Google+

Harry Potter in FRBR

3

14 June 2006 by Ian Davis

Over on the FRBR blog, William Denton had produced a FRBR-ised version of that fine literary work Harry Potter and the Goblet of Fire. At last a FRBR example that I can identify with! It’s quite a detailed example and I thought I’d have a go at representing it in RDF using the FRBR schema Richard and I developed last year.

Some points to note in my interpretation:

  • I chose to use a mixture of successorOf and successor to describe the relationship of the book to others in the series. For consistency I perhaps could have described each work individually and just used the successor property. This would be closer to the pattern suggested by Dan Connolly. However, I wanted to show the RDF data in approximately the same order as William’s original example, so having the inverse properties helped with clarity.
  • I chose to make the audio versions expressions of the work although the placement of them in the original example implied that they were related to the written expression. This kind of thing is still the subject of debate in FRBR circles. Perhaps I could have related them to the written expression like the translations, but I’m not entirely sure which FRBR relationship to use, perhaps transformation?
  • In keeping with the usual FRBR practice I made the movie and game versions of the book separate works since they involve new creative input.

When reading this, it’s helpful to have the original example open in another window.

@prefix frbr: <http://vocab.org/frbr/core#> .
@prefix dc: <http://purl.org/dc/elements/1.1/> .
@prefix dct: <http://purl.org/dc/terms/> .
@prefix rdfs: <http://www.w3.org/2000/01/rdf-schema#> .
@prefix foaf: <http://xmlns.com/foaf/0.1/> .

_:hpgf
  a frbr:Work ;
  dc:title "Harry Potter and the Goblet of Fire"@en ;
  dct:abstract """Fourteen-year-old Harry Potter joins the Weasleys at the
    Quidditch World Cup, then enters his fourth year at Hogwarts Academy
    where he is mysteriously entered in an unusual contest that challenges
    his wizarding skills, friendships and character, amid signs that an old
    enemy is growing stronger."""@en ;

  frbr:creator _:jk ;

  frbr:successorOf [
    a frbr:Work ;
    dc:title "Harry Potter and the Prisoner of Azkaban"@en ;
    frbr:creator _:jk ;
    frbr:successorOf [
      a frbr:Work ;
      dc:title "Harry Potter and the Chamber of Secrets"@en ;
      frbr:creator _:jk ;
      frbr:successorOf [
        a frbr:Work ;
        dc:title "Harry Potter and the Philosopher's Stone"@en ;
        frbr:creator _:jk ;
      ]
    ]
  ] ;

  frbr:successor [
    a frbr:Work ;
    dc:title "Harry Potter and the Order of the Phoenix"@en ;
    frbr:creator _:jk ;
    frbr:successor [
      a frbr:Work ;
      dc:title "Harry Potter and the Half-Blood Prince"@en ;
      frbr:creator _:jk ;
      frbr:successor [
        a frbr:Work ;
        rdfs:label "Untitled seventh volume"@en ;
        frbr:creator _:jk ;

      ]
    ]
  ] ;

  frbr:adaption [
    a frbr:Work ;
    dc:title "Harry Potter and the Goblet of Fire"@en ;
    rdfs:label "The movie version" ;
  ] ;

  frbr:adaption [
    a frbr:Work ;
    dc:title "Harry Potter and the Goblet of Fire"@en ;
    rdfs:label "The game version" ;
  ] ;

  frbr:realization [
    a frbr:Expression ;
    rdfs:label "Original text" ;
    frbr:embodiment [
      a frbr:Manifestation ;
      rdfs:label "Bloomsbury (UK) hardcover" ;
    ] ;
    frbr:embodiment [
      a frbr:Manifestation ;
      rdfs:label "Bloomsbury (UK) paperback" ;
    ] ;
    frbr:embodiment [
      a frