[tcs-lc] enumeration of relationsihp types

Nozomi Ytow nozomi at biol.tsukuba.ac.jp
Wed May 11 06:44:42 PDT 2005


Hi Roger,

> The primary use of the schema is to communicate between machines

Agreed, so the exchange schema should have a sufficient capability
to express necessary data.  Both sufficient and necessary depends
consumer and provider.  I hope that the extension mechanism is used
only 'exceptional' cases to be harmless.


> Homonyms are dealt with as part of the NameObject. If a name is a later 
> homonym of another name then you can add a pointer to the older
> name.

I'll think of it more deeply.... would be right...

> Could you give an example of a relationship we haven't thought of?

I picked up following relationship types from Flora of Japan as
unmappable (or, uneasy but mappable):

based on
comb. nov.  (by NameObject?)
equivalent  (varnacular of?)
excl. fig.
in syn. 
not seen
refer 
stat. nov.
synonym p. p. (synonym + include?)
nom. nud.
homonym  (by NameObject?)
nov.
nec  (exclude?)
corrected spelling

Some may be coverd by NameObject, some may be covered by combination
of two or more relationship types, as guessed above.  Some are useless
for some consumers, but it depends on counsumers decisions.  I prefer
to export them if possible because I can't tell its value.

Cheers,
JMS
--
Dr. Nozomi "James" Ytow
Institute of Biological Sciences / Gene research center
University of Tsukuba
Tsukuba, Ibaraki 305-8572
Japan


More information about the Tcs-lc mailing list