|
On 12/16/14 10:52 AM, Mike Bennett
wrote:
Hi Kingsley,
I just need to clarify a couple of things with regard to those
links:
Naming in the FIBO standard has UpperCamelCase for classes and
lowerCamelCase for properties, as per OWL naming conventions.
Business facing labels are upper and lower case with spaces, to
be human readable (as previously noted, we were never
"discussing matters with a technical audience").
In this situation you solve the problem as follows:
<#SomeClass>
a owl:Class ;
rdfs:label "SomeClass";
skos:altLabel "Some Class";
skos:prefLabel "Some Class" ;
rdfs:comment "{language-tagged-human-reader-oriented-sentences}" .
A consumer of the data above can then apply reasoning and inference
en route to constructing <a
href="">{value-of-annotation-property}</a>, ditto
language tag preferences so we drop the habit of only writing for
English speakers i.e., use RDF language tags to aid crowd-sourcing
of content in many human natural languages . That's how we solve the
issue in our Linked Data browser which will choose the value of
skos:prefLabel, and determine language preference etc..
Based on what's presented at
<http://linkeddata.uriburner.com/c/9BN322EX>, Here's some
nanotation to actually fix the problem, using the hueristic above,
via this post:
## Nanotation Start ##
<http://www.omg.org/spec/EDMC-FIBO/FND/Utilities/AnnotationVocabulary/nameOrigin>
rdfs:label "nameOrigin" ;
skos:altLabel "name origin" ;
skos:prefLabel "nameOrigin" .
## Nanotation End ##
This is exactly what you are
recommending.
Okay.
As far as I can see, the COSMO
resource you have linked to (which is not FIBO!) follows the
same convention as we do, with some minor differences in
convention.
I've de-selected the COSMO named graph [1][2].
IMPORTANT: The two links you have given are not links to FIBO or
to the subsumption tree of any part of FIBO. These links both go
to a re-framing of FIBO concepts within COSMO, including
sub-class assertions which are not part of FIBO and never have
been.
I have not yet had the time to look at the COSMO work and
comment on it, but please be aware that FIBO itself is framed
within its own mid level and upper ontologies all the way up to
Thing and no reference is made in the FIBO standard to the COSMO
hierarchy of concepts.
Please can you modify your description of these concepts
accordingly.
What is the canonical URI of the FIBO Ontology? Should I still be
referring to the wrong ontology or version?
Links:
[1] http://linkeddata.uriburner.com/c/9DFXMZEO -- FIBO modulo COSMO
[2]
http://linkeddata.uriburner.com/fct/rdfdesc/usage.vsp?g=http%3A%2F%2Fwww.omg.org%2Fspec%2FEDMC-FIBO%2FFND%2FUtilities%2FAnnotationVocabulary%2FnameOrigin
-- Showing how I exclude COSMO from the FIBO ontology in my browser
[3] http://linkeddata.uriburner.com/c/9BN322EX -- example of what's
solved using the label heuristic I suggested .
Kingsley
Many thanks,
Mike
On 16/12/2014 15:24, Kingsley Idehen wrote:
On
12/16/14 9:19 AM, Mike Bennett wrote:
Hi John,
As it happens, there was a conversation about "Thing" versus
"Entity" on
this very forum at the time I was thinking about what words to
use.
I came down on the side of "Thing" for the following reasons:
1. I wanted to make it explicitly clear to any reader of the
model, that
this was not "Yet Another Data Model" but a model of real
things.
Sometimes it helps to call upon the Ango-Saxon language for
words that
have not been muddied by previous usages. I had previously
tried to
present computationally independent models of concepts without
doing
this, and the result was always that they were reviewed as
though they
were data models. How to make it clear that the model is
intended to be
a model of real things in the world? A new, untainted word was
required.
2. I decided to use the OWL language, against the advice from
many
well-informed folks including yourself, because we needed
something
which was recognized in the technology community and which
would have
tooling going forward, and because the OWL language, having
"Thing" at
the top, made it explicitly clear that it was a model of
things in the
world and not a model of someone's data. So it matched my
instinct to
use the word "Thing" as above.
