sounding ‘knowledge economy’ looks on the ground.”
  • A theme throughout the book is tensions in the patent system, including:
    • broad/narrow
    • functional/structural
    • certainty/ambiguity (“definite & certain claims”, yet wiggle room)
    • protection/disclosure (“the name of the game is the claim”, yet “you get what you disclose”)
    • national/international (protection is per-country, but disclosure is worldwide)
    • past/future (see “Why let them do this?” in Chapters 1 and 2 below)
    • exclusion/diffusion (see non-rhetorical question in Chapter 2: “how can anyone think that a system of exclusion will promote diffusion?”)
  • A related theme is the necessarily contradictory nature of the patent system: my notes contained the phrase “schizo mishmash” I had casually tossed off; one of the AI RAG systems I’m using saw this phrase and really ran with it, in a way that sounded like a mere critique of the patent system, rather than as a possibly necessary attribute of any functioning patent system (see Chapter 2).
  • Shorthand: P (plaintiff, or patent owner) v. D (defendant, accused infringer)

Chapter 1: “$1 Billion … for That?!” (Caltech v. Apple/Broadcom) and “Why Not Sue Over This?” (Google AI Transformer Patents Not Asserted Against ChatGPT)

  • Opening lengthy example of Caltech v. Apple/Broadcom: how do a few patent claims turn into a billion-dollar judgment (albeit one in which the damages amount was appealed)?
    • The patents at issue (US 7,116,710; 7,421,032; 7,916,781; and 8,284,833)
    • Staking out claims in 2001, using in 2016 (“loading/buying the gun in 2001, pulling the trigger in 2016, getting paid in ____?”) — throughout book, we’ll ask in different ways “Why let them do this?”, why allow past inventions to “tax” future uses?
    • Focus includes 7,421,032 somewhat unusual Claim 11 with Tanner graph within claim
    • Irregular Repeat–Accumulate Codes (IRA), a type of Low-Density Parity-Check Code (LDPC; [Gallagher 1963])
    • How would a chip or mobile phone “use” a patent?
    • WiFi 802.11ac includes LDPC option: section 20.3.11.6 (“LDPC codes”)
    • Infringement by implementing (in a certain way) an optional portion of a standard?
    • Patent infringement is specific and boring-looking (“in these two lines of the source code, D embodies element [1d] of our patent claim…”), not general and hand-waving (“they boosted our stuff”) — the boring details might be seen as mere “tickets of admission” to the main event of closing argument before a jury, but the case will be occupied by more of the former than the latter
    • What low-level details would show infringement?
    • A patent claim is a map or checklist of what infringement would “look like”:
      • Claiming a method or process (a series of steps)
      • Claiming an apparatus or device (an inter-connected list of components)
    • Patent claim “limitations”, and the claim as a parts list
    • Caltech’s infringement evidence; experts
    • Caltech claim charts unavailable; trying to infer claim charts with AIs, and with GreyB consulting company in India/Singapore
      • [TODO: use GreyB GPS beta to generate claim charts for ‘032 v. 802.11ac 20.3.11.6 — GPS now supports comparison of claims with text in PDF files]
    • [TODO: need brief discussion up-front of types of infringement, pointing out which ones asserted or not in this case: making, using, selling, offering for sale, importing, inducement, contributory]
    • $1B damages; including implications for remedies/royalties/damages of government funding for the underlying research
    • Caltech is non-practicing entity (NPE); university plaintiffs as “trolls”?
    • Apple and Broadcom’s invalidity evidence; prior art from conferences, etc.
    • Claim construction: “repeating”, “irregularly”, “random”; relation to earlier claim construction in Caltech v. Hughes Network
  • And contrasting example on why Google didn’t assert its Transformer patent against OpenAI’s ChatGPT
    • Common narrative: Google invented the Transformer architecture (Vaswani 2017 paper, “Attention Is All Your Need”), and even had a related patent (US 10,452,978; Shazeer et al., filed June 2018, grant Oct. 2019)  but it was OpenAI that ran with it: ChatGPT in Nov. 2022
    • Show discussions of this (including why Google didn’t sue OpenAI) with Google AIs (NotebookLM and Gemini) and with ChatGPT-5; especially useful chat with Perplexity.ai on Transformer patents.
    • AI chatbots agree that ‘978 Claim 1 requires encoder-decoder, whereas GPT-3.5 (basis for ChatGPT) decoder-only
    • Also see Enablement/Disclosure Chapter XXX below for comparison of ‘978 with [Vaswani 2017]
    • Initial analysis with Perplexity.ai Patents of Google decoder-only patent US 11,556,786 Claim 1 (Shazeer et al., filed Oct. 2018, grant Jan. 2023)
      • See earlier OpenAI GPT-1 paper (June 2018; Radford et al.), which predates filing of ‘786 — but OpenAI’s own ChatGPT-5 pointed out that relevant priority date for ‘786 is Oct. 2017; Perplexity.ai does preliminary examination of provisional 62/578,358 written-description support of ‘786 Claim 1.
      • Temporarily ignoring ‘786 Oct. 2017 priority date, see Perplexity.ai claim charts of ‘786 Claim vs. GPT-1 paper at end of this chat
    • Initial claim chart generated by GreyB GPS (GlobalPatentSearch.ai) beta of ‘786 Claim 1 vs. GPT-1 paper [TODO: carefully check Perplexity and GreyB claim charts]
    • Google Gemini on how Google’s ‘978 encoder-decoder does not read on decoder-onlyChatGPT, and ChatGPT on how OpenAI’s GPT-1 paper does not anticipate earlier priority date of ‘786: Will IP litigation involving AI companies look at AI chatbot output as “admissions against interest“?
    • Using file wrapper to see which prior art was before the examiner (e.g. Google Transformer ‘978 file wrapper, June 2018-Oct. 2019)

Chapter 2: “Promoting Progress of Science and the Useful Arts”: How Does Patent Litigation (and the Patent System Generally) Fit In?

