[tmql-wg] Basename duplication ... ?
Thu, 24 Jun 2004 07:16:06 +0000
I have a basic -though I hope not trivial / frequently answered- question.
The question relates to TMs and not TMQL, but it has arisen in relation to
what I feel is a fundamental point about TMQL(s). In trying to better express
the latter, I did some research and found the following:
"F.6.2 Topic name duplicate suppression
"[If] basename B1 ... is equal to [basename B2] [in the same
scope] [a] basename B3 is created which contains a base
name equal to [basename B1] ... . All variants of B1 are added
to B3. All variants of B2 are added to B3. ... B1 and B2 no
longer exist." (ref.1)
"It is allowed in topic maps for different topics to have the
same name, something that taxonomies and thesauri do
not allow. In practice this is something that occurs all the
time, and so it is important to support it. Many classification
systems avoid duplicate names by including some dis-
ambiguation in the name itself. For example, Paris the city
and Paris the hero of Greek mythology may have been
differentiated as 'Paris (France)' and 'Paris (Greek myth.)'. In
topic maps this is not necessary, and the types, occurrences,
and associations of the topics will generally distinguish them
([footnote 4.] "You may have heard of the "topic naming
constraint", which is a rule in topic maps that only allows
duplicate names in certain limited cases. This rule has now
been removed from topic maps, so the statement is true.")
So can one have two _different_ topics in the same (set of) topicmap(s) with
the same basename?
Apologies again for being off topic, but it has (obvious) relevance to
(1) "XML Topic Maps (XTM) 1.0." xtm1-20010806.html v 1.16 14:31:44 pepper at
_"Informative"_ Annex F.
I have only now (as I type) noticed that this Annex is _not_ normative. Even
though the (presumably Normative) section section 3.10 in the body of the
document requires merging "according to the _rules_ in Annex F" (emphasis
added) and the Annex itself, declares (section F.1) "This annex describes the
minimal requirements for a _conformant_ XTM processor. .. [a] processor is
_required_ to ...." (emphases added).
(2) "Metadata? Thesauri? Taxonomies? Topic Maps! Making sense of it all"
Lars Marius Garshol www.ontopia.net/topicmaps/materials/tm-vs-thesauri.html