AI Archaeology
Mining Forgotten Documents
SOFTWARE/UI PATENT #52026-05-09

1959-04-08 Pentagon Convening → 1959-06-04 CODASYL Founded → 1960-01-08 Spec Approval → 1960-08-17 First Run on the RCA 501 — Charles Phillips of the US DoD Convened a Short-Range Committee of 6 Commercial Vendors (Burroughs / IBM / Minneapolis-Honeywell / RCA / Sperry Rand / Sylvania) + 3 Government Agencies (US Air Force / Navy David Taylor Model Basin / National Bureau of Standards) Chaired by Joseph Wegstein, Drafted COBOL 60 With Grace Hopper's FLOW-MATIC (Remington Rand, 1955-1959) as Its Direct Parent Language, and Published it as a Government Printing Office Document — and Yet Wikipedia EN COBOL, Wikipedia EN Grace Hopper, Wikipedia EN FLOW-MATIC, the Yale CS Hopper Story, and Britannica Hopper All Carry No COBOL-Related Patent Number: Eligibility Wall Excavation #7 (SW Subseries DB Form: Eligibility Wall (a) pre-judicial era × (c) government-contract hybrid form #1)

About This Excavation Memo

Primary-source URLs confirmed and full text not read (working range: 6 secondary sources — Wikipedia EN COBOL, Wikipedia EN Grace Hopper, Wikipedia EN FLOW-MATIC, the Yale CS Hopper Story (cs.yale.edu/homes/tap/Files/hopper-story.html), the Britannica Hopper entry, and gracehoppers.wordpress.com FLOW-MATIC explainer). No patent number for the COBOL 60 spec, FLOW-MATIC (the direct parent of COBOL), or Hopper's B-0 / A-0 compilers was found within today's verify scope, so this memo is written as 'a patent-absence excavation log = structural record of Eligibility Wall (a) pre-judicial era × (c) government-contract hybrid form.'

1. COBOL CODASYL Committee Basics

FieldValue
Pentagon conveningApril 8, 1959, US DoD Charles Phillips (Director, Data System Research Staff) convened major computer vendors and government agencies at the Pentagon
Reason for conveningDoD had 225 computers in service, 175 on order, and had spent over $200M on implementing programs. Portable programs were needed
CODASYL foundedJune 4, 1959, Conference on Data Systems Languages (later Committee on Data Systems Languages)
Short-Range Committee6 commercial vendors (Burroughs / IBM / Minneapolis-Honeywell / RCA / Sperry Rand / Sylvania) + 3 government agencies (US Air Force / Navy David Taylor Model Basin / National Bureau of Standards)
ChairJoseph Wegstein (NBS, also concurrent on the ALGOL 60 committee)
Direct parent languageGrace Hopper FLOW-MATIC (Remington Rand, developed 1955-1959, public 1958, partial input from IBM COMTRAN). Hopper's 1980 recollection: 'COBOL 60 is 95% FLOW-MATIC'
Spec approvalJanuary 8, 1960 by the executive committee, sent to the Government Printing Office
Disclosure venueGovernment Printing Office printed 'COBOL 60'; an electronic reproduction is available on archive.org (cobol-60)
First compile success1960-08-17, first COBOL program ran on the RCA 501
Compatibility demo1960-12-06/07, RCA and Sperry Rand (Univac) ran the same program on both computers
Patent numberNone found within today's verify scope (Wikipedia EN, Yale CS Hopper Story, Britannica — none cite any patent number)

2. Core: Structure of 'Eligibility Wall (a) pre-judicial era × (c) Government-Contract Hybrid Form'

(a) Verify Status of Secondary Sources

  • WebSearch ""COBOL" OR "FLOW-MATIC" Grace Hopper Remington Rand patent USPTO" → COBOL development history, FLOW-MATIC explainers, and Grace Hopper's career are abundantly indexed, but 0 hits for COBOL- or FLOW-MATIC-related patent numbers
  • Wikipedia EN COBOL: detailed chronology (Pentagon convening 1959-04-08, CODASYL founded 1959-06-04, Short-Range Committee composition, 1960-01-08 approval, GPO publication, 1960-08-17 RCA 501 first run, 1960-12-06/07 compatibility demo) — no patent reference, no IP-clause reference
  • Wikipedia EN Grace Hopper: FLOW-MATIC (Remington Rand 1955-1959), A-0 compiler (1952), B-0 compiler (1956), and contribution to COBOL — no patent reference
  • Wikipedia EN FLOW-MATIC: developed on Remington Rand UNIVAC I 1955-1959, public in 1958, derived as AIMACO (Air Force) before being merged into COBOL — no patent reference
  • Yale CS Hopper Story (cs.yale.edu/homes/tap/Files/hopper-story.html): from Hopper's Harvard Mark I era to COBOL — no patent reference
  • Britannica Grace Hopper Kids: awards and career — no patent reference
  • gracehoppers.wordpress.com FLOW-MATIC explainer: verbatim citation of Hopper's 1980 recollection that FLOW-MATIC supplied 95% of COBOL — no patent reference

(b) DoD Government Funding × CODASYL Committee × 6-Vendor Cooperative-Disclosure Structure