  • Patent litigation and patent claims: Why should we care?
  • US Constitution goal: “promote progress of science and the useful arts”; what that phrase meant in the 18th century, in particular why and in what particular ways the Founders expected patents to promote progress; the following from Dutton, The Patent System and Inventive Activity During the Industrial Revolution, 1750-1852; and Walterscheid, The Nature of the Intellectual Property Clause: A Study in Historical Perspective:
    • Reward by monopoly (Adam Smith, Lectures of Jurisprudence)
    • Monopoly-profits incentive (similar to Smith, but utilitarian, focused on social benefit rather than individual reward)
    • Exchange for secrets; centrality of disclosure (Liardet v. Johnson, 1778)
    • Protection of investment (Alexander Hamilton, Report on Manufactures, 1791)
    • Natural-law right of inventors (France 1791 statute), rejected by Jefferson et al.?
  • Briefly contrast 18C “science”/copyright and “useful arts”/patent [TODO: must incentive theory covered here for patents also apply to copyright which is commingled in the same Constitutional clause?]
  • Does the patent system (and patent litigation and patent claims in particular) function as an incentive to R&D, to innovation and to economic growth generally, particularly through tech disclosures that otherwise wouldn’t occur?
  • Whatever the Y/N answer to that question, what do the patent system, patent litigation, and patent claims DO?
    • What is social impact of different parts of patent?: specification vs. claims
    • Impact/incentives on WHO?: not just inventor; incentives also created for potential defendants, and for investors as well as inventors
    • Incentives created by patent system/litigation: being incentivized to do WHAT?
  • Patents are future-oriented: e.g. Caltech case: in 2001 staking claims over much later activity: “Why let them do this?”, why allow the past a “tax” on the future?
  • NOT a rhetorical question: “How could anyone think that restriction -> diffusion?”
    • Part of the reason is simple: patent system is “leaky”, in that large % of patent applications not granted, but applications still published, and used as prior art; large % of granted patents not renewed; most patents not litigated; patents disclosed internationally but only territorial protection (see Chapter XXX)
    • More important reason is that the restrictions and requirements imposed by patent system have a channeling function, pushing inventive activity in certain directions (see below)
  • Large effects (both private and social) from small details
  • [TODO: some here in Chapter 2 on the necessarily contradictory nature of the patent system, covered mostly in Conclusion chapter]
  • The patent system viewed as a lottery, treadmill, or marathon — some self-referential or self-perpetuating quality, with patent system incentivizing more patents without necessarily incentivizing invention, but “treadmill” and “routinizing” quality also likely promotes progress, as noted below:
  • Routinizing inventions: regularly expect about 6,000 US patents per week — the book’s Conclusion will cover relation of economics work by Romer, Mokyr and others to the “regularizing” and “routinizing” function of the patent system; briefly here:
  • Patent system urging/channeling inventors in several different directions at once:
    • thinking big (generalize from what inventor did to largest possible claim scope)
    • + thinking small (distinguish from prior art; focus on what is the point of novelty [PON]?)
    • + thinking ahead (what will the Monopoly board look like in 5 years?)
  • The centrality of patent claims and “limitations” — “The name of the game is the claim limitation“
  • Fine-grained disaggregation and parsing, not a “holistic” approach (with a few exceptions)
  • Boundary-drawing in a virtual space — some analogy to machine learning?
  • Comparisons, matching (larger match from series of smaller), “forensics,” bright lines, penumbras, scarecrows, howitzers, and sheep in wolves’ clothing [TODO: explain each catchphrase such as scarecrow, howitzer, etc.; since a “penumbra” is a shadow, it is perhaps not the best word to describe how a patent claim can encompass a slightly larger region than is apparent from its text]
  • Patent litigation is both rare and crucial to the patent system: The need to kick the can down the road, or “rational ignorance”
    • Though note new emphasis under Trump administration PTO/PTAB on “settled expectations”
  • A somewhat odd form of property: inherently-unstable templates? [TODO: is “unstable” the right word to describe somewhat uncertain scope and validity of patent claims?; and “odd” also reflects inherently contradictory nature of patent system, as described in Conclusion]
  • What does society get for all this?:
    • patent incentives to “promote progress of … the useful arts” (not so much “the useful arts” themselves; more about a particular type of “progress”)
    • disclosure vs. secrecy;
    • the value of representing inventions as claims; [TODO: make sure this value is adequately covered in Chapter 3]
    • how could litigation over patents possibly “promote progress”? (again, not a rhetorical question)
  • Patents as incentive to avoid trade secrets (TS): it is TS, not vague “the commons”, that is the real-world alternative to patents
  • Incentives: on who? where? how?
  • Relation of patents to risky R&D; distinguish R (research) vs. D (development)
    • Economists: “windfall” awards as compensation for “dead-end” research?;
    • Inventors that couldn’t commercialize own invention: take Musk “patents are for losers” literally?
  • Patent litigation & the “knowledge economy” (avoid facile AI-generated statements like “patents are the engine [or drive train or …] of the knowledge economy — but there is something to say here)
  • Is all this complexity necessary? Could the patent system be simplified to avoid even more litigation?
  • Let’s Make Patents Boring Again: procedure (e.g. something as petty-sounding as enforcing claim chart requirements) to cut through substantive difficulties
  • If AI is “the last invention man need make” (per IJ Good 1965), will we still need patent-system incentives?
  • How do patents compare to alternative IP and non-IP incentives for R&D?
  • COVID-19 vaccines as an example of the positive, negative, or neutral role of patents in both creating and distributing vaccines

Chapter 3: Patent Infringement, Claims, and Limitations

  • Patents: a negative right to exclude others from making/using invention (NOT a license for patent owner to make/use invention)
  • Patents: enforceable boundaries around inventions
  • Patents have boundaries: what is property?; is IP really “property” like real estate?
  • Relation of patents to “IP as a whole” (is there such a thing?; or a collection of separate regimes?)
  • How patents relate to the patent owner’s products:
    • what does it mean for a patent to “cover” a product marketed by the patent owner?;
    • not a “picture patent”: generally not a verbatim textual transcription of what is in a product; instead, an abstraction from what appears in a product, and from what inventors did
    • patent owner may not have a product at all: why no “working requirement”?
  • Infringement, and the right to exclude
  • Patents are not self-acting, not self-enforcing, and the “patent police” are not on their way; this is why there is patent litigation (and even non-litigation occurs “in the shadows of” litigation)
  • Crucial role of litigation in the patent system, including to test the true scope of patents (beyond necessarily brief PTO testing)
  • Patent litigation in federal civil court (experienced attorneys can skip this section)
    • Venue & forum shopping
    • Briefly here on other “venues”; ITC; PTAB
    • Trial, and the role of judge & jury; need for lay opinion even on deep technical issues (how does the case “smell” to the community?)
    • Most litigation is pretrial, and most cases don’t go to trial (see section on settlement)
    • Appeals and the CAFC
  • Showing infringement
    • Earlier said “making/using”; actually: making, using, selling, offering for sale, or importing
    • Briefly here on indirect infringement (more later on contributory infringement & inducement, including “knowing” and “intent” issues normally irrelevant in strict-liability patent infringement — except willful infringement with treble damages)
  • Patent infringement is based on patent claims and limitations
    • Briefly here on design and plant patents, without claims/limitations
    • Some parts of claims that look like limitations are actually non-limiting (preconditions, most preambles)
    • Note assertions that “the name of the game” is not the claim anymore, e.g. instead “you get what you disclose” (Root) or trial prep requires a thoroughly “holistic” approach throughout the case?
  • The value (both to patent owners and to society) of representing inventions as claims
    • As noted above, a patent claim is generally an abstraction, not a “picture patent” that exactly duplicates what inventors did
    • Why would we want that?; contrast tech imperative to “just do it and move on”, “move fast and break things” [TODO: add more from notes]

