Re: Synonyms and namespaces was: Re: DTD Notation raises a question

[Jonathan Borden <jborden@mediaone.net>]

John Cowan wrote:

> Jonathan Borden wrote:
>
>
> > Actually this is not true. SSE is not limited to humans, JC/CJ are.
>
>
> Fair enough.  SNOMED doesn't really handle the full range of
> veterinary terms.

Ah. SNOMED. One must be very careful in using SNOMED outside of pathology
(it is a product of the American College of Pathology). This is a really
important issue and one very close to my heart. A very practical problem
with using pathology terms (i.e. SNOMED codes) outside of describing a piece
of tissue which has been fixed in formalin, sliced up and examined under a
microscope (i.e. a live person) is that one may only be guessing at what the
pathology _would show_ if it were examined, and if it hasn't been then you
are making the very bad mistake of assigning the wrong name to an entity.

Assigning the wrong name, leads to bad problems like giving the wrong
medication, performing the wrong surgery etc etc. Most of the time it is far
better to assign a much less precise name, that is correct e.g.

"dementia" or "rapidly progressive dementia"

e.g. clinical:dementia -isCausedBy-> path:SSE

Well, I guess this gets back to the fact that all vocabularies are local...

-Jonathan