RE: Cases

I admit the subtlety. This isn't really about "laymen's terms", though,
it's more about human-readable rdfs:labels that domain experts should be
assigning in their ontologies.

It ends up being a quibble over the phrase "the target is not really
fixed". The "target" is a broad intuitive understanding of strong
rdfs:labels like "has as the subject" combined with consistently
preserving these terms in HTML representations.

It's an idealistic argument and I have no doubt that UX designers will
dumb-down the terms for aesthetic reasons regardless of the consequences
for improved Google ranking and searchability.

I honestly won't be offended if the use case gets dropped.

Jeff 

> -----Original Message-----
> From: public-xg-lld-request@w3.org [mailto:public-xg-lld-
> request@w3.org] On Behalf Of Antoine Isaac
> Sent: Thursday, January 13, 2011 4:41 PM
> To: public-xg-lld
> Subject: Re: Cases
> 
> Jeff, the case is subtle *and* significant.
> The issue is that by "alignment" we are rather aiming at cases where
> the several artifacts (e.g., thesauri) have to be connected together.
> Here the "human vocabulary" you mention would be too immaterial a
> notion to fit the cluster: you have a source but the target is not
> really fixed.
> 
> As mater of fact I have a colleague (Rinke Hoekstra) at the university
> who worked on alignming expert legal terminology to laymen's terms for
> the same domain. I guess this fits your general goal. But these laymen
> terms were encoded in a vocabulary, which changes the situation quite
a
> bit.
> 
> Antoine
> 
> > Antoine,
> >
> > To my mind, the Subject Search use case is "vocabulary alignment" in
> the
> > sense of aligning RDF vocabularies to human vocabulary via
> rdfs:label.
> > The WW2 illustration is meant to illustrate how Google's quote
> operator
> > can be used as a poor man's SPARQL endpoint IFF the rdfs:labels are
> > actually carried into the HTML representation. I think it would be
> hard
> > to rationalize this point in any other cluster. I won't be offended,
> > though, if the use case was dropped for being too subtle or
> > insignificant.
> >
> > Jeff
> >
> >> -----Original Message-----
> >> From: Antoine Isaac [mailto:aisaac@few.vu.nl]
> >> Sent: Thursday, January 13, 2011 4:09 PM
> >> To: Young,Jeff (OR); Ray Denenberg; Gordon Dunsire; public-xg-lld
> >> Subject: Cases
> >>
> >> Dear all,
> >>
> >> As mentioned in today's call, we have been doubting whether two
> cases
> >> should be assigned to the Vocabulary alignment cluster [1]:
> >> - Component Vocabularies  [1]
> >> - Subject Search [2]
> >>
> >> These cases do not mention alignment openly. In fact they don't
> > feature
> >> multiple vocabularies that should be aligned or merged.
> >>
> >> So we would gladly welcome two kinds of assistance here:
> >>
> >> 1. case owners (Jeff and Ray) or curators (Gordon) to make a bit
> more
> >> explicit the "alignment" dimension of the cases, if they can, or to
> >> identify another cluster from [4] where they'd fit better, if they
> >> agree with us that the current clustering is not optimal;
> >>
> >> 2. other cluster's curators to "adopt" these two cases, if they
> think
> >> they would fit well their own cluster. Note that we already started
> > the
> >> job by identifying the general goal of the cases [5] and noting
down
> >> the associated vocabularies--and these cases are indeed very
> >> interesting ones!
> >>
> >> Thanks a lot for your help,
> >>
> >> Marcia, Michael, Antoine
> >>
> >> [1]  http://www.w3.org/2005/Incubator/lld/wiki/Cluster_VocAlign
> >> [2]
> >>
> >
>
http://www.w3.org/2005/Incubator/lld/wiki/Use_Case_Component_Vocabulari
> >> es
> >> [3]
> http://www.w3.org/2005/Incubator/lld/wiki/Use_Case_Subject_Search
> >> [4]  http://www.w3.org/2005/Incubator/lld/wiki/UseCases
> >> [5]
> >>
> >
>
http://www.w3.org/2005/Incubator/lld/wiki/Cluster_VocAlign#Scenarios_.2
> >> 8Case_Studies.29
> >
> >
> 
>

Received on Thursday, 13 January 2011 22:37:41 UTC