Chapter 4: A Small Example: Printer Toner

  • A sample patent claim (for printer toner)
    • Fuji Xerox patent (US 8,951,704), related to printing
    • Dependent claims, and their uses
  • A sample method claim (for an image-forming method)
  • Other types of patent claims: CBM, system, CRM (media), sequences/structures, design patents
  • General distinctions: function vs. structure; device/apparatus vs. process/method, etc.
  • Patent claims are made of “limitations”
    • Limitations are elements/components/parts of devices, or steps of methods/processes
    • Briefly here on sub-limitations (much more in Chapter 17  on parsing lengthy/complex/multi-part claim limitations)
    • Almost a parts list or BOM (bill of materials), but not quite
  • How a selective “parts list” effectively marks a patent boundary
  • Patent claims are devices/tools for testing infringement (and for testing their own validity/invalidity over prior art — NOT a complete “recipe” for making/using the invention as such) [interesting that AI chatbots regularly assert the “complete recipe” fallacy]
  • Claim should enable one to “say what’s in, AND what’s out” (see “operationalism”?)
    • Possibly contrast patent claim approach (similarity to rules in GOFAI symbolic AI) vs. how machine learning classifies after training on a large number of imperfect examples — see PatClaim.ai on how machine learning can nonetheless be helpful when working with patent claims
  • Claim scope’s inverse relation to size/complexity: Less is More
  • “The name of the game is the claim” ==> “The name of the game is the claim limitation“
  • Briefly here on role of claim construction in patent litigation (mostly in Chapter XXX below)

Chapter 5: An Extended Example: “Facade Server” Patent

  • (This section analyzes a claim, not as an abstract claim-construction exercise, but also without a specific target product [which in Chapter 7], but somewhere in the middle.)
  • Treating a patent claim as a parts list…
  • … But not just a parts list
  • What’s in a name?: infringement & nomenclature/terminology
  • A simple software patent claim (the book’s major example: “facade server”, and how/why it was selected) — US 7,472,398, granted to Hewlett Packard (HP) in 2008; NOT part of large HP portfolio sold to Qualcomm ~2014?
  • Walking through claim limitations
  • “Facade server”: interpreting terms in a patent claim (specific example of using claim construction)
  • The “core” of the claim (or at least what parties fight hardest over) may not look like much
    • How patent claims change while the PTO is examining the patent application; claim evolution; PON (“point of  novelty”)

Chapter 6: The Role of Claim Construction in Patent Litigation

  • Below, will sometimes seem to refer to passively “understanding” or “interpreting” claims as such, but purpose of claim construction is as a tool, to guide use of claims in disputes over infringement, non-infringement, invalidity and validity.
  • “Words, words, words”: the importance of nomenclature, terminology
    • even though engineers tend to view this as “lawyer games” or “angels dancing on the head of a pin”;
    • patent system has chosen words/language as means to create actionable descriptions of inventions;
    • engineers tend to think in drawings more than in words, and of course the patent system has drawings, but words are generally how the property lines are drawn (some exceptions: design patents; chemical structures; example of Tanner graph in Caltech patent in Chapter 1)
  • “Literal” infringement doesn’t require verbatim identicality between claim language and e.g. product literature or source code
    • e.g. a so-called “gizmo” may literally infringe on what the patent claim calls a “widget”
  • Lack of standardized terminology:
    • inconsistent meanings across patents from different owners, and even from same owner, or in same patent
    • patent owner can be “own lexicographer”: make up terms, or use existing terms in idiosyncratic way; then need to use specification as dictionary
  • The (rebuttable) presumption of consistency in claim terminology meaning
    • consistency of interpretation across claims, across P’s multiple patents, and multiple cases, would be one claim-construction factor?
  • “Plain and ordinary meaning” is the default,
    • But this means P&OM to the PHOSITA (person having ordinary skill in the art) having just read the specification (this is one reason why claims come at the very end of the patent doc, after the specifications?)
  • Rule is to first do claim construction (CC), and only then apply to accused product (Relation of CC to the accused device)
    • BUT in practice, it’s done the other way around: litigants first figure out which claim terms they care about, and then they dispute the meaning of those particular terms; not every term in claim is construed (analogy to © practice?)
  • Hierarchy of claim construction sources; intrinsic vs. extrinsic evidence for claim construction:
    • of course, first the specification for the patent
    • including file wrapper as intrinsic evidence
    • dictionaries and experts are extrinsic evidence
  • Relationship of claims to the rest of the patent
    • context/definition, but not “importing” from spec into claim;
    • PHOSITA (“person having ordinary skill in the art”) reads claims, at the end of the patent, in the context of what came before (non-claims portion: specification, drawings, cited prior art, even the “file wrapper”)
  • Markman hearing: often dispositive of case
    • parties may try post-Markman to shift their infringement/invalidity positions, but courts try to curb “shifting sands” and “musical chairs” in post-Markman revised claim charts — see Chapter XXX on Claim Charts
  • Using the other side’s implied claim construction against it
    • party’s claim construction implied from its infringement posture;
    • simplest example: P says X in D’s product matches this claim limitation for purposes of infringement analysis, so D can assert that same X in the prior art matches the same limitation in P’s patent claim)
  • Claim differentiation:
    • using children (dependent claims) to understand their parent (independent claim);
    • using small differences between claims, or between claim limitations
  • Using (but not “importing”) non-limiting examples from the specification
    • using examples and embodiments in the specification, as (usually non-limiting but nonetheless revealing) guides;
    • proper nouns in the spec are especially helpful in searching (sometimes, though rarely, proper nouns right in the claim)
  • Don’t rush past non-technical words, word endings, or seemingly-vague words
    • e.g. “whereby” vs. “wherein” (which are terms of art), or
    • e.g. “based on”
  • The importance of word suffixes: -ed vs. -ing vs. -able
  • Using adjectives to understand claim terms
  • Preambles and other generally non-limiting claim elements
    • e.g. preconditions (required to be present for the claim to work, but need not be done by an infringer)
  • Equivalence claim construction — see Chapter XXX
  • Means-plus-function claim construction — see Chapter XXX
  • “Canons” of claim construction
    • “for each and every canon, there is an equal and opposite canon”;
    • using claim-construction “canons” as tools in patent litigation
  • “Careful what you ask for, you just might get it”
    • why P should be careful asking for broad claim construction;
    • “that which infringes if after, anticipates if before”, so broad scope of infringement also means broad scope of prior art for invalidity
  • Threading the needle: how broad and narrow claim/limitation scope relate to plaintiff’s and defendant’s postures for infringement and invalidity
  • Remember to USE the claim construction; working with multiple potential claim constructions
    • having fought hard for a particular claim construction, remember to actually USE it in analyzing infringement and invalidity;
    • amazing how often a party’s claim charts reflect raw unconstrued claim language, rather than employ hard-won favorable claim construction
  • Working with multiple potential claim constructions — see Chapter XXX on Claim Charts, and Chapter XXX on Multiplicity
  • Relation of claim construction in one case to claim construction of same patent in other cases with different defendants (see also RJ, non-mutual CE?)
  • What is tech legitimacy, outside patent context, to the way that technical descriptions and terminology are handled in patent system?; any worthwhile methodology for technical analysis?
    • e.g. determining the compliance of a product with a technical standard

Chapter 7: Literal Infringement Analysis: Comparison of a Patent Claim with an Accused Product or Service

