On 9/17/13 1:14 AM, Meehan, Thomas wrote:
Karen,
Yes, I mean that if you have something like:
- example:book1 dc:title example:unknownTitle .
- example:book2 dc:title example:noTitle .
Thomas, this does not work because the object of dc:title is the title
of the work (actually, the defintion is "A name given to the resource".)
Therefore, you have just declared the title of the work to be
"example:unknownTitle." That obviously isn't right, but most importantly
it isn't helpful. It isn't just a question of dcterms requiring a
literal, it has to do with the meaning of dc:title. You need a property
with a different meaning.
If you do this, then most dc-aware programs would display:
Title: example:unknownTitle
It would actually probably be more useful to say:
dc:title "unknown title"
although again you would be declaring that string to be the actual title
of the item, which then gets us into the "Title varies" issue -- that
is, you cannot prevent someone from creating a resource and giving it
the actual title "unknown title."
Now, if you were to have the object of dc:title defined as "Content as
Text", then, as the conversation continued between Esme and I, you could
have a blank text (which would be saying that dc:title = ""), and, since
the object of dc:title would be a URI (not a string), you could
potentially mint a new property with the semantics of "title status"
that could take an object like "example:unknownTitle."
kc
"
then this approach only works for the dc:title element (putting aside for now
what you said about dc:title requiring a literal). For any other elements you
would need to define a similar example:unknownThing:
- example:book2 example:claspNote "brass" .
- example:book2 example:claspNote example:noClasp .
The same would go for the following approach:
- example:book2 example:hasATitle egboolean:false .
- example:book2 example:hasAClasp egboolean:false .
I hope these make sense. An approach that could be used for any element might
be useful. I admit at this stage that I need to reread the second half of the
thread again. I can't decide in my head if this is something that the rules
(RDA, etc) should handle. If so, RDA in particular has an entertaining variety
of ways of doing so.
Thanks,
Tom
---
Thomas Meehan
Head of Current Cataloguing
Library Services
University College London
Gower Street
London WC1E 6BT
t.mee...@ucl.ac.uk
-----Original Message-----
From: Code for Libraries [mailto:CODE4LIB@LISTSERV.ND.EDU] On Behalf Of
Karen Coyle
Sent: 16 September 2013 16:23
To: CODE4LIB@LISTSERV.ND.EDU
Subject: Re: [CODE4LIB] Expressing negatives and similar in RDF
On 9/16/13 2:05 AM, Meehan, Thomas wrote:
Don: As I understand it, the open world view implies knowledge not
asserted for whatever reason, whereas sometimes a negative is a definite
(and ultimately verifiable) fact, such as a painting simply not having a title.
I
think you're ultimately right about unknown things.
Esmé's solution does seem to work, although would perhaps require
redefinition for every element (title, place of pub, presence of clasp,
binding, etc.). I did wonder if a more generic method existed.
Can you say more about what you mean by "redefinition for every element"?
kc
Thank you,
Tom
---
Thomas Meehan
Head of Current Cataloguing
Library Services
University College London
Gower Street
London WC1E 6BT
t.mee...@ucl.ac.uk
-----Original Message-----
From: Code for Libraries [mailto:CODE4LIB@LISTSERV.ND.EDU] On Behalf
Of Donald Brower
Sent: 13 September 2013 14:46
To: CODE4LIB@LISTSERV.ND.EDU
Subject: Re: [CODE4LIB] Expressing negatives and similar in RDF
At a theoretical level, doesn't the Open World Assumption in RDF rule
out outright negations? That is, someone else may know the title, and
could assert it in a separate RDF document. RDF semantics seem to
conflate unknown with nonexistent.
Practically, Esme's approach seems better in these cases.
-Don
--
Donald Brower, Ph.D.
Digital Library Infrastructure Lead
Hesburgh Libraries, University of Notre Dame
On 9/13/13 8:51 AM, "Esmé Cowles" <escow...@ucsd.edu> wrote:
Thomas-
This isn't something I've run across yet. But one thing you could
do is create some URIs for different kinds of unknown/nonexistent titles:
example:book1 dc:title example:unknownTitle
example:book2 dc:title example:noTitle etc.
You could then describe example:unknownTitle with a label or comment
to fully describe the states you wanted to capture with the
different categories.
-Esme
--
Esme Cowles <escow...@ucsd.edu>
"Necessity is the plea for every infringement of human freedom. It
is the argument of tyrants; it is the creed of slaves." -- William
Pitt,
1783
On 09/13/2013, at 7:32 AM, "Meehan, Thomas" <t.mee...@ucl.ac.uk>
wrote:
Hello,
I'm not sure how sensible a question this is (it's certainly
theoretical), but it cropped up in relation to a rare books
cataloguing discussion. Is there a standard or accepted way to
express negatives in RDF? This is best explained by examples,
expressed in mock-
turtle:
If I want to say this book has the title "Cats in RDA" I would do
something like:
example:thisbook dc:title "Cats in RDA" .
Normally, if a predicate like dc:title is not relevant to
example:thisbook I believe I am right in thinking that it would
simply be missing, i.e. it is not part of a record where a set
number of fields need to be filled in, so no need to even make the
statement.
However, there are occasions where a positively negative statement
might be useful. I understand OWL has a way of managing the
statement This book does not have the title "Cats in RDA" [1]:
[] rdf:type owl:NegativePropertyAssertion ;
owl:sourceIndividual example:thisbook ;
owl:assertionProperty dc:title ;
owl:targetIndividual "Cats in RDA" .
However, it would be more useful, and quite common at least in a
bibliographic context, to say "This book does not have a title".
Ideally
(?!) there would be an ontology of concepts like "none", "unknown",
or even "something, but unspecified":
This book has no title:
example:thisbook dc:title hasobject:false .
It is unknown if this book has a title (sounds undesirable but I
can think of instances where it might be handy[2]):
example:thisbook dc:title hasobject:unknown .
This book has a title but it has not been specified:
example:thisbook dc:title hasobject:true .
In terms of cataloguing, the answer is perhaps to refer to the
rules (which would normally mandate supplied titles in square
brackets and so
forth) rather than use RDF to express this kind of thing, although
the rules differ depending on the part of description and, in the
case of the kind of thing that prompted the question- the presence
of clasps on rare books- there are no rules. I wonder if anyone has
any more wisdom on this.
Many thanks,
Tom
[1] Adapted from
http://www.w3.org/2007/OWL/wiki/Primer#Object_Properties
[2] No many tbh, but e.g. title in an unknown script or
indecipherable hand.
---
Thomas Meehan
Head of Current Cataloguing
Library Services
University College London
Gower Street
London WC1E 6BT
t.mee...@ucl.ac.uk
--
Karen Coyle
kco...@kcoyle.net http://kcoyle.net
m: 1-510-435-8234
skype: kcoylenet
--
Karen Coyle
kco...@kcoyle.net http://kcoyle.net
m: 1-510-435-8234
skype: kcoylenet
