> > I have none. Never said I did.
>
> Fine.
>
> >
> > I only made a remark about the TIME datatype that when you are
> > applying math to it (eg: substracting one TIME from another), you
> > don't always get a TIME as the result.
>
> Time - Time gives a time in seconds(plus milliseconds).

No it doesn't... It gives a "period of time", a "duration" ...

TIME (the datatype) means: time of day.

> this seconds doesn't mean of meridian info of course.
> If you think the TIME must be meridian number, i say nothing.
> There was TIME and TIME clause before coordinate systems i think.

This is not what the datatype "TIME" is about. It's "time of day".

I think we can agree that "14:00 - 08:00" doesn't give us "time of day
06:00", does it?

> > Then you started about time being just a numerical value and you
> can
> > do anything you like with it.
> >
> > Which is wrong on the conceptual level. A TIME - TIME doesn't
> > result in a TIME value. Period. Yes, it fits the internal
> implementation,
> > but it's still not a TIME value.
> >
>
> Here is the difference between me and you:
> I'm using TIME datatype as an frequnece ruler, but you not.
> TIME is a frequence info in my mind. Thats all.
> It is my opinion only, this is not a must.
>
> If my solution doesn't fit Arnoldo's or my customers or my friends
> or science requeriments than i am ready to change my minds.
> I should show my any calculation mistake with math, if there is.
> This makes me happy more than 4 times.

See above.

With regards,

Martijn Tonies
Database Workbench - tool for InterBase, Firebird, MySQL, Oracle & MS SQL
Server
Upscene Productions
http://www.upscene.com
Database development questions? Check the forum!
http://www.databasedevelopmentforum.com