  • Investigating infringement of a software patent claim
  • Brainstorming search terms (nomenclature, terminology, synonyms, non-limiting examples); using LLM semantic search
  • Using specifics (often non-limiting examples or embodiments) from the spec: proper nouns, product names, model/part numbers
  • Example of a possibly-infringing product
  • Limitation-by-limitation comparison of a patent claim with an accused product
    • Parsing claims into limitations (“beyond the semicolon”)
    • “All-Elements Rule”
    • For each limitation, show SPAFF: at least one “structure, process, algorithm, feature or function” of the accused product
  • Structuring the comparison of a single claim limitation with a product attribute/feature/element
    • Parsing lengthy/complex limitations into sub-parts for analysis
    • Sub-limitations, both explicit and implicit
    • Deceptively simple-looking limitations (e.g. “a facade server”; when patentee “acts as his own lexicographer”)
    • Features/attributes/adjectives of limitations
    • Negative limitations, and absences (e.g. with closed “consisting” preambles)
    • Claim limitations with choices
    • Contrast patent claim up-front explicit features/limitations with AI finding implicit features (also contrast with TS litigation often involving specific TSes only described after the fact of employees leaving company)
  • P’s difficulties showing infringement & D’s ability to show non-infringement
  • An example infringement analysis concluding “NO”, and other examples of possibly-infringing products
    • This section includes research by GreyB, as a detailed example of a real-world patent-infringement search);
    • AI-based search would use semantic similarity/difference with less emphasis on specific keywords (but ultimately must come back to claim language, as construed by the court)

Chapter 8: Equivalence, Function/Way/Result, and Interchangeable Parts Analysis

  • Equivalence is an important example of what a “knowledge economy” “looks like on the ground”
  • The function/way/result (F/W/R) test for equivalence
  • The “insubstantial differences” test (interchangeability); “parts is parts”?
  • Equivalence is per-limitation, not holistic
  • Handling equivalence in claim charts: show WHY equivalent by separately showing substantially similar (1) function (purpose, role), (2) way (implementation), and (3) result (output)
  • Restrictions on using equivalence: how important is PHE?
  • Underlying policy for the doctrine of equivalence (possibly basis is more history than logic?)
  • Relation of equivalence to other areas of patent law (can it non-superficially be viewed as a sort-of mirror image to obviousness?; note that equivalence does not have “invention as a whole” perspective; see obviousness Chapter XXX below)
  • Importance of equivalence in definition of patents as property (patent owner possesses area as claimed, plus a bit more: reasonably equivalent technology, even if “after-arising“)
  • Relation of equivalence to the “notice” requirement (patent strict liability & notice)
  • [Doesn’t quite go here:] Design-around, and its relationships with equivalence (including PHE) and with non-infringing alternatives (NIA)

Chapter 9: “Means for…”: So-Called “Functional Claiming”

  • “Means for”: so-called functional claiming (need to find structural details in the specification)
  • “Means for…” in claim += the means themselves are in the specification (LPRs require SAM: “identity of the structure(s), act(s), or material(s) in the Accused Instrumentality that performs the claimed function”)
  • No such thing as “functional claiming”
  • When means+function governs/operates (other triggers besides “means for…”)
  • Means+function is per-limitation, not per-claim
  • When means+function applies, one needs to actually USE it (e.g., claims chart for “means for…” limitation must employ specific structure(s) from the spec)
  • Relation of means+function to other areas of patent law
  • Underlying policy for allowing means+function claim limitations?
  • Note patent example in important older AI book: Herbert Simon, Sciences of the Artificial sections on “Functional Description and Synthesis” and “Limits of Adaptation”

Chapter 10: Defending a Patent Infringement Case

  • (Introduction to this, and the next several chapters on Anticipation, Obviousness and Enablement)
  • Defendant’s options include:
    • Showing non-infringement (show absence of 1+ elements, ideally from each & every asserted claim)
      • Flipping around the All-Elements Rule for non-infringement: only need one missing claim limitation
    • Counter-claims (D shows P infringes D’s patents)
    • Briefly here on design-around and NIA (non-infringing alternatives); mostly in Chapter XXX on damages
    • Showing invalidity:
      • anticipation, obviousness, lack of enablement; but: presumption of validity; need clear & convincing evidence;
      • see individual Chapters 11 (mostly on Anticipation), 12 (Obviousness), and 13 (Enablement & Disclosure) below
    • Inter-partes review (IPR) at PTAB — challenging patent validity becoming more difficult under Trump administration
    • Claim construction disputes; balancing narrow construction on infringement & broad construction on invalidity
    • Asserting on-sale & public use bars: P may have tripped itself up with its own use of the invention, more than one year earlier
    • D’s own prior use (presumably secret, otherwise would be prior art)
    • Asserting 101 subject matter (Alice, AI/ML, business methods), AI inventorship
    • Motions to compel, dismiss, exclude, preclude, strike, stay re: P’s pleadings, contentions, experts, evidence, etc.
    • Summary judgment (SJ), including “failure of proof” (Celotex)
    • Declaratory judgment (DJ) actions
    • Reducing damages: royalty rate, royalty base, lost profits, non-infringing alternatives
    • Settlement; grudging license (losing litigation may result in involuntary license)
  • Unstable property: unpredictability, and the risk of patent invalidity
    • Reader may ask why “instability” of P’s patent is covered here in chapter on D (defense), but D poses the major threat to the stability of P’s patent
      • D (defendant) is arguably as crucial to the patent system as P (plaintiff): patent system needs a pool of Ds to weed out “bad patents,” and to keep patent scope (infringement assertions) within reasonable bounds
      • Risk to P’s patent of exercising it against D: Possibly “Use It AND Lose It”
      • Becoming a defendant in a patent infringement case motivates the defendant to try to invalidate the patent
      • Patent system needs highly motivated Ds to challenge patents; this sounds perverse or Panglossian, but given the PTO’s limited time to examine patents, handing a “dollop of state power” to P (see Chapter 2) in effect uses P to incentivize D to challenge patents
    • Note changes under Trump administration, including new focus on “settled expectations,” PTO “one and done” proposal (characterized as “where patents that survive any initial validity challenge – regardless of its quality or completeness – becomes virtually immune from subsequent IPR proceedings”)

Chapter 11: Patent Invalidity: Pulling the Rug Out from Under Plaintiff’s Property

