Who was First to Use the Relational Model?
In the early 1970's, IBM's System R "was intended to provide
a high-level, nonnavigational, data-independent interface to many users
simultaneously, with high integrity and robustness." By the end of the 1970's,
IBM had experimented with "a full-function, multi-user version." One of the main
results of that effort was the development of a structured query language, or
At roughly the same time in the 70's, the product known as
Ingres was being developed by scientists at UC Berkeley. Funding for Ingres
came from a variety of sources, chief among them being various branches of the
military. The Ingres project used its own query language (QUEL), but SQL is
what has become the standard.
There were some significant differences between System R and
Ingres. The source code for System R was kept private, whilst that of Ingres
was freely available. System R was meant to be sold; Ingres was freely
distributed to the academic community. Additionally, System R, being produced
in the commercial arena, was competing with other resources within IBM. Ingres
was new and could go any direction as needed.
System R, did, however, create the opportunity for another
system to come along. IBM was sitting atop a gold mine and did not realize it.
Lawrence Ellison recognized the potential of what System R had to offer, and
that gave rise to the company and relational database system named Oracle.
In 1977, Lawrence Ellison, along with cofounders Bob Miner
and Ed Oates, founded a company named Software Development Laboratories. In
1979, as the renamed Relational Software Inc., the first commercial version was
released, running on the Digital PDP-11. Interestingly, the first released
version was Version 2 - the "2" conveying the idea that all the bugs had been
worked out since Version 1, of which, there was no such version. In 1983, the
company was renamed to Oracle (it was a name of a project Ellison had worked on
for the CIA). Today, Oracle is the largest relational database system vendor
and the world's second largest independent software company.
Originally designed to run on the VMS/VAX computer system,
Oracle today can be found on every version of UNIX, most versions of Windows,
and a slew of other platforms. In fact, Oracle currently lists support for 77
platforms under its "Certify & Availability" link on its support Web site
(MetaLink). The Oracle Web site shows a timeline of significant events and
Starting out as a pure database company, the company's chief
strength today is still in the RDBMS market. Oracle also has a successful line
of products used for development, presentation and reporting (Forms and
Reports); Web services (Application Server); and enterprise management software
(Oracle Applications, recently augmented by its acquisition of PeopleSoft).
For an interesting perspective of Oracle's beginnings, read The Difference Between God
and Larry Ellison: Inside Oracle Corporation: God Doesn't Think He's Larry
Ellison. The book paints a somewhat less than flattering view of Mr.
Ellison as a person, but as a shark in corporate waters, he is a very efficient
and successful killing machine. How many times can a company get away with
promising functionality that does not exist yet, or inflate/manipulate sales
figures without being caught by stockholders or regulatory agencies? If the
product is that good, it appears the answer to the first question has no limit.
The answer to the second question - well, that practice did catch up with
Oracle and it nearly sunk the company.
Perhaps the greatest selling strength Oracle possesses today
is its support for a wide range of platforms. SQL Server is very good in many
ways, and is relatively inexpensive when compared to Oracle. Unfortunately, for
SQL Server, it is now and forever tied to the Windows operating system.
Virtually all great inventions start with an idea. The music
industry's CD-ROM format, for example, started because a music aficionado
wanted a better way to hear and preserve what was on vinyl records (do you even
know the inventor's name?). Years and years went by before that idea caught on,
but when it did, it signaled the demise of the cassette tape.
What are those key moments in time and how can you (or will
you) recognize them when something brilliant is proposed? Does the
implementation of your idea sit on the shelf for a decade? Does your employer
take steps to downplay the significance of your proposal? And how hard would
you be willing to work to sell the idea? Your use of Oracle today is due in
large part to Dr. Codd's persistence and Lawrence Ellison's vision. What is the
next (or current) unappreciated model or IT-related invention waiting to be
Listed below are some references used in this article. If
you haven't explored the history of Oracle, these represent a good starting
point to begin your investigation.
Codd, E.F. "A Relational Model of Data for Large Shared Data
Banks." Communications of the ACM 13.6 (1970): 377-387. 20 May 2005 <http://doi.acm.org/10.1145/362384.362685>
United States. National Academy of Sciences. Funding a
Revolution: Government Support for Computing Research. Washington: National
Academy Press, 1999. 20 May 2005 <http://www.nap.edu/readingroom/books/far/ch6.html>
Wilson, Mike. The Difference Between God and Larry
Ellison: Inside Oracle Corporation: God Doesn't Think He's Larry Ellison. HarperCollins:
New York, 1998.
See All Articles by Columnist Steve Callan