In 1959 the COBOL spec was developed as a hybrid of form (a) pre-judicial era × form (c) government-contract-mandated public disclosure:

  1. Form (c) government-contract-mandated public disclosure: DoD's Charles Phillips brought $200M (two hundred million dollars) of government investment and a 225-machine fleet to bear on the Pentagon convening. Federal procurement regulations of the period (forerunners of today's FAR/DFARS) carried clauses making public disclosure the default for deliverables of government-funded work, and the COBOL 60 spec was mandatorily published as a Government Printing Office document. Same structure as Day 25 ep89 SW-003 BBN IMP (1969 ARPA contract → BBN Report 1822 disclosure → Internet STD 39)
  2. Form (a) pre-judicial era: 1960 sits 12 years before Gottschalk v. Benson, with no judicial precedent on U.S. software-patent eligibility yet established. COBOL 60's core — English-like statements, the PROCEDURE DIVISION structure, the DATA DIVISION hierarchy, the FILE DIVISION — are all algorithm- and data-structure inventions, and drafting Claim 1 was difficult under the period's 'mathematical method = unpatentable' doctrine
  3. Effect of the 6-vendor cooperative form: With Burroughs / IBM / Minneapolis-Honeywell / RCA / Sperry Rand / Sylvania as 6 commercial vendors agreeing to simultaneous disclosure inside the Short-Range Committee, any later COBOL-related patent application from any of them would be mutually rejected as prior art

This hybrid form reads as the problem-space precursor of post-1980s open standards (POSIX, TCP/IP, HTTP, JSON). The three-element overlay of government funding + multiple commercial vendors cooperating + patenting difficulty in the pre-judicial era built a mechanism that effectively foreclosed any single-company patent strategy after the fact.

(c) Note on the Patent Record of FLOW-MATIC, COBOL's Parent

The direct parent of COBOL — Grace Hopper's FLOW-MATIC (Remington Rand 1955-1959) — also carries no patent number in any source today: Wikipedia EN FLOW-MATIC, the Yale CS Hopper Story, and gracehoppers.wordpress.com. Hopper had developed B-0 (1956) and A-0 (1952) compilers while at Remington Rand, and no patent record was found within today's verify scope for those either.

Hopper's 1980 verbatim recollection 'COBOL 60 is 95% FLOW-MATIC' shows COBOL as the direct evolution of FLOW-MATIC, and the patent absence of FLOW-MATIC further reinforces COBOL's pre-judicial era form. Tracing ownership transfers Remington Rand → Sperry Rand → Sperry → Unisys (1986 merger) yields no FLOW-MATIC / COBOL-related patents in any secondary source.

3. To Be Strict (3 Required Items)

Confirmed Facts

  • The chronology — 1959-04-08 DoD Charles Phillips's Pentagon convening, 1959-06-04 CODASYL founding, 1960-01-08 executive committee approval, 1960-08-17 first run on the RCA 501, 1960-12-06/07 RCA-Univac compatibility demo — is detailed in Wikipedia EN COBOL and confirmed
  • That DoD had 225 computers in service, 175 on order, and over $200M spent on program implementation is verbatim recorded in Wikipedia EN COBOL
  • The Short-Range Committee composition (6 commercial vendors: Burroughs / IBM / Minneapolis-Honeywell / RCA / Sperry Rand / Sylvania, + 3 government agencies: US Air Force / Navy David Taylor Model Basin / National Bureau of Standards) and Wegstein as chair are verbatim recorded in Wikipedia EN COBOL
  • That Grace Hopper's FLOW-MATIC (Remington Rand 1955-1959) is COBOL's direct parent is confirmed in Wikipedia EN COBOL and FLOW-MATIC; Hopper's 1980 recollection 'COBOL 60 is 95% FLOW-MATIC' is verbatim cited

Author's Interpretation

  • 'Eligibility Wall (a) pre-judicial era × (c) government-contract hybrid form' is post-hoc framing built from comparison with ep88 FORTRAN '(a-1) corporate-lab solo type,' today's ep91 LISP '(a-2) pure academic-disclosure type,' and ep92 ALGOL 60 '(a-3) international-committee cooperative type'. Specialists may push back that 'COBOL is led by form (c) government contract and (a) pre-judicial era is secondary' or 'the 1959 DoD government-contract clauses differ from the 1969 ARPA contract'
  • 'The 6-vendor cooperative-disclosure machinery forecloses any later patent strategy' is post-hoc framing; whether the 6 vendors at the time explicitly intended IP defense is unverified

Where the Comparison Breaks

  • A comprehensive USPTO Patent Center search for 1955-1972 COBOL / FLOW-MATIC-related patents from Burroughs / IBM / Minneapolis-Honeywell / RCA / Sperry Rand / Sylvania was not performed today. It is possible that one of the 6 vendors took COBOL-related patents internally
  • 'DoD $200M investment, 225 machines in service' is secondary-source information via Wikipedia EN; the original DoD 1958-1959 budget records were not examined
  • Hopper's B-0 / A-0 compilers may have been protected as Remington Rand internal Trade Secret, and 'not patented' and 'not disclosed' are separate questions. Sperry Univac / Unisys archive examination is needed

References