  • Chapter focus mostly on Anticipation, but first some general points on Invalidity
  • Using a Patent Claim to Look for Invalidity, and as a Lens Through Which to View Prior Art
  • Patent invalidity: how D can pull the rug out from under P’s property
    • (rebuttable) presumption of validity; need clear & convincing evidence, more than preponderance of the evidence needed to show infringement;
    • “That which infringes, if after, anticipates if before”;
    • how dependent claims intersect with infringement & invalidity;
    • bottom line: no “reinventing the wheel”, but encouraging inventing slightly different (but non-obvious) wheels
  • Investigating prior art for a software patent claim
  • How a patent claim is compared with prior art: similar to infringement comparison, except:
    • Clear & convincing evidence > preponderance of the evidence
    • If all limitations cannot be found in a single reference, possible fall-back to combining references; see obviousness, Chapter XX
  • Prior art already examined by the PTO; reading the file wrapper
  • Searching for prior art:
    • both keyword approach, and LLM semantic search (e.g. Perplexity.ai);
    • how PTO is using AI (including to search drawings — though mostly for design patents)
  • Searching for a negative limitation (required absence) in prior art
  • What qualifies as prior art: foreign, NPL (non-patent literature) theses, real-world practice with limited documentation, edge cases
  • An extended prior-art example
  • Limitation-by-limitation analysis of a prior-art reference; as with infringement, it all comes down to individual limitations
  • Other prior-art examples (this section includes research by Patspade IP as a further detailed example of a real-world prior-art search)

Chapter 12: Obviousness — Both “Crown Jewel” and Mere Fallback

  • Both a mere fall-back position from anticipation, AND the “crown jewel” of the patent system
  • Combining multiple prior-art references, when a single reference does not fully anticipate (primary/secondary references); does combination of many references start to look non-obvious? (at least at PTO, apparently no)
  • “Motivation to combine”; TSM+
  • “Prior art” for obviousness can include PHOSITA’s common knowledge, beyond particular references
  • The PHOSITA’s perspective:
    • PHOSITA presumed to know all relevant prior art (Winslow tableau), but only within field and analogous/relevant/pertinent fields, which would have been obvious to consult at time by one working on same problem (PC clamshell case);
    • prior art pertinent if same field, regardless of problem to be solved, or if reasonably related to particular problem with which inventor was concerned;
    • can AI LLM stand in for the PHOSITA role of ordinary-skilled, not-inventive person having complete knowledge of the relevant prior art?
  • “Teaching away”
  • Some “holism” in obviousness analysis (“the invention as a whole” in 35 USC 103; but NOT the “gist” or “thrust” of the invention)
  • Obviousness in claim charts: handling combination of multiple references — see Chapter XXX on claim charts below
  • Secondary considerations: are they really secondary?
  • Underlying policies, including avoiding hindsight bias
  • Economists (e.g. Suzanne Scotchmer, Innovation and Incentives, and Pia Weiss, Patent Policy) on the role of obviousness in urging inventors away from minor contrived distinctions from the prior art (see Chapter 2 above)
  • Relation of obviousness in definition of patents as property (ability to invalidate patent by later combining multiple earlier prior-art references that were only separately accessible at the relevant time)

Chapter 13: Enablement and the Underlying Purpose of the Patent System: Do Patents Disclose Useful New Information?

  • Lack of enablement as another way to invalidate patents
  • Crucial to the patent “contract”: P gives disclosure in exchange for temporary property over claims; enablement and the underlying purpose of the patent system
  • Enablement basics from 35 USC 112(a)
  • What types of information must be disclosed? What need not be disclosed?
    • AI-generated summaries frequently exaggerate: ‘inventor must disclose everything,” whereas need only disclose minimum necessary for PHOSITA to make/use the invention without “undue experimentation” (so some experimentation not undue)
    • Issue with enablement of AI/ML inventions?; but “black box” okay since disclosing how to make/use doesn’t require understanding why it works
  • Enabling the full scope of the claim, including (but not limited to) the claim as P construes/asserts in litigation
  • Enablement also required for prior art (except patent owner’s own on-sale, public-use); science fiction unlikely to be prior art
  • If “the name of the game is the claim,” why do we care about the spec?; the spec is the price paid to play “the game”
  • Is P required to provide NEW info?; no: while patent claims must be novel, it is common to have multiple patents sharing the same identical spec, so long as the older spec still enables the new claims
  • Does the patent system truly treat disclosure as its ultimate goal?; and is so, what happened to “best mode” — isn’t that something we would really want to be disclosed?
  • Codification, tacit knowledge, “sticky knowledge”, trade secrets: Michael Polanyi objection to patents based in part on “sticky” nature of tacit knowledge
  • Do patents actually disclose new information that would be useful to engineers? (see Ouellette, “Do Patents Disclose Useful Information?“)
  • And, are patent disclosures used by engineers for R&D, non-patent purposes? (see survey in Ouellette paper)
  • Consider the possibility that patent disclosures exist largely to supply prior art for the patent system itself, such that disclosures are more self-referential (for the system’s own benefit) than truly enabling the PHOSITA engineer. Even if true, this would still be reasonable and necessary to prevent future non-inventive patents?
  • Patents vs. trade secrets
  • The “ostrich position” (company policy to deliberately avoid reading patents), willful infringement, and indirect infringement
  • How do academics, businesses, and governments use patents?; e.g. using patents to identify geographical clusters, importance of location
  • Comparison of Google Transformer patent (US 10,452,978) with [Vaswani 2017] NeurIPS paper — as disclosure, and as dated asset:
    • What does one do that the other doesn’t?;
    • ‘978 result of xx hours @ PTO (see file wrapper cited above) + prior art + subsequent citations-to
    • [Vaswani 2017] @ NeurIPS: peer review; subsequently cited by xx (most-cited academic paper ever)?
    • Umm, any question which one directly resulted in more promotion of “progress of the useful arts”?

Chapter 14: Remedies, Damages, and Patent Valuation “Show Me the $”)

  • Remedies (ideally, this would be towards the front, not the back, of the book, since “everything works backwards from damages”)
  • Injunctions and TRO; eBay v. MercExchange and NPEs; equity rules
  • Reasonable royalties
    • Georgia-Pacific factors;
    • importance of royalty base as well as royalty rate;
    • no more 25% “rule of thumb” (see Experts chapter XXX);
    • “comps”; “Zillow for patents”?
  • Royalties and standards: SSOs, SEP, and FRAND
  • Remedies when the patent covers only part of the product: apportionment, “stacking”, EMV, SSPPU (smallest saleable patent-practicing unit)
  • Treble damages and willful infringement (knowledge/intent)
  • Lost profits, Panduit factors, including absence of non-infringement alternatives (NIAs)
  • Statute of limitations (SOL), 6 years
  • Settlement, licensing, and monetary value of a patent case
  • Patent valuation: a claims-based approach (“Doh, what else would it be based on?”)
    • How valuation relates to damages
    • Valuation is based on the patent, generally NOT on the underlying technology
    • Cost, market, and income valuation methods
    • Patent value is likely context-specific, not entirely within the four corners of the patent
    • Potential invalidity’s impact on patent value
    • Patents as options?
    • Oddly, patent valuations almost never reference the claims: Using what proxies for actual value?
    • Patent value & claims & claim limitations: Some simple rules
    • The impact on valuation of the All Elements rule, design-around, PON, and next-best comparison
    • Valuation & patent portfolios
    • Using abandonment of patent to measure max value
    • Relative ranking of patents within a portfolio, and pegging to known dollar values from abandoned patents, and from licensing
    • Experts & patent damages/valuation: see Chapter XXX
    • Contrast purely market-based value (“market price is by definition correct”); pricing failures, bubbles, “greater fool” — damages in context of litigation vs. valuation generally: can see value in context of this specific defendant
    • Difficulty of pricing “something new under the sun” — but still value related to next-best alternative
    • Trump PTO proposal to price maintenance fees at % patent value — how would measure?
    • Any social interest in $ value of patents?; of course matters to P & D in litigation, but importance beyond the parties?; why society cares about ability to attach $ value to patent claim when compared to products on the market
    • [… etc. …; TODO: add more from notes]
  • ITC non-importation order, and “domestic industry” (working/practicing requirement)
  • Interaction of technical experts with economics/accounting experts (e.g. tech expert may provide econ expert with list of part numbers to include/exclude in damages calculation spreadsheets) — see Chapter XXX
  • Attorneys fees, American Rule
  • Litigation funding (Bentham, et al.)
  • Patent infringement insurance (for potential defendants; any insurance for patent invalidity?)

