Re: Temporal database - no end date

-CELKO- wrote:
> You had better look fast. This book was a failure on several levels.
> Date uses the Chronon model of time, which was discredited by Zeno's
> paradoxes. The vast majority of temporal database researchers use the
> half-open interval model for time.

And as we all know: in science, everything is determined by majority vote.

> A chronon is a quantum of time that originally had a fad in physics and
> the term spread -- remember the tachyon fad (a hypothetical particle
> that always travels faster than the speed of light) that disappeared
> with a one page logical proof of impossibility? Physics is bad about
> that kind of thing.

And the logical impossibility of tachyons is of course directly applicable to how we choose to model limited aspects of reality in our database systems.

> Date's book invents its own language rather than giving SQL, so
> programmers cannot use it in their apps. You also wind up with the
> weird thing that sequentially numbered salesman can be handled with the
> same operators as durations.

And in Java, you can use the same operators on a list of Apple objects as on a list of Orange objects. Weird!

> So, the book was rejected by both the working and academic communities.
> Sales were awful in spite of Date's & Darwen's name on it. What really
> did it in was an unfortunate attack on Snodgrass by Date that got to be
> minor scandal in the RDBMS community and an embarrassment to
> Morgan-Kaufmann.

A scandalous attack? Can you elaborate?

-- 
Jon