Below is a section of RDF from a page-object. We keep this right in RELS-EXT 
and let Fedora do the work of updating for us.

---
A. Soroka
Online Library Environment
the University of Virginia Library

<rdf:RDF ... xmlns:uva="http://fedora.lib.virginia.edu/relationships#";>
        <rdf:Description rdf:about="info:fedora/uva-lib:1038869">
...
                <uva:hasPreceedingPage 
rdf:resource="info:fedora/uva-lib:1038868"/>
                <uva:isFollowingPageOf 
rdf:resource="info:fedora/uva-lib:1038868"/>
                <uva:hasFollowingPage 
rdf:resource="info:fedora/uva-lib:1038870"/>
                <uva:isPreceedingPageOf 
rdf:resource="info:fedora/uva-lib:1038870"/>
...
        </rdf:Description>
</rdf:RDF>

On Aug 31, 2011, at 3:25 PM, Scott Prater wrote:

> Hi, Adam --
> 
> Can you give us a sample of the double linked-list triples?  I'd like to see 
> how you express that in RDF.
> 
> Is this something you express in the RELS-EXT, or triples you add to the RI 
> outside Fedora?
> 
> thanks,
> 
> -- Scott
> 
> On 08/31/2011 02:18 PM, aj...@virginia.edu wrote:
>> I agree with Scott's point that "page sequences can be considered a property 
>> of containers ... rather than a property inherent in the page object 
>> itself", and because of a well-known updating difficulty in the RI, it is 
>> not possible to use RDF containers in RELS-*. See:
>> 
>> https://jira.duraspace.org/browse/FCREPO-656
>> 
>> There is, however, a fair way to use the RI to this end-- you can construct 
>> a linked list, or a doubly-linked list amongst the pages. That's what we do 
>> (a doubly-linked list). We're therefore relying on the RI to be fast, which 
>> it is (it is, after all, an _index_-- the repository is the data store). 
>> Updating is fairly easy (create object with two relationships, then alter 
>> four relationships of the preceding and following pages). We have no trouble 
>> querying into the RI for results fast enough for a page-turner presentation.
>> 
>> But then, we are strongly committed to RDF for as much structural metadata 
>> as we can cram into it. {grin}
>> 
>> ---
>> A. Soroka
>> Online Library Environment
>> the University of Virginia Library
>> 
>> 
>> 
>> 
>> On Aug 31, 2011, at 2:11 PM, Scott Prater wrote:
>> 
>>> Hello, Joszef --
>>> 
>>> I'll send you a couple of sample objects in a separate email.
>>> 
>>>> Not to mention, that I don't really have a clear vision, how to store page 
>>>> orders in RELS... :)
>>> 
>>> Nor do we.  You could store the sequence triple in the object itself,
>>> something along the lines of<myobject>  <hasPageSequence>  "1", but then
>>> you'd have to query every single object to build a list of pages (not a
>>> real big deal in the resource index, but still, a little clunky).  And
>>> what if you forget to scan a page, and your numbering gets all whacked,
>>> and you have to go back and add a page later (something that occurs more
>>> often than we would like to admit)?  You'll need to update the page
>>> sequence triple in every following page object.
>>> 
>>> An even more subtle problem crops up if your object has one page number
>>> in one context (say, a plate in a book) and another page number in
>>> another context (say, the same plate in an art exhibit catalogue):   if
>>> you were to create these two relations in the page object, how would you
>>> express in a triple that I'm page 1 of book A, and page 3 of book B?
>>> This is a use case which demonstrates that page sequences can be
>>> considered a property of containers ("I'm a book, and I have this
>>> content at position X"), rather than a property inherent in the page
>>> object itself.
>>> 
>>> That would be okay, except that there's no way to express in a book
>>> object's RELS-EXT triple that book object A contains page object A1 with
>>> the attribute page sequence "1".  You can do that in METS, though.
>>> 
>>> -- Scott
>>> 
>>> 
>>> --
>>> Scott Prater
>>> Library, Instructional, and Research Applications (LIRA)
>>> Division of Information Technology (DoIT)
>>> University of Wisconsin - Madison
>>> pra...@wisc.edu
>>> 
>>> ------------------------------------------------------------------------------
>>> Special Offer -- Download ArcSight Logger for FREE!
>>> Finally, a world-class log management solution at an even better
>>> price-free! And you'll get a free "Love Thy Logs" t-shirt when you
>>> download Logger. Secure your free ArcSight Logger TODAY!
>>> http://p.sf.net/sfu/arcsisghtdev2dev
>>> _______________________________________________
>>> Fedora-commons-users mailing list
>>> Fedora-commons-users@lists.sourceforge.net
>>> https://lists.sourceforge.net/lists/listinfo/fedora-commons-users
>> 
>> 
>> ------------------------------------------------------------------------------
>> Special Offer -- Download ArcSight Logger for FREE!
>> Finally, a world-class log management solution at an even better
>> price-free! And you'll get a free "Love Thy Logs" t-shirt when you
>> download Logger. Secure your free ArcSight Logger TODAY!
>> http://p.sf.net/sfu/arcsisghtdev2dev
>> _______________________________________________
>> Fedora-commons-users mailing list
>> Fedora-commons-users@lists.sourceforge.net
>> https://lists.sourceforge.net/lists/listinfo/fedora-commons-users
> 
> 
> -- 
> Scott Prater
> Library, Instructional, and Research Applications (LIRA)
> Division of Information Technology (DoIT)
> University of Wisconsin - Madison
> pra...@wisc.edu


------------------------------------------------------------------------------
Special Offer -- Download ArcSight Logger for FREE!
Finally, a world-class log management solution at an even better 
price-free! And you'll get a free "Love Thy Logs" t-shirt when you
download Logger. Secure your free ArcSight Logger TODAY!
http://p.sf.net/sfu/arcsisghtdev2dev
_______________________________________________
Fedora-commons-users mailing list
Fedora-commons-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/fedora-commons-users

Reply via email to