Chapter 15: Pre-Filing Investigation, “Informal Discovery,” and Monitoring the Marketplace

  • Monitoring the marketplace for patent infringement, e.g. using potential defendant’s own patent filings as indicia
  • Why is there no “patent police”?; why is monitoring patent infringement different from what ASCAP/BMI do for music copyright?; do so-called “trolls” (or litigation funders) perhaps function as quasi “patent collection societies”?; see Merges on patent owner given a “dollop of state power”
  • Which patents or claims to assert from a larger portfolio?; portfolio ranking; see Chapter XXX on Multiplicity
  • Who to assert patents against?; finding potential/candidate targets (including within chain of commerce)
  • Which particular products to accuse, based e.g. on 6 years SOL?
  • Difficulty detecting much patent infringement: contrast e.g. trademark infringement which apparent; but not as difficult as detecting trade secret infringement?
  • [TODO: below, add links to relevant portions of Claim Charts book draft]
  • Reasonable pre-filing investigation under Rule 11, Iqbal/Twombly (“plausibility”), and Local Patent Rules (LPRs)
  • Legal and ethical issues in “informal discovery” (“IP, PI”)
  • “Acquire the target”: Difficulties in acquiring access to possibly-infringing products/services/in-house methods (“instrumentalities”)
  • Reverse engineering (or its equivalent) as a litigation tool for detecting and assessing infringement; “product forensics” — note reverse engineering (and source-code examination) changes for AI; hardware reverse engineering
  • What to do when detecting infringement is difficult before discovery
  • “Information and belief” (I&B) & difficulties learning about products; I&B over-use when publicly-accessible detailed information has not been diligently mined; preliminary/partial reverse engineering as a basis for legitimate I&B
  • Pre-filing enforcement, the demand letter, and declaratory judgment (DJ) (see excellent Martens/Sganga Pre-litigation Patent Enforcement)

Chapter 16: Bringing (and Defending) a Patent Infringement Suit: Pros, Cons, and Plausibility

  • Patent litigation is often a Bad Idea: “War” is a last resort
  • Some seemingly-contradictory general points about licensing vs. litigating (a false dichotomy)
  • Private goals of patent litigation
  • Shield: “Protect our product”, “protect our patent”, and the “right to exclude”; try to create a moat around product
  • Sword: Patent litigation to make (or recover) money; monetization
  • Other goals: Retaliation, harm, deterrence, PR, “weaponization”
  • The major de facto goal: Leverage for settlement, licensing (“licensing vs. litigation” a false dichotomy)
  • Cost-benefit analysis of bringing a patent litigation case
  • Attorney fees, the American Rule, contingency fees, and litigation funding
  • [TODO: need more on “trolls”, non-practicing entities]
  • Checklist of necessary but insufficient preconditions for litigation
  • WHEN to sue, in relation to patent life (early war-chest vs. end-of-life; which best 6-year period?), and in relation to D product’s life cycle (sunk costs)
  • Risks of suing for patent litigation: “use it and risk losing it” (P motivates D to work to invalidate P’s patent)
  • Defendant’s choices: Why ever not settle or license? (good & bad reasons)
  • Attorney and expert choices
  • A plausible case — social importance of “plausibility” under Iqbal/Twombly; relation to Rule 11, Local Patent Rules (LPRs); see Chapter XXX above on reasonable pre-filing investigation
  • Briefly on civpro issues like venue, jurisdiction here
  • Summary judgment (SJ) & “failure of proof” [TODO: does this go here?]
  • [TODO: what is relation of this chapter to Chapter XXX below on Strategy/Tactics?; that chapter more on strategy/tactics after litigation has started?]

Chapter 17: Claim Charts and the Local Patent Rules (LPRs)

  • [TODO: YUCK, outline below does a poor job summarizing the book’s extensive projected chapter on claim charts]
  • Claim Charts under LPRs, but also under Rule 11, and Iqbal/Twombly plausibility standard
  • Local Patent Rule (LPR; also called Patent Local Rules) requirements for infringement & invalidity contentions
    • Infringement and invalidity contentions should actually contend something, not merely juxtapose claim language with a quote from product literature, as in many color-coded claim charts
    • For example, “the gizmo limitation in ‘123 [1.d] is embodied in the widget element of D’s product, because…” (why D’s widget == P’s gizmo, or is equivalent, or is not, or why widget in prior art == gizmo in P’s claim, etc.)
    • Sufficient infringement/invalidity contentions (claim charts) can drive other side’s mandatory disclosure requirements
  • See author’s articles and draft book on claim charts; many more claim chart issues than in brief list below:
  • Extended example: facade server claim chart
  • Claim charts: Underlying purposes & policies
    • Intended to replace contention interrogatories
    • Intended to crystallize party positions, avoiding post-Markman “musical chairs”, “shifting sands”
  • Claim chart requirements from Local Patent Rules
    • “each” x3: “A chart identifying specifically where each limitation of each asserted claim is found within each Accused Instrumentality”
    • [… more …]
  • Claim chart types & scenarios, including:
    • initial charts (formerly “PICs”),
    • infringement & invalidity contentions,
    • handling equivalence (DoE) with function/way/result,
    • handling means-plus-function using means (implementation) from specification,
    • handling alternate claim constructions in claim chart,
    • working/practicing chart (for ITC “domestic industry” requirement; how NPEs/”trolls” try to show “domestic industry”),
    • expert report charts,
    • obviousness (primary & secondary references),
    • post-Markman claim chart revisions (“shifting sands”), supplemental charts, final charts
  • Drafting claim charts:
    • Parsing claims into rows (“going beyond the semicolon”),
    • handling lengthy limitations,
    • sub-limitations (pros & cons of calling out sub-elements in separate rows of chart),
    • deceptively simple-looking limitations,
    • non-limiting clauses,
    • “weaving” claim limitations + claim construction (including e.g. means/function) + specific pinpointed fact re: the limitation’s presence or absence
  • Specificity:
    • pinpointing facts,
    • “location where” (point to specific place where each limitation is embodied or practiced),
    • names for locations,
    • level of detail,
    • accused instrumentalities,
    • relevant dates,
    • source-code references (filenames, function/method/class name, line numbers)
  • Common claim-chart problems
    • e.g. mere juxtaposition without explanation; mixing & matching across multiple accused products
  • Semi-automatic generation of claim charts with AI (e.g., GreyB GPS (GlobalPatentSearch.ai ; textual equivalent of claim charts from Perplexity.ai Patents beta e.g. here)
  • Briefly on charts for design patents (Crocs example)

Chapter 18: Discovery in Patent Litigation: Getting Information from the Other Side

  • Discovery: Getting materials from, and producing materials to, the other side
  • A patent = right to discovery, given a plausible case
  • This section also covers “informal discovery” (pre-discovery investigation) and mandatory disclosures under LPRs (specific disclosure requirements in part triggered by other side’s claim charts)
    • First diligently mine public information re: accused products (also potential prior-art products) — there’s likely more than you think; see Chapter XXX on pre-filing investigation
  • Types of materials in discovery
    • AI-generated materials & authentication, deepfakes, hallucinations; generally on need for attorney/expert caution
  • Only briefly here on the major FRCP discovery mechanisms: rogs, deps, doc production, etc.
    • Contention rogs discussed at Chapter XXX on claim charts, which intended to largely replace contention rogs
  • Focus here more on technical materials, than on economics/accounting
  • Who/What/When/Where/Why/How for discovery in patent litigation (see here for source-code example):
    • WHO from?: Possession, custody & control, and third parties; third-party subpoenas
    • WHO gets?: protective orders (POs), confidential business info (CBI), and attorney-eyes only (AEO), prosecution bar
    • WHO knows more?: including possible “burden of explanation” (see old flour barrels case)
    • WHAT is relevant, and/or likely to lead to relevant evidence?
      • Intent/knowledge evidence (emails etc.) — arguably irrelevant under strict-liability patent infringement, but: willful infringement, and indirect infringement (inducement, contributory); party may try to elevate its own requirements, to allow “smoking gun” evidence that otherwise irrelevant?
    • WHAT/WHICH to ask for?: specificity vs. “fishing expeditions”
      • Using public information to fine-tune formal discovery requests; how internal/confidential info relates to public info (e.g., how confidential source code relates to actual products on the market; how company emails relate to public pronouncements)
    • WHAT NOT produced?: briefly on attorney/client privilege (including unfortunately contrived attorney communications); attorney work product; expert drafts?
    • HOW should it be produced?: “native” and file formats; under protective order (PO), on computer isolated from internet
    • HOW MUCH discovery?: costs & burdens
      • Discovery mutually-assured destruction (MAD) balance upset by “trolls” (asymmetrical discovery from NPEs)
    • WHEN from?: past, present, and future docs
    • WHEN to produce?
    • WHERE to look for material to produce?
    • WHERE to produce?: mostly re: source-code production under PO
    • WHY are you asking? — “Stop and think” (FRCP ACN on discovery) vs. kitchen-sink requests; “Careful what you ask for, you just might get it”
    • HOW use?: or, what are you going to do with it?
  • Incomplete productions, withheld or missing docs, and spoliation
  • E-discovery and ESI: what’s the big deal?
  • Source code: a key example of discovery for patent litigation; including changes in cases re: AI/ML
    • Source code as an example of how infringement evidence relates to actual infringing acts (making, using, selling)
  • Discovery strategy/tactics, “games”
    • Knee-jerk responses (not every burden is unduly burdensome) are as common as knee-jerk requests (“give us the kitchen sink”)
  • Underlying policy considerations: why discovery matters
    • Public importance of discovery, though most important cases are sealed, so importance of the process for “knowledge economy,” rather than of specific evidence?
  • Discovery requests tailored to what experts will need; early involvement of consulting experts (see Chapter XXX) in drafting discovery requests
  • Discovery & case theories
    • D’s discovery requests to P re: invalidity, and re: P’s infringement theory (contention rogs -> claim charts)

Chapter 19: Expert Witnesses and Expert Evidence in Patent Litigation

  • The expert “eye”: need someone to “see” and explain presence/absence of claim limitations
  • Matching facts to claim limitations: pattern matching & forensics
  • Experts can “create facts”: another brick in the wall
  • Relation of technical & non-tech (economics, accounting, industry) experts
  • Expert choices:
    • specialist vs. generalist;
    • academic vs. industry;
    • early vs. late (sunk costs vs. fait accompli)
    • single vs. bifurcated (separate infringement and invalidity experts)
    • testifying expert vs. non-testifying consultant
    • fully-immersed (hands-on) vs. hands-off experts
  • Expert relation to hiring party:
    • witness not “team member”
    • not a conduit for inadmissible evidence (though expert may reasonably rely on otherwise-inadmissible evidence, if…)
    • ghostwritten expert reports
  • Testifying expert relation to non-testifying consulting expert; pros & cons of consulting experts
  • Experts, claim construction, obviousness, and the PHOSITA
  • Expert opinions:
    • Expert needs to offer opinions, vs. faux-naive “I just let the facts speak for themselves”
    • Conclusory opinions vs. providing basis and methodology (“because…”), alternate conclusions considered; the Wizard of Oz rule (“because because because…”)
    • What are opinions?; opinions are not merely beliefs (see Dennett), not even belief of expert (ipse dixit)
    • Basis/methodology/experience provide expert opinion’s value, vs. “well, that’s just your opinion”
    • Weasel words (see Jackson, “Understanding forensic science opinions“), “consistent with”, degrees of certainty, and “fit”
  • Expert reports:
    • Report drafting, non-discoverable drafts, and ghostwriting; including AI-generated parts of reports (there’s already an infamous case involving Anthropic)
    • FRCP 26 requirements
    • Rebuttal reports
    • Expert affidavits
    • Report timing e.g. re: Markman
    • Drafts & “rolling report”
    • Amending reports
    • What the expert “considered” (or didn’t) vs. relied upon (or didn’t)
  • Expert’s litigation methodology compared to non-litigation practice
  • Experts: Who, what, when where, why, how, how much? [TODO: match format of WHO/WHAT/etc. in Discovery chapter]
    • WHY: the role of experts in patent litigation
    • WHO: types of experts, and their relations to others
    • HOW: methodology and bases for expert opinion — FRE 702, Daubert, Kumho Tire
    • [… etc. …]
  • Expert depositions & other expert discovery
  • Battle of the experts: why there is legitimate disagreement
  • Using “Daubert factors” (actually FRE 702 + FRCP 26; “Don’t Say Daubert“) to structure report and testimony
  • Falsifiability, peer review, error rate, non-litigation use of method, standards for methods
  • Presenting any AI/ML-based results with basis, e.g. semantic results similarity/difference results presented with Euclidean distance (see PatLitig.ai on experts and AI), and semantics results must be brought back to claim language
  • Experience, qualifications, and ipse dixit
  • Battle of the experts; legitimate expert disagreement
  • More on expert witnesses & expert evidence, source-code book chapter on expert evidence, and chapter on expert reports

Chapter 20: Strategy, Tactics, and Alternatives to Patent Litigation

  • Patent strategy: A claims-based approach?
  • Some goals for patents, and for patent litigation
  • Pros: Some reasons behind (or causes for) patent litigation
  • Cons: Some bad reasons for patent litigation, and reasons NOT to litigate (see Chapter XXX pros/cons)
  • Non-litigation uses of patents: e.g. “signaling” to investors; part of inventor’s CV; egoboo; academic studies using patents
  • Patents as tools for collaboration and bargaining
  • Alternatives to patenting:
    • Both other IP and non-IP alternatives
    • Private alternatives e.g. trade secrets;
    • Social alternatives: non-patent inducements for R&D
  • Patent valuation: A claims-based approach; see Remedies/Damages Chapter XXX above
  • [Possibly a “potted history” of patent strategy, focusing on P/D collusion, and D defense associations during the Industrial Revolution]
  • “Trolls”:
    • which non-practicing entities (NPEs) are “trolls”?; e.g. Caltech v. Apple/Broadcom in Chapter 1?;
    • monetization of an otherwise-unused patent as a normal and expected use of the patent system?;
    • no working requirement, but working preference, including no injunction if $ damages sufficient (eBay v. MercExchange), and no import block without “domestic industry”
  • Settlement of patent litigation
  • Patent licensing
  • Pre-litigation patent enforcement
  • Strategy & tactics within litigation, including defense
  • Patent portfolios: quantity & quality
  • Industry-specific strategies e.g. pharma biosimilars, evergreening?
  • [Some overlap here with Chapter 16 on pros/cons of bringing suit; need to explain why Patent Portfolios in chapter on Strategy/Tactics (also in Chapter XXX on “multiplicity” issues]

Chapter 21: Multiplicity Issues

  • Multiple claims, multiple patents asserted in a single case
  • Multiple accused products, and choosing “representative instrumentalities”
  • Multiple (even conflicting) theories of infringement (how does expert witness handle hiring party request to provide support for all?)
  • Multiple prior-art references: either obviousness combination, or alternative prior-art references
  • Multiple potential claim constructions, pre-Markman
  • Multiple defendants, and joinder (AIA change); the problem of so-called “joint” or “divided” infringement; third parties; D joint defense/association
  • Multiple locations, including multidistrict litigation (MDL)
  • Multiple international jurisdictions (including attempted consolidation of invalidity hearings across multiple patent-granting jurisdictions)

Chapter 22: International Issues

  • Change YUCK title: patents in global context are not mere “issues”
  • This US-centric book may look narrow/parochial in a few years, if it doesn’t already
  • Slow evolution (with backward steps) to IP internationalization, e.g. UPC
  • Patent infringement/litigation is largely territorial (protection in one country)
  • Major contradictions: protection is territorial (per-country), but disclosure everywhere (foreign prior art is still prior art; AIA change to US treatment of some foreign prior art; see my old “Not Invented Here” paper)
  • International IP is one reason why patents generally not viewed as part of industrial policy (see Chapter 2)
  • International IP doctrines, e.g. MFN, national treatment, territoriality
  • International aspects of US patent litigation
    • ITC importation actions: things/res on docks (DHS), not signals; “domestic industry”
    • Export of components & US patent infringement
    • Cross-border issues (e.g. Canada in Blackberry case)
  • Contrast WIPO vs. WTO/TRIPS vs. bilateral (after US soured on TRIPS (one nation, one vote)
  • A bit on Europe (EPC), UK (post-Brexit), UPC, China [TODO: clearly need more coverage of each than suggested by single bullet point; also need Korea, India]
  • Globalization, trade, trade sanctions; is IP inherently a trade issue, apart from earlier US assertions in pushing TRIPS?; military/security restrictions

Chapter 23: Trends in Invention, Patenting, and Patent Litigation

  • Using AI for trends research e.g. finding “holes” in claim space
  • Patent “landscaping”: going beyond pretty four-color heatmaps; focusing on claims and litigation
  • Some of the following based on Suleyman, Coming Wave; and NotebookLM notebook with quarterly reports
  • Right now, below are mostly author’s very rough notes to self (some multiple industries grouped into single bullet point don’t make sense)
    • See GreyB “Future Products” [TODO: possibly put into NotebookLM, get Gemini-assisted summaries]
    • TODO: list top companies in each area, based on rough count of # patents/applications US & int’l
    • TODO: how do industry-specific ETFs relate to patent portfolios?
  • AI, ML, GPT, Transformers, attention, LLM, deep learning, neural network, genetic programming (GP)
  • Use of AI in other fields, e.g. neural net -> lithium reconfiguration (?), vaccine development
  • Security, military, drones: e.g. how IHAK ETF relate to patent portfolios?
  • 3D printing e.g. in construction/housing
  • Quantum: main methods of implementation; top companies specializing in (e.g. D-Wave, MagiQ); value of ungranted patent applications (1 year lag); cryptography response (don’t confuse w/ “crypto”)
  • Green/sustainability, energy, climate change, solar, batteries
  • Nuclear energy, including fusion; note long-secret nuclear patents
  • Robotics, vehicles, EV, autonomous vehicles (Waymo, Level 5)
  • 6G: Ericsson, Huawei (see GreyB analysis of Huawei 6G), Samsung, Qualcomm patents
  • VR, metaverse?: use pat stats (including # applications, int’l) to measure hype vs. reality?; or patenting activity itself reflects hype?
  • Crypto, DiFi??: same hype question
  • Brain/computer interface: Neuralink, Merge Labs (Sam Altman)
  • Biotech, genetic sequences, synthetic bio, CRISPR (ongoing Doudna v. Charpentier CRISPR-Cas9 litigation), gene editing
  • Space, satellites, LEO: SpaceX, Blue Origin
  • Entertainment, social media (key products for input to “knowledge economy”, and training for AI)

Chapter 24: Conclusion

  • Recap of major points, themes of book
  • [TODO: Possibly make Conclusion focus more on social goals of patent system, with Chapter 2 more on private goals for patent owners and potential defendants?]
  • Necessarily-contradictory system; benefits of imperfections [TODO: add more from notes]
  • Taking some fact-investigation methods from here, and applying to other areas of software-related litigation
  • Desire for greater IP predictability? (Some/many may prefer it the way it is; pros & cons of unpredictability for attorneys)
  • More on how patents relate to economics of innovation, economic growth: revisit themes of Chapter 2 now that we’ve worked through claims, litigation specifics
    • Relate the book to economics of Romer, Mokyr, Scotchmer et al.

Appendices

  • Caltech v. Apple/Broadcom details
  • Glossary & Abbreviations
  • Checklists, e.g. issues to consider when assessing other side’s, or one’s own, infringement/invalidity contentions
  • Bibliography
  • Cases Cited

Outline above may insufficiently cover:

  • Trolls, NPEs
  • On-sale, public-use bar
  • Indirect infringement (inducement, contribution)
  • Design patents
  • 101 subject matter becoming more important?
  • Venue, forum shopping, EDTX, WDTX, Delaware?
  • Appeals, CAFC
Print Friendly, PDF & Email