On Fri, May 11, 2018 at 4:48 AM, Paul A Jungwirth
<pj@illuminatedcomputing.com> wrote:
> I'm traveling and can't write much more at the moment, but I'll try to
> reply more fully in a week or two.
Sorry it took awhile to continue this discussion! If people are
interested in implementing temporal features in Postgres, I wrote an
annotated bibliography here about the research:
https://illuminatedcomputing.com/posts/2017/12/temporal-databases-bibliography/
A shorter version, focused on the patches offered by Dingös et al, is here:
https://www.mail-archive.com/pgsql-hackers@postgresql.org/msg324631.html
There is a lot of history, and temporal features tend to be pretty
drastic and far-reaching (e.g. a "temporal primary key" is more like
an exclusion constraint). It's really worth reading a bit from the
research. The Date/Darwen/Lorentzos book is very good, although I'd
take some of it with a grain of salt (e.g. no nulls).
I think SQL:2011 is so limited & over-specific, it would be great to
offer it on top of some more generalized infrastructure. The Dingös
work is great, so that's what I'd like to see. With their new
operators, it would be very easy to build the stuff from the standard.
This paper is only 10.5 pages (not counting bibliography & appendix)
and easy to understand:
https://files.ifi.uzh.ch/boehlen/Papers/modf174-dignoes.pdf
Paul
