- From: Alistair Miles <a.j.miles@rl.ac.uk>
- Date: Mon, 13 Feb 2006 15:19:09 +0000
- To: "Svensson, Lars" <svensson@dbf.ddb.de>
- CC: public-esw-thes@w3.org
Hi Lars,
The very early 'Thesaurus Interchange Format' on which SKOS is based had
a property called 'classificationCode' which was intended for notations.
SKOS Core used to have a very badly named property called
'skos:externalID' which was intended to provide a non-URI identifier for
a concept ('external' as in 'outside the web' - that was my bad idea,
apologies - see section 3.4 in [1]). It replaced the
'classificationCode' property. In the 2004 SWAD-Europe report 'RDF
Encoding of Classification Schemes' [4] I used a sub-property of
skos:externalID to represent a PACS classification code.
Later in 2004 I made a proposal to replace the 'skos:externalID'
property with something call 'skos:localID' - see [2]. The idea was that
the value of a 'localID' is a literal that identifies a concept uniquely
within the scope of a thesaurus or classification scheme. We ultimately
decided that the dc:identifier property could be used for this purpose
(see [3]) and decided to deprecated skos:externalID and replace it with
dc:identifier. The idea was that each thesaurus or classification scheme
could create a sub-property of dc:identifier to capture their own
notations, although we never added anything to the SKOS Core Guide to
indicate how to do this.
However, I have been thinking for a while that a SKOS 'notation'
property would be a good idea after all. Such a property would have a
meaning that goes beyond dc:identifier, but is common to all
classification schemes and thesauri.
I would be happy to support a proposal for a 'skos:notation' property. I
would suggest that the property had the following meanings:
- The main purpose of the property is to provide a human-friendly
identifier for the concept, whose value is not a recognisable word or
collocation of words from any natural language.
- The value of the property should be a literal that uniquely
identifies the concept within the scope of a given thesaurus or
classification scheme.
- The value should be a typed literal, where the datatype defines both
the scope of reference, and the lexical space of allowed values.
So e.g. the classification scheme with URI <http://example.com/myScheme>
defines a datatype for notations and allocates the URI
<http://example.com/myScheme/Notation> to this datatype. It then uses
the skos:notation property as in:
ex:concept01 a skos:Concept;
skos:prefLabel 'love'@en;
skos:altLabel 'affection'@en;
skos:notation '10.024.354'^^<http://example.com/myScheme/Notation>.
Cheers,
Al.
[1] http://www.w3.org/2001/sw/Europe/reports/thes/1.0/guide/20040504/
[2] http://lists.w3.org/Archives/Public/public-esw-thes/2004Aug/0064.html
[3] http://lists.w3.org/Archives/Public/public-esw-thes/2004Sep/0002.html
[4] http://www.w3.org/2001/sw/Europe/reports/thes/8.5/
Svensson, Lars wrote:
> [This question has been on the list before [1][2], but I found no
> ultimate answer and thus feel free to post again. Sorry for any
> inconvenience!]
> Classification schemes like the Universal Decimal Classification, MCS,
> or PACS usually have two human-readable labels: One containing a
> notation (usually a combination of digits and letters following a
> specific syntax, so that it's possible to see super-/subclass
> relationships and ideally to identify precoordinated notations), the
> other one being kind of free-format text only. Both labels are intended
> for human use. For the second one, I would probably use skos:prefLabel,
> but is there a specific label for use with notations? I wouldn't be
> comfortable to use skos:altLabel, since the semantic is quite different,
> but I can't find a skos:notation property or anything similar. Have I
> missed anything?
>
> Thanks
>
> Lars
>
>
> [1] http://lists.w3.org/Archives/Public/public-esw-thes/2005Jan/0075
> [2] http://lists.w3.org/Archives/Public/public-esw-thes/2005Feb/0033
>
--
Alistair Miles
Research Associate
CCLRC - Rutherford Appleton Laboratory
Building R1 Room 1.60
Fermi Avenue
Chilton
Didcot
Oxfordshire OX11 0QX
United Kingdom
Email: a.j.miles@rl.ac.uk
Tel: +44 (0)1235 445440
Received on Monday, 13 February 2006 15:19:27 UTC