brief: initial proposal — five-entry catalog of computing terms traced to primary sources (bit, byte, software, bug, daemon)
brief: Computing Terms: Origins on the Record
1. Filing
- Pillar: Catalogs
- Working title: Computing Terms: Origins on the Record
- Slug: computing-etymology
- Researcher: Lewis Aldea, Staff Researcher
- Date filed: 2026-06-10
2. Angle
The founding vocabulary of computing — bit, byte, software, bug, daemon — carries origin stories that have been repeated into legend. This catalog reads the primary sources: the actual memo, the actual logbook, the actual conference paper. The recurring finding is not that the legends are false, but that the real stories are flatter, stranger, and more contingent than the standard accounts — and that the gap between the folk story and the document is usually the interesting part.
3. Pillar justification
Catalogs. Each entry covers one term: the folk story, the primary source, and the gap between them. The series is open-ended — there are dozens of computing terms with traceable primary documents and no natural stopping point. The format is niche in the way Catalogs should be: nobody is going to fund a book called The Primary Sources for Computing Vocabulary, but it is exactly the kind of patient, specific project that accumulates value entry by entry. Several of these chains run through archives, conference proceedings, and internal memos that reward deliberate searching. This is not Cross-references (no mechanism being carried across fields) and not From the Stacks (not a single artifact being opened) — it is a standing obsession with one specific kind of document: the origin record.
4. Prior art
Queries run:
- Searched institutional memory for "computing etymology bit byte daemon bug software"
- Searched institutional memory for "etymology vocabulary computing history origin terms"
Findings: Both queries returned zero results. No prior pieces on this topic in the archive. Net new.
5. Primary sources
Bug:
[1] Thomas A. Edison, letter to William Orton (President, Western Union), March 1878. Quote: "You were partly correct, I did find a 'bug' in my apparatus, but it was not in the telephone proper. It was of the genus 'callbellum.'" Archival home: Edison Papers, Rutgers University, directed by Paul Israel. Sold at Swann Galleries auction, 2024.
[2] Thomas A. Edison, letter to Theodore Puskas, November 13, 1878. Quote: "Bugs—as such little faults and difficulties are called—show themselves..." Edison Papers, Rutgers University. Quoted in IEEE Spectrum, "Did You Know? Edison Coined the Term 'Bug'" (accessible at spectrum.ieee.org).
[3] Grace Murray Hopper, Mark II Computation Log, September 9, 1947. Entry: "First actual case of bug being found." Moth taped to page. Smithsonian National Museum of American History, Object 1994.0191.1. Digitized; publicly viewable.
Bit:
[4] John W. Tukey, Bell Laboratories Memo, January 9, 1947. Coins "bit" as contraction of "binary digit." Survives as carbon copy in the Herman Heine Goldstine Papers, American Philosophical Society. Cataloged: "Bell Laboratory Memo with the First Use of the Word 'Bit,' John Tukey, 1947."
[5] Claude E. Shannon, "A Mathematical Theory of Communication," Bell System Technical Journal, vol. 27, no. 3, July 1948, pp. 379–423. Shannon credits Tukey: "The choice of a logarithmic base corresponds to the choice of a unit for measuring information... the resulting units may be called binary digits, or more briefly bits, a word suggested by J. W. Tukey." Accessible via pmc.ncbi.nlm.nih.gov (PMC3629598) and the Bell System Technical Journal archives.
Byte:
[6] G. A. Blaauw, F. P. Brooks Jr., W. Buchholz, "Processing Data in Bits and Pieces," IRE Transactions on Electronic Computers, vol. EC-8, no. 2, June 1959, pp. 118–124. First published use of "byte." Accessible via IEEE Xplore; likely paywalled but may be in library holdings.
[7] Werner Buchholz, ed., Planning a Computer System: Project Stretch (New York: McGraw-Hill, 1962), Chapter 4. Elaborates the byte concept. Earlier Stretch project memos from June 1956 establish the term internally; those are held at the Computer History Museum (archive.computerhistory.org/resources/text/IBM/Stretch/).
Software:
[8] Richard R. Carhart, "The Systems Approach to Reliability," Proceedings of the Second National Symposium on Quality Control and Reliability in Electronics, Institute of Radio Engineers, January 1956, p. 149. First documented print use of "software" — but meaning human personnel in a system, not computer programs. Proceedings cataloged at Stanford SearchWorks (id 377313); likely requires library access.
[9] John W. Tukey, "The Teaching of Concrete Mathematics," American Mathematical Monthly, vol. 65, no. 1, January 1958, pp. 1–9. First documented print use of "software" to mean computer programs. Quote: "Today the 'software' comprising the carefully planned interpretive routines, compilers, and other aspects of automative programming are at least as important to the modern electronic calculator as its 'hardware' of tubes, transistors, wires, tapes and the like." Traced by Fred R. Shapiro (Yale Law School) via JSTOR search. Accessible via JSTOR with institutional access.
Daemon:
[10] Clerk Maxwell, Theory of Heat (London: Longmans, Green, and Co., 1871), Chapter XII, "Limitation of the Second Law of Thermodynamics." Origin of the Maxwell's Demon thought experiment. Digitized via Google Books and Internet Archive. Accessible.
[11] The Jargon File, s.v. "daemon" (Eric S. Raymond, ed.). Documents that DAEMON (Disk And Execution MONitor) is a retroactive backronym applied to programs named after Maxwell's Demon by the MIT CTSS team, ca. 1963. Available at catb.org (currently 503; use Wayback Machine snapshot or archived copies).
6. Key claims
Claim 1: "Bug" for technical malfunction predates Hopper's 1947 logbook entry by nearly 70 years; Edison used the term in letters as early as March 1878. Hopper's logbook entry itself reads "First actual case of bug being found," indicating the word was already established vocabulary, not coined there — the irony is that the most-cited origin document for the term is itself evidence the term predates the document. — Sources [1], [2], [3]
Claim 2: Tukey coined "bit" in a January 1947 Bell Labs memo; Shannon popularized it in 1948 and explicitly credited Tukey. The originating document survives as a carbon copy in a colleague's papers (Goldstine's), not Tukey's own. — Sources [4], [5]
Claim 3: Buchholz coined "byte" in June 1956 for IBM's Stretch project, deliberately misspelling "bite" to prevent its abbreviation to "b" (which would collide with the abbreviation for "bit"). At coinage it referred to variable-length bit groups, not 8 bits specifically; the 8-bit standardization came later with IBM System/360. — Sources [6], [7]
Claim 4: The first documented print use of "software" (Carhart, January 1956) uses the word to mean human personnel, not computer programs. Tukey's 1958 usage — first to mean programs — is what the computing world adopted. The word was effectively repurposed, not coined, for its current meaning. — Sources [8], [9]
Claim 5: The MIT "daemon" (CTSS, ca. 1963) was named after Maxwell's Demon; the DAEMON acronym (Disk And Execution MONitor) is a retroactive backronym, confirmed as such by the Jargon File. The Maxwell's Demon origin predates computing by nearly a century. — Sources [10], [11]
7. Open questions
- Edison letter archives: IEEE Spectrum credits the Edison Papers at Rutgers University (Paul Israel, director) as the source for both 1878 letters. Exact archival call numbers not confirmed this shift. Writer should verify via the Edison Papers project (rutgers.edu) before treating these as primary citations rather than secondary accounts.
- Carhart's exact text and context: The 1956 proceedings appear to require library access. Stanford SearchWorks lists holdings. Writer needs to confirm Carhart's use of "software" means people (not programs) in full context, not just in the summary found this session. This is load-bearing for Claim 4.
- Niquette's claimed 1953 coinage: Paul Niquette claims he coined "software" verbally in 1953 in his memoir. His website (niquette.com) was inaccessible this session (503). The claim appears undocumented beyond his own account — no contemporary records, no witnesses cited. Writer should flag as unverifiable claimed prior rather than ignoring or accepting it.
- Daemon acronym timeline: The Jargon File confirms DAEMON is a backronym but does not date when the acronym was retroactively applied. If dateable, this sharpens the daemon entry.
- Series framing: Five initial entries. Whether the title "Computing Terms: Origins on the Record" and the folk-story / primary-source / gap entry format should be fixed at this stage, or left for the writer to establish in the first installment, is an editorial judgment for the night editor.
8. Length estimate
Researcher estimates: 1,500–2,500 words for the initial five-entry installment. Writer may revise: Yes — final length to be determined by what the material supports.
— Lewis Aldea, Staff Researcher