Search the web
Sign In
New User? Sign Up
rdfpath
? Already a member? Sign in to Yahoo!

Yahoo! Groups Tips

Did you know...
Hear how Yahoo! Groups has changed the lives of others. Take me there.

Best of Y! Groups

   Check them out and nominate your group.
Having problems with message search? Fill out this form to ensure your group is one of the first to be migrated to the new message search system.

Messages

  Messages Help
Advanced
Implementation Notes   Message List  
Reply | Forward Message #27 of 965 |
In trying to implement the RDFPath spec, I noticed that:

- The . selection for the current context is used but not defined.
- # is sort of a bad operator for grabbing type information. How about %?
- The spec distinguishes between resources, arcs and literals. Aren't they
all resources? I mean resources vs. literals I can sort of see, but arcs?
- It's not clear one goes in the whatever() functions.
- It's not clear where a plain URI vs. a QName can be used.
- The separation of filters and selections in sections 3 and 4 imply that
they are different things, however the examples show them being used
seemingly interchangeably. Which is it?
- It should also be noticed that the ' character (used to quote URIs) can
be included in URIs, which makes things messy. Can we use <...> instead?

Now I'm about half-way through my implementation and I realize the syntax
doesn't make much sense to me. Alright, time for some examples of my own
(I'm trying to be N3-like):

** Node Queries

/
Returns all nodes.
[]
Returns all anonymous nodes.
[foaf:mbox "me@..."]
Returns all nodes where the value of the property foaf:mbox equals
the literal "me@...".
[foaf:mbox "me@..." ; foaf:homepage <http://www.aaronsw.com>]
Returns nodes where foaf:mbox equals the literal of my email address
and foaf:homepage equals the resource of my homepage.
[a rdf:Bag]
Returns nodes of type rdf:Bag.
<http://example.org/#foo>
Returns the resource <http://example.org/#foo>.
"Whatever"
Returns the literal "Whatever".

** Sub-queries
(In these, the first path segment can be replaced with any of the node
queries above.)

/dc:creator
Returns values of the dc:creator property for all resources.
/[a rdfs:ContainerMembershipProperty]
Returns the values of all container membership properties (_1, _2, etc.)
for all resources.
"Aaron Swartz"\dc:creator
Returns all resources with a dc:creator property of "Aaron Swartz".

(The last one may be a bit odd, but I think it's very powerful. It allows
you to go backwards in queries.)

** Query Refinements

/dc:creator [a w3c:StaffMember]
Returns all of the w3c:StaffMembers who are listed as the dc:creator of
a resource.

I don't do arcs or reification because I don't understand how they fit into
a query context.

There's more I want to add to this, but I'm at a conference and don't have
time to finish it, but I wanted feedback.

What do you think?

--
[ Aaron Swartz | me@... | http://www.aaronsw.com ]




Tue May 22, 2001 8:49 pm

aswartz@...
Send Email Send Email

Forward
Message #27 of 965 |
Expand Messages Author Sort by Date

In trying to implement the RDFPath spec, I noticed that: - The . selection for the current context is used but not defined. - # is sort of a bad operator for...
Aaron Swartz
aswartz@...
Send Email
May 22, 2001
8:50 pm

... One might want to query a graph to find out what the relation was between two resources. Yes, the relation (property) is also a Resource, but it's being...
David Allsopp
dallsopp@...
Send Email
May 23, 2001
8:48 am

... I don't believe it is and don't think it should be treated any differently. ... Hmm, yes. That could be interesting. Perhaps we should start collecting ...
Aaron Swartz
aswartz@...
Send Email
May 24, 2001
10:01 pm

... Good idea. I will set up a document. regards, stefan...
Stefan Kokkelink
skokkeli@...
Send Email
May 27, 2001
10:29 am

... I would be glad if we had an RDFPath spec ;-) ... changed. ... why not, changed. ... I don't think so. Look at the example graphs in the spec. There is no...
Stefan Kokkelink
skokkeli@...
Send Email
May 23, 2001
12:39 pm

... I don't think this is relevant. Just because it looks one way in a graph serialization doesn't mean that the properties aren't resources. One could ...
Aaron Swartz
aswartz@...
Send Email
May 24, 2001
10:12 pm

... XPath is a path language that describes paths in a tree. RDFPath should (as the name suggests ;-) be a path language that describes paths in graphs. When...
Stefan Kokkelink
skokkeli@...
Send Email
May 27, 2001
10:51 am

... perhaps this is misleadling because it doesn't use the filter syntax. Should perhaps be changed to /dc:creator[%w3c:StaffMember] (costs one additional...
Stefan Kokkelink
skokkeli@...
Send Email
May 23, 2001
12:49 pm

... The RDF M&S says that Properties are a subset of Resources, and that the object of a triple is either a Resource or a Literal (i.e. it means, but doesn't...
David Allsopp
dallsopp@...
Send Email
May 23, 2001
4:19 pm

RDF M&S describes four sets, 'Resources', 'Literals', 'Properties' and 'Statements': 1) The triple itself is a member of the 'Statements' set. 2) The triple's...
Lee Jonas
ljonas@...
Send Email
May 24, 2001
2:01 pm

... yep. ... hmm, perhaps you are right. but do you agree that we schould talk about nodes *and* statements? (An RDF "model" is not only a set of statements,...
Stefan Kokkelink
skokkeli@...
Send Email
May 24, 2001
4:15 pm

... Interesting point -- I think this is correct. Thinking about things this way, I think that using a resource as a predicate implies its "existence" and thus...
Aaron Swartz
aswartz@...
Send Email
May 24, 2001
10:27 pm

Yes, of course the draft should refer to all relevant aspects of RDF - Resources & Literals instead of nodes; Statements & Properties instead of arcs & labels....
Lee Jonas
ljonas@...
Send Email
May 25, 2001
8:50 am

... that is exactly what I wanted to say, sorry for the confusion. but this raises a questions: if we use "/" as a query returning all the resources, do we...
Stefan Kokkelink
skokkeli@...
Send Email
May 24, 2001
3:58 pm

... Hmm, I think I see your point now. I think the answer is that / talks about subjects, \ talks about objects and the stuff in between are the predicates. ...
Aaron Swartz
aswartz@...
Send Email
May 24, 2001
10:17 pm

... Yes, but literals can be interpreted as resources using the data: scheme. -- [ Aaron Swartz | me@... | http://www.aaronsw.com ]...
Aaron Swartz
aswartz@...
Send Email
May 24, 2001
10:14 pm

... Interpreted by whom? By which I mean, will an RDF parser read them as Resources (with a URI) or Literals (without a URI)? Or do you mean that the...
David Allsopp
dallsopp@...
Send Email
May 25, 2001
8:35 am

... It can do either. Although, as Dave noted, there are some problems with this interpretation. -- [ Aaron Swartz | me@... | http://www.aaronsw.com ]...
Aaron Swartz
aswartz@...
Send Email
May 25, 2001
1:46 pm

... At one level, if we are able to distinguish between nodes and arcs, then we are necessarily using/treating their resources differently (if they aren't...
David Allsopp
dallsopp@...
Send Email
May 25, 2001
8:46 am
Advanced

Copyright © 2009 Yahoo! Inc. All rights reserved.
Privacy Policy - Terms of Service - Guidelines - Help