|
|
|
In the early eighties I was an undergraduate student at the University of Helsinki. This was in the middle of the Cold War. The placement of Finland in the East/West partition was a bit fuzzy, so Helsinki was one of the few Western places academics from the East block could easily visit. In Helsinki we had a steady stream of visitors from the Soviet satellites on various ``friendship'' exchange programs. Nobody paid much attention to these visitors. However, I heard one day that one them was working on ``null values,'' something which I was interested in. So I went to talk to Tomasz Imielinski, a ``friendship'' visiting graduate student from Warsaw. Tomasz explained the tables-construct to me, and I was sold. Meanwhile, as Jeffrey D. Ullman and Alberto O. Mendelzon describe in their reminiscences, the chase and an arsenal of other tools were being developed at Princeton.
Later I wrote my PhD-thesis on tables, and worked on incomplete information with several database people. Today, incomplete information is experiencing a renaissance, due to its central role in information integration applications. Serge Abiteboul's recent comment (PODS'99, invited talk) [2] that ``[tables] are a typical representative of great tools that remain unfortunately mostly unused'' is somehow emblematic of the story.
Copyright © 2000 by the author(s). Review published with permission.