Permaculture Database

[Paul d'Aoust <paul@heliosville.com>]

On Fri, 2007-30-03 at 11:47 -0700, Sean Maley wrote:

> What happens when you find that embarrassing typo after all of the
> fields have been populated and everyone is clamoring to enter data?

Although I like the idea of using some human-readable token as a primary
key, I have to agree with Sean. Typos would cause huge problems, and
what would you agree on for the name? The Latin names are always
changing; for example, the Chilean guava goes by Myrtus ugni, Ugni
molinaceae, and another name that escapes me. I think Ugni molinaceae is
the canonical name now, but that's not constant. Common names are just
as inconstant, and are a bit parochial (not that Latin names aren't, I
suppose...)

> In a combined data set, I'd think the ITIS number would be better,
> since PFAF only deals with plants and we need to combine all species.

'course, if we were using this as a primary key, this prevents soil
characteristics, inputs and outputs, fertilisers, etc from being
first-order objects in the taxonomic tree. I think a guid or an auto_inc
would be best.

> Feel free to dig up my star schema discussions from the past.  I'm
> just a DBA, so maybe I don't know any better.

hopefully you won't mind my being lazy and asking you one quick
question: were you arguing for or against star schemas? I know you
mentioned OLAP; is that denormalised approach what you mean by star
schemas? (I've never had any experiences with OLAP methodologies before.
They frighten me.)