Source code ch.18: Searching

Chapter 18: Searching source code for patent-claim-relevant elements and steps

18.1 Searching the code: What to look for

18.1.1 BOX: “Let’s not make a big deal out of this”?
18.1.2 “Synonyms” and examples: Generating search keywords
18.1.3 BOX: The doctrine of equivalents
18.1.4 Common programming terms, jargon, abbreviations
18.1.4B Standardizing software patent terminology: US PTO efforts (see GAO report)
18.1.5 Terms reflecting implementation details (or, Searching for X, by searching for Y used to implement X)
18.1.6 Dictionaries, learned treatises, and other reference works
18.1.6.1 BOX: A cautionary note about dictionaries
18.1.7 BOX: Two cheers for patent classification systems (using patent classification to assist code searching; but PTO classification is purpose- rather than implementation-based?)
18.1.8 Code naming conventions
18.1.9 Connectives
18.1.10 Naming problems & searching problems
18.1.11 Turning patent claims “back into” the language of the underlying invention (software patent claims may be patent-attorney-encoded versions of code) [analogy: if patent claim states “liquid bisects an open cylindrical vessel”, then the correct search terms may be “glass half empty” or “glass half full”]

18.2 Searching the code: mechanics

18.2.1 Regular expressions and grep (including on Macs)
18.2.2 findstr on Windows machines
18.2.3 mkndx/find (see chapter 1: VBS code; C code; awk code)
18.2.4 Understand (SciTools)
18.2.5 dtSearch
18.2.6 Lucene-based tools (including Understand “Instant Search” and DocFetcher)
18.2.7 BOX: Searching by counting, and establishing a baseline, and a bird’s eye view

18.3 Searching the code: methodology issues

Potential Daubert questions regarding how expert/examiner decided what code to look at, not look at
Are there standard methodologies here?; academic literature on source-code retrieval (see bibliography)
E-discovery cases regarding search/retrieval experts, methodology (ABA, Litigators on Experts at 366-367)