> Whether you store '123ABC' in CHARSET OCTETS or in
> CHARSET ASCII or in CHARSET NONE, doesn't make any
> difference in size. All will take 6 bytes.
>
> Perhaps you are misled by the automatic display-as-hex
> in ISQL, but using it this way, you cannot gain any size advantage.
>
> You can use the compres your keys to fit into 8 bytes of
> CHARSET OCTETS, but this isn't pretty at all.

Hi Peter,

Granted, I am a bit confused... between the Charsets, Collations,
Case Sensitivity, and multi-byte characters... my head is a bit
swimming...

I am getting used to FB, but I need to finalize my ERD very soon and
I just wanted to make sure that my decisions now would not come back
to "haunt" me later...

When I create a brand spanking new Database, I am prompted for
a "Default" Charset... should I leave it a None or select a charset?
(we have never neede any accents or multi-byte chars, nor do I see
this changing... however, I do not want to be short-sighted for
little gain either...)

I always hear about horror stories about users "pasting" in some text
from a MS Word document containing a bunch of weird characters and
unknowningly causing trouble within the db...

I am trying to make my PK and FK domains as "quick and lite" as
possible... so from reading past posts, I thought OCTETS was
the "lightest" way to go... however, I noticed that ASCII does not
behave as "naughty" as OCTETS... is this better... or should I not
specify any character set?

I sure appreciate any advice (I could sure use Helen's book right
now! :)

--Raymond