W3C home > Mailing lists > Public > public-esw-thes@w3.org > April 2014

Fwd: SKOS comment - How to better explain why skos:broader is not transitive

From: David Booth <david@dbooth.org>
Date: Fri, 04 Apr 2014 15:54:25 -0400
Message-ID: <533F0DF1.8040806@dbooth.org>
To: public-esw-thes@w3.org
-------- Original Message --------
Subject: SKOS comment - How to better explain why skos:broader is not 
transitive
Date: Fri, 04 Apr 2014 15:14:44 -0400
From: David Booth <david@dbooth.org>
To: public-swd-wg@w3.org

The fact that skos:broader is not transitive is confusingly
non-intuitive, because it is quite natural to assume that if A
skos:broader B, and B skos:broader C, then A skos:broader C.

The confusion is well known, and the primer duly explains how
skos:broader is *not* transitive, but skos:broaderTransitive *is*
transitive:
http://www.w3.org/TR/skos-primer/#sectransitivebroader

However, I think I just realized *why* this is so non-intuitive, how
this problem could have been avoided, and how the primer (and other
documents) can now explain it better.  And it is extremely simple.

The problem is that the *name* of the skos:broader predicate is
misleading.  The concept it formally denotes is *not* the concept that B
is broader than A.  Rather, it is that B is *immediately* broader than A
(in a potentially larger hierarchy of broader-ness).  If the predicate
had been named skos:immediatelyBroader (or something similar) then the
reader would much more readily realize that the predicate does *not*
denote the concept of broader **in general** (which would be
transitive), but the concept of a concept being *immediately* broader
than another concept.

Since it is too late to rename this predicate, I suggest that the next
version of the primer -- and any new documents on SKOS -- explicitly
explain the non-transitive nature of skos:broader this way, and
explicitly acknowledge that the decision to name this concept
"skos:broader" was admittedly misleading, for this reason.  I think this
will help a lot of readers understand it more easily.  I know that I
personally puzzled over this for a long time, whereas I would have
understood it right away if it were explained this way.

Thanks,
David
Received on Friday, 4 April 2014 19:54:53 UTC

This archive was generated by hypermail 2.4.0 : Friday, 17 January 2020 19:46:36 UTC