Zum Inhalt springen

SQL Was Called SEQUEL Until an Aircraft Manufacturer Objected

Zusammenfassung

SQL — the query language that manages essentially every relational database in the world — was originally named SEQUEL: Structured English Query Language. Donald Chamberlin and Raymond Boyce at IBM published it under that name in 1974. IBM was forced to rename it to SQL (Structured Query Language) because SEQUEL was already a registered trademark of the Hawker Siddeley aircraft manufacturer in the UK. The name SQL was chosen as a minimal variation. Database administrators have been arguing about whether to pronounce it “sequel” or “ess-cue-ell” ever since.

The 1970 Paper That Made SQL Possible

In 1970, Edgar Codd published “A Relational Model of Data for Large Shared Data Banks” — a paper that defined the mathematical foundations of relational databases. Codd’s model specified that data could be organized in tables (relations), that queries could be expressed as operations on sets, and that the physical storage of data should be independent of its logical representation.

Codd was a mathematician at IBM’s San Jose Research Laboratory. His paper described the relational model in terms of relational algebra and relational calculus — formal mathematical notations. The model was correct and powerful but expressed in a form that required mathematical training to use.

SEQUEL: Making Codd’s Model Usable

Donald Chamberlin and Raymond Boyce at IBM set out to design a language based on Codd’s model that non-mathematicians could use. Their 1974 paper, “SEQUEL: A Structured English Query Language,” described a language that expressed database queries in something approximating English syntax:

SELECT EMPNAME
FROM EMPLOYEE
WHERE SALARY > 50000

The goal was accessibility: a manager or analyst without mathematical training should be able to write and read database queries. The SELECT-FROM-WHERE structure maps roughly to English sentence structure, which was deliberate.

Chamberlin later described SEQUEL’s design philosophy: “The principle we tried to follow was that a person who could not read computer programs could read SEQUEL and understand what it was doing.”

The Trademark Problem and the Rename

When IBM sought to standardize the language, it discovered that “SEQUEL” had been registered as a trademark by Hawker Siddeley Group — a British aircraft and engineering conglomerate — for one of their products. Rather than engage in trademark litigation, IBM quietly renamed the language SQL. No announcement was made; documentation was updated, and the name change became visible only gradually.

The resulting pronunciation ambiguity has never been resolved. The original “sequel” pronunciation carries through from the original name (many database administrators and the founders themselves say “sequel”). The official ISO standard name is “SQL,” which invites letter-by-letter pronunciation (“ess-cue-ell”). Both are in common use; neither is officially incorrect.

The Database Wars That Followed

IBM’s internal SQL implementation was called System R (prototype, 1974-1979) and later became part of IBM’s DB2 product line. However, Oracle Corporation — founded by Larry Ellison, who read Codd’s paper before IBM productized it — shipped a commercial relational database using SQL in 1979, before IBM’s commercial release. The result was that Oracle, not IBM, dominated the commercial relational database market that IBM’s own research had created. The full story is in The Database Wars and Larry Ellison and Oracle.

📚 Sources