Richard Knapp wrote:
> IMHO it seems a better fit to have
> a common format for all the members and roles instead of having a
> special one for those on instrument and those doing other duties.
Okay, I'm glad you stated it that way. That is why I asked.
When one goal is have the ability to record for history _anyone_ who
was on that survey party, then person instrument tends to look a bit
limited thus everyone should definitely assign them a role and list
each person on the team thus crosslinking to the instrument
information.
devinkouts@earthlink.net wrote:
> We're in agreement. I strive for maximum flexibility and try to avoid
> narrowing the options a user has for entering data into the XML file.
> How could I ever hope to guess all the lables a role on a cave survey
> could be given (e.g. water boy)? The software developers with whom I
> developed the Cave Survey data in XML requirements beat me up severely
> over the flexibility issue (figuratively).
Don't get me wrong, I'm not suggesting a cross link is a bad thing,
in fact, I'm sure John Halleck could easily find data in caves we were
both involved in typing data for which contain "Water Boy" like roles.
Flexibility can have it place, but I believe it would be more useful
if you eventually specify some values of roles better than just the
present "DTD" whcih says that role is any string,
i.e. CDATA. In fact you need to eventually do better in
order to provide the cross links you claim you wish to have and I agree
is a great thing.
role=bookie, role=sketcher and role=book all look fine for many humans,
but I believe it is more useful when the values are constrained to
define certain predefined values that connect each instrument (probably
just use the same names as the measurement names), maybe a few other
predefined terms for existing roles. Additional free-form
values would also be allowed so that you wouldn't be limited to
only recording the known roles.
Good stuff,
-Paul
This archive was generated by hypermail 2b30 : Mon Apr 02 2001 - 18:00:01 CEST