Source Code & Software Patents: A Guide to Software & Internet Patent Litigation for Attorneys & Experts
by Andrew Schulman (http://www.SoftwareLitigationConsulting.com)
Detailed outline for forthcoming book
Chapter 30: Remedies for patent infringement (damages and injunctions)
30.1 Introduction
- Use of source code in litigation re: damages, including lost profits and reasonable royalty
- Use of source code in litigation re: injunctions, preliminary injunction (PI), and ITC importation order
30.2 Damages: Lost profits
- Lost profits are the primary measure of patent damages; reasonable royalty is a fall-back position
- Panduit 4-factor test, and post-Panduit changes (e.g. no longer all-or-nothing)
- Two Panduit factors include technical considerations, possibly informed by source-code examination:
- Presence of absence of non-infringing alternatives (see 30.7 below)
- Patent owner’s ability to meet consumer demand which would have existed but-for D’s infringement; in addition to marketing and sales ability, this is in part also an issue of technical ability (see working/practicing at 30.8 below)
- Entire market value rule (see 30.11 below)
30.3 Damages: Reasonable royalty
- Fall-back position from lost profits
- Georgia-Pacific (G-P) many-factored test
- Several G-P factors include technical considerations, possibly informed by source-code examination:
- Extent of D’s use of the invention, including % of product/profits/revenues attributable to invention (see 30.9 below)
- Infringer’s improvements to its commercial embodiment of the invention (see 30.10 below)
- Existence of non-infringing alternatives would likely play a role in hypothetical licensing negotiations (see article on non-infringing alternatives & reasonable royalty, and 30.7 below); in Uniloc v. Microsoft, the ancient 25% royalty “rule of thumb” was rejected in part because the “rule” took no account of non-infringing alternatives
- Technical vs. consumer-demand tests for attributing commercial success to the invention
- Evidence from source-code licenses (see 30.15 below)
30.4 Damages: Treble damages from willful infringement
- Evidence of “slavish copying” from source code
- Evidence of willfulness from source-code comments, or emails etc. re: source code
- Is it possible to show NOT willful, e.g. when all limitations are all in code, but they are dispersed, developed at different times?
- While slavish copying can show deliberate/willful copying, it is not necessary to show wilfullness (see Stryker v. Intermedics)
30.5 Attorneys fees
- Awarding reasonable attorneys fees to prevailing party in “exceptional cases” (35 USC 285)
- D’s source-code discovery abuse (see chapter 9 on discovery)
- P wrongfully bringing case without reasonable pre-filing investigation (see Rule 11, etc. in chapter 6 on pre-filing investigation)
30.6 Injunctions, preliminary injunction (PI), and equitable considerations
- Traditional four equitable considerations:
- Irreparable harm
- Inadequate remedy at law (i.e., damages alone cannot make P whole); or, for PI, likelihood of success on the merits
- Balance of hardships between P and D
- Public interest
- No more presumption of irreparable harm, from mere act of infringed patent?
- eBay v. MercExchange on damages as adequate remedy in law for non-practicing entities (NPEs, “trolls”)
- To show non-NPE status, demonstrate working/practicing invention (see 30.8 below)
- Relevance of source code for future, unreleased products, re: injunction (see 30.13 below)
- Separability/integration as issues in injunction feasibility (relative hardships) and injunction enforcement:
- Can D easily isolate/extract/split-off infringing component from the rest of the product?;
- P will for purposes of relative-hardships say it is easy for D to remove infringement, but it will desire inseparability/integration when using entire market value rule to determine lost profits
- Injunction enforcement, contempt proceedings (Tivo v. Echostar):
- Is D’s new code not colorably different from (substantially similar to?) code which has been enjoined?
- Use of copyright-like source-code comparisons in post-injunction contempt proceedings
- Assessing attempted post-injunction design-around (see also 30.7 below)
- ITC importation order requires domestic industry, which however may even be from unwilling licensee (see working/practicing at 30.8 below)
- Quick source-code examination, or product reverse engineering, to meet tight deadlines of PI or temporary restraining order (TRO) hearings
30.7 Non-infringing alternatives & design-around
- Relevant to both lost profits (30.2 above) and reasonable royalty (30.3 above)
- Does P need to proactively show absence of non-infringing alternatives?
- Non-infringing alternatives can belong to a third party
- Cost of design-around: reasonable royalty must be less than the cost of design-around
- Next-best alternative: value of patented invention, minus cost of next-best alternative
- Assessing post-injunction design-around in contempt proceedings (see also 30.6 above)
- [MORE xxx]
30.8 Showing working, practicing, commercial embodiment, and/or domestic industry
- US patent law does not contain a working/practicing requirement; hence the phenomenon of non-practicing entities (NPEs) or “trolls”
- However, working/practicing is relevant in several areas:
- showing possibility of irreparable injury for injunction (i.e., non-“troll” status under eBay v. MercExchange);
- technical prong of domestic industry showing for ITC importation action; can be shown even with the acts of an unwilling licensee from previous litigation
- ability to meet consumer demand for Panduit lost profits test (30.2 above) has a technical aspect: in addition to marketing/sales ability, technical ability to practice the invention can clearly be shown by fact of already practicing/working the invention; can’t be speculative; see e.g. Hebert v. Lisle
- contrast showing domestic industry in ITC action, by pointing to working/practicing by unwilling licensee, with need to show one’s own ability, or ability which can be acquired, for Panduit test of ability of meet consumer demand
- a factor in the G-P reasonable royalty test (30.3 above; see e.g. Riles v. Shell);
- though non-remedies related, working/practicing also relevant to the commercial-success factor of nonobviousness secondary considerations
- Local Patent Rules on demonstration of working/practicing, e.g. ND Cal. LPR 3-1(g): “If a party claiming patent infringement wishes to preserve the right to rely, for any purpose, on the assertion that its own apparatus, product, device, process, method, act, or other instrumentality practices the claimed invention, the party shall identify, separately for each asserted claim, each such apparatus, product, device, process, method, act, or other instrumentality that incorporates or reflects that particular claim.”
