OASIS Mailing List ArchivesView the OASIS mailing list archive below
or browse/search using MarkMail.


Help: OASIS Mailing Lists Help | MarkMail Help

emergency message

[Date Prev] | [Thread Prev] | [Thread Next] | [Date Next] -- [Date Index] | [Thread Index] | [List Home]

Subject: RE: [emergency] Naming Conventions

Title: RE: [emergency] Naming Conventions
For Human Physical Descriptions, as pointed out in HumanML, the
public safety industry by and large relies on standard descriptors
that are not at the level of detail as in the medical industry.  Overlaps
there will have to be handled because the records systems for
public safety (ie, police RMS) don't need the level of detail of
medical.  I'd be surprised if that changes.   That was one
reason for the use of abstract types in the HumanML schema.
The best we could do was pop up a metalevel and provide
supertypes, so to speak.  I note that JusticeXML appears to
have taken a similar path, and I think that is what Elliott
may be pointing to.  Naming conventions, per se, won't
fix the problem of having two concrete types that really
are different types.
Sorry to belabor this.  All I meant to convey was that
vocabulary based standards are reasonable, but the
more one imports and includes the greater the likelihood
of collisions if the domains are not aligned.   It isn't always
practical to take schemas from a different industry/domain
and attempt to factor it into another. That is why the
planning documents for a schema project of any kind
have to be quite specific about the domain, and not,
as is traditional in some projects, written purposefully
vague.  One should not late bind domains.
Good Friday to all.
-----Original Message-----
From: Rex Brooks [mailto:rexb@starbourne.com]

Thanks for the input Len. Your experience is valuable, as usual. My own concerns in this arena have been fairly limited to the overlaps with HumanML (especially the Medical connection) and GEOVRML, and providing new kinds of tools, i.e. 3D tools, that will use the standards this TC develops, among others.

I had not given the database issue much thought, but what you say makes sense.

[Date Prev] | [Thread Prev] | [Thread Next] | [Date Next] -- [Date Index] | [Thread Index] | [List Home]