3. The word "Entity" which you proposed at the time, while
more
precisely correct, suffered from the problem which Pat notes
in a later
email, namely that it has a well-worn meaning in the data
modeling
world. The last thing we needed was something that could be
mistaken for
a model of database "entities".
Of course the word "Thing" is often interpreted as being a
concrete,
continuant, independent thing. To get around this, we made
sure that
everything in the model was framed in terms of three sets of
partitions:
Independent Thing / Relative Thing / Mediating Thing;
Continuant Thing /
Occurrent Thing; and Concrete / Abstract. By appending the
word "Thing"
to those, it is hopefully clearer that the term "Thing" at the
top of
the model encompasses all of those, and not simply those
things you can
poke with a stick.
Hi Mike,
Assuming I have the right version, I have a FIBO viewing page
[1] that could assist everyone in this burgeoning debate. Thus
far, in regards to definitions of the *nature* of entities and
relations described by this ontology, I am not seeing the
owl:Thing anomaly [2].
In my experience "Entity" is better than "Thing" especially,
when discussing matters with a technical audience. Personally,
an Ontology should be constructed for an audience that can
understand its technical underpinnings. Alternative approaches
always lead to problems.
An few aesthetic tweak suggestions for FIBO:
1. Class Names should start with upper case and phrases should
be CamelCase
2. Property Names should be in lower case and camelCase re.,
phrases.
The above makes a world of difference, it even aids those who
are encountering ontologies (for the first time) from other
realms e.g., RDBMS and DBMS folks in general.
Links:
[1] http://linkeddata.uriburner.com/c/8JZYHP
-- FIBO overview via a LInked Data browser
[2] http://linkeddata.uriburner.com/c/8JZX7W
-- FIBO autonomous agent subsumption tree root .
_________________________________________________________________
Message Archives: http://ontolog.cim3.net/forum/ontolog-forum/
Config Subscr: http://ontolog.cim3.net/mailman/listinfo/ontolog-forum/
Unsubscribe: mailto:ontolog-forum-leave@xxxxxxxxxxxxxxxx
Shared Files: http://ontolog.cim3.net/file/
Community Wiki: http://ontolog.cim3.net/wiki/
To join: http://ontolog.cim3.net/cgi-bin/wiki.pl?WikiHomePage#nid1J
--
Mike Bennett
Director, Hypercube Ltd.
89 Worship Street, London EC2A 2BF
Tel (UK): +44 20 7917 9522
Tel (US): +1 646 583 2095
www.hypercube.co.uk
_________________________________________________________________
Message Archives: http://ontolog.cim3.net/forum/ontolog-forum/
Config Subscr: http://ontolog.cim3.net/mailman/listinfo/ontolog-forum/
Unsubscribe: mailto:ontolog-forum-leave@xxxxxxxxxxxxxxxx
Shared Files: http://ontolog.cim3.net/file/
Community Wiki: http://ontolog.cim3.net/wiki/
To join: http://ontolog.cim3.net/cgi-bin/wiki.pl?WikiHomePage#nid1J
--
Regards,
Kingsley Idehen
Founder & CEO
OpenLink Software
Company Web: http://www.openlinksw.com
Personal Weblog 1: http://kidehen.blogspot.com
Personal Weblog 2: http://www.openlinksw.com/blog/~kidehen
Twitter Profile: https://twitter.com/kidehen
Google+ Profile: https://plus.google.com/+KingsleyIdehen/about
LinkedIn Profile: http://www.linkedin.com/in/kidehen
Personal WebID: http://kingsley.idehen.net/dataspace/person/kidehen#this
|
smime.p7s
Description: S/MIME Cryptographic Signature
_________________________________________________________________
Message Archives: http://ontolog.cim3.net/forum/ontolog-forum/
Config Subscr: http://ontolog.cim3.net/mailman/listinfo/ontolog-forum/
Unsubscribe: mailto:ontolog-forum-leave@xxxxxxxxxxxxxxxx
Shared Files: http://ontolog.cim3.net/file/
Community Wiki: http://ontolog.cim3.net/wiki/
To join: http://ontolog.cim3.net/cgi-bin/wiki.pl?WikiHomePage#nid1J (01)
|