- See chapter 26 (claims tables) for discussion of working/practicing tables
- Question is embodiment: what commercial products does the patent “cover”?; this is related to patent marking: what does it mean to say that x is a “patented product”?
30.9 Role of the invention in the product, and percentage of revenue/profit attributable to the invention
- Relevant to both lost profits and reasonable royalty
- Sometimes the % of D’s product, represented by P’s invention, is measured by counting lines of code (LOC)
- See e.g. code counts in Lucent v. Gateway re: “date picker” feature in Microsoft Outlook; while the result (that date picker caused only a small % of consumer demand for Outlook) was reasonable, the LOC measure was naive
- A small amount of code can represent a large % of a product’s value (which value may then translate into consumer demand, revenues, profits)
- Even a rarely-used feature can be the basis for consumer demand, e.g. electronic lock/key’s functionality during the rare occurrence of an attempted break-in; if doesn’t operate properly in rare boundary condition, essentially doesn’t work
- Other software metrics, apart from LOC; see Capers Jones et al.
- In some cases, the % of the product represented by the infringement cannot be measured entirely by static examination of source code, and will depend in part on run-time behavior (see e.g. WMP encoder in Lucent v. Gateway; size of routing cache in Bedrock)
- How the source-code examiner can assist in measuring infringement’s role in a commercial product
- During the source-code exam, remember to measure the total amount of source code provided in several different ways; isolating/extracting and “measuring” the relevant infringing portion will be more difficult
- Comments in source code, and/or non-source documents re: source code may help determine the value of infringement (e.g. developer email on why a particular approach was or wasn’t taken, stating “Doing it this way will save us tons of $”, but make sure that the non-source doc is really referring to the same thing as the code; see chapter 29 on correlating source- and non-source docs)
- [What about D’s evidence indicating that using the invention has caused problems, high tech-support costs, etc.?]
- Generally courts look to consumer demand rather than supply-side technical considerations
30.10 Infringer’s improvements
- One Georgia-Pacific factor for reasonable royalty looks at the extent of the infringer’s improvements to the inventor
- In source code, a core of infringement may be surrounded by a large number of small tweaks, special casing, bug fixes, etc.; see xxx
- Not even patentable improvement avoids infringement, but may account for commercial success
- D may argue that its history of improving the core invention, rather than the invention itself, accounts for commercial success; for example, xxx [MORE]
30.11 Entire market value rule
- Relevant to lost profits from “convoyed sales” [MORE]
- Here, the question is not whether a product embodies the patent, but rather whether a product, which does NOT practice or embody the patent, nonetheless is a component of, and/or has a functional relationship with, a product which DOES embody or practice the invention, such that infringer should be liable for sales relating to these “non-patented” products; see Rite-Hite v. Kelley
- Component
- Functional relationship
- These are technical rather than consumer-demand issues; see also litigation over “integration”/separability in e.g. Microsoft antitrust cases
30.12 A % of What?
- Often, the parties care more about what product is being used as the baseline for damages, rather than the % of the product
- For example, P may prefer to have even a tiny % of Microsoft Windows, rather than a larger % of something else
- P generally is interested in siting infringement at a small but arguably inseparable portion of a major product; only asking for a small % can give P and its experts an aura of “reasonableness”
- D generally is interested in siting infringement at a minor product or even a give-away (“oh that old thing”), though this may be contradicted by its earlier “crown jewels” assertions during PO and discovery negotiations
- D’s efforts to downplay the importance of the infringing product may be undermined by its own marketing touting “tight integration”, etc.
- For the entire market value rule (see xxx above), is otherwise-ancillary invention inseparably part of a functionally-integrated big-ticket product?; see Rite-Hite
30.13 Future products
- Relevant for injunctions
- [Generally no damages for future infringement, but likely affects settlement negotiations, ongoing licensing fees]
- Purported relevance for remedies may be used as rationale for discovery request for source code to future, unreleased products (see DisplayLink v. Magic Controls)
30.14 Past products and damages period; dating infringement
- Six-year statute of limitations (SOL) on patent infringement damages
- Laches
- Tolling of laches period when infringement was latent, not readily ascertainable, no should-have-known?; see chapter 6 on ease/difficulty of detecting infringing in products on the market, and especially in processes/services
- Importance of establishing first date for infringing source code
- Difficulty in some circumstances of determining source-code dates (see ch.xxx)
- When did infringement begin?: D would like very early, predating patent priority date, to establish invalidity (or create prior user rights for business methods); but failing that, D will prefer that infringement start as late as possible to reduce damages
- Dates of course also relevant to anticipation, invalidity, prior user rights for business methods, etc.; see ch.xxx
30.15 Source code licenses
- Relevant to reasonable royalty G-P calculations
- Earlier licenses of source code which are now known to have included the infringing code, but which were not viewed at the time as patent licensing [MORE xxx]
- See also inferring low value from infringement’s incorporation in open-source software licensed for $0 [MORE xxx]
30.16 Damages and experts
- Economist reliance on technical expert, including possibly for technical aspects of market definition
- Economist’s own source code, e.g. spreadsheet macros used in damages models; see case xxx where economic model buried in uncommented code with inscrutable names
- In patent cases, Daubert challenges are more common re: economics experts and their methodology; technical expert methodology comparatively goes unchallenged (see ch.xxx)