<XML> Abstract Data Model for Taxonomy
Jean-Marc Vanel
jmvanel at FREE.FR
Thu Feb 24 14:59:22 CET 2000
Leigh Dodds a écrit :
> 1. Firstly is it possible to express items like
> "Feature often present by mis-identification"
Sorry, my english is too weak. Does this mean a character value that is
often reported by confusion with another taxon ?
> I note you've already highlighted seasonal variations as a problem.
My solution: the default for a Property is to record the state for:
- middle of active season
- adult specimen
Otherwise, create a new Property. Note that beside the specified
sub-elements of tax:Property (appliesTo, appliesToClasses, name), it is
possible to add extra information like, say,
<bio:season>spring</bio:season>. This is a property (rdf:Property) about a
property (tax:Property).
> To that I'd add geographical,
My (simple) solution: if anything is different in some region, create a
lower level tax:TaxonomicClass, containing just the differing Features;
according to model, other Features will be inherited.
> 2. It seems to presuppose a Linnean viewpoint i.e. Kingdom,
> Phylum, etc.
I didn't write that, in fact this model is so generic that it could almost
be applied to car models!
> Gregor has previously pointed out that the
> reality is much more complex. Multiple hierarchies
> can be produced. For example, how would the model be used to express
> DNA/Protein data? Physiological versus Molecular hierarchies?
And belonging to phytosociological associations, also. Although I didn't
think of that, this is no problem. You just have to publish on the Web a
Schema with, say, a ProteinClass and a ProteinSubClass, declaring that both
are rdfs:subClassOf of tax:TaxonomicClass, and connecting them using
tax:lowerClasses and tax:upperClass, etc. And hoping that many will connect
to your Schema... Well, this is the freedom of the Web, but I hope that a
standard will be found. But I wanted to show that extendibility is
perfectly possible, with the possibility that the new semantic is declared
as such, so that tools can understand the new semantic.
> 3. I assume that the list of Feature/Properties will not be
> fixed, but can be extended at will?
Yes
> However I'm concerned that the presence of items like 'TaxonomicClass' in
> the model doesn't
> capture the range of flexibility that Kevin, Gregor and the others
> have stated as a requirement.
Is this requirement like the Molecular hierarchies, or something else ?
Cheers
Jean-Marc
--
<person>
<first_name>Jean-Marc</first_name>
<name>Vanel</name>
<project>Worlwide Botanical Knowledge Base -
making botany available on Internet
<a href="http://wwbota.free.fr/" >site</a>
</project>
<homePage>http://jmvanel.free.fr/</homePage>
<a href="mailto:jmvanel at free.fr">mail (eventually put "wwbota" in subject
to route your mail in relevant folder)</a>
</person>
--------------F6D7B7F21F8B3633F46915EA
Content-Type: text/html; charset=us-ascii
Content-Transfer-Encoding: 7bit
<!doctype html public "-//w3c//dtd html 4.0 transitional//en">
<html>
Leigh Dodds a écrit :
<blockquote TYPE=CITE>1. Firstly is it possible to express items like
<br>"Feature often present by mis-identification"</blockquote>
Sorry, my english is too weak. Does this mean a character value that is
often reported by confusion with another taxon ?
<blockquote TYPE=CITE>I note you've already highlighted seasonal variations
as a problem.</blockquote>
My solution: the default for a Property is to record the state for:
<br>- middle of active season
<br>- adult specimen
<p>Otherwise, create a new Property. Note that beside the specified sub-elements
of tax:Property (<tt>appliesTo, appliesToClasses, name</tt>), it is possible
to add extra information like, say, <tt><bio:season>spring</bio:season>.
</tt>This is a property (<tt>rdf:Property</tt>) about a property (tax<tt>:Property</tt>).
<blockquote TYPE=CITE>To that I'd add geographical,</blockquote>
My (simple) solution: if anything is different in some region, create a
lower level <tt>tax:TaxonomicClass</tt>, containing just the differing
Features; according to model, other Features will be inherited.
<br>
<blockquote TYPE=CITE>2. It seems to presuppose a Linnean viewpoint i.e.
Kingdom,
<br>Phylum, etc.</blockquote>
I didn't write that, in fact this model is so generic that it could almost
be applied to car models!
<blockquote TYPE=CITE>Gregor has previously pointed out that the
<br>reality is much more complex. Multiple hierarchies
<br>can be produced. For example, how would the model be used to express
<br>DNA/Protein data? Physiological versus Molecular hierarchies?</blockquote>
And belonging to phytosociological associations, also. Although I didn't
think of that, this is no problem. You just have to publish on the Web
a Schema with, say, a ProteinClass and a ProteinSubClass, declaring that
both are <tt>rdfs:subClassOf of tax:TaxonomicClass, </tt>and connecting
them using tax:low<tt>erClasses </tt>and tax:<tt>upperClass</tt>, etc.
And hoping that many will connect to your Schema... Well, this is the freedom
of the Web, but I hope that a standard will be found. But I wanted to show
that extendibility is perfectly possible, with the possibility that the
new semantic is declared as such, so that tools can understand the new
semantic.
<blockquote TYPE=CITE>3. I assume that the list of Feature/Properties will
not be
<br>fixed, but can be extended at will?</blockquote>
Yes
<blockquote TYPE=CITE>However I'm concerned that the presence of items
like 'TaxonomicClass' in the model doesn't
<br>capture the range of flexibility that Kevin, Gregor and the others
<br>have stated as a requirement.</blockquote>
Is this requirement like the Molecular hierarchies, or something else ?
<br>
<p>Cheers
<br>Jean-Marc
<p>--
<br><person>
<br> <first_name>Jean-Marc</first_name>
<br> <name>Vanel</name>
<br> <project>Worlwide Botanical Knowledge Base -
<br> making botany available on Internet
<br> <a href="<A HREF="http://wwbota.free.fr/">http://wwbota.free.fr/</A>" >site</a>
<br> </project>
<br> <homePage><A HREF="http://jmvanel.free.fr/">http://jmvanel.free.fr/</A></homePage>
<br> <a href="<A HREF="mailto:jmvanel at free.fr">mailto:jmvanel at free.fr</A>">mail (eventually put "wwbota"
in subject to route your mail in relevant folder)</a>
<br></person>
<br> </html>
More information about the tdwg-content
mailing list