Re: Extending my question. Was: The relational model and relational algebra - why did SQL become the industry standard?

From: Mikito Harakiri <>
Date: Tue, 18 Feb 2003 14:32:23 -0800
Message-ID: <lBy4a.20$>

"Steve Kass" <> wrote in message news:b2u9bg$50p$  > Mikito Harakiri wrote:
> >This is just a series of random definitions, well spiced with curly
> >bracketed symbols. Or is there any wonderful theory about it? Where in
> >did you see "We have 10 unrelated alternative definitions, pick anyone
> >like"?
> Um, everywhere? Pick up half a dozen graph theory textbooks,
> and you're likely to see nearly as many definitions. Some analysis
> books define irrationals in terms of Dedekind cuts, others with
> limits.

Plus there is nonstandard analysis. Each of these is like a separate theory, however.

Now when we write


what theory does it belong to? What can we deduce from this "definition"? How does it relate to


for example?

> Riemann integrals get defined in all kinds of ways.

I know only one.

> The
> function f(z) = e^z has plenty of useful definitions.

Deduced from each other.

> Fortunately,
> mathematics is well-enough founded that the many alternate definitions
> of a construct are provably equivalent. I would be surprised to
> hear any mathematician suggest that the existence of many equivalent
> foundational definitions of a concept indicates that the concept is
> useless. There is no "one way" to define a mathematical idea using
> sets.
> To throw away all of mathematics because there are equivalent
> formulations of ideas makes as much sense as to throw away
> numbers because Europeans, Chinese, and Arabs write them
> differently.

I think we agree on all that. We disagree upon how "well founded" or "well established" set reduction is. Received on Tue Feb 18 2003 - 23:32:23 CET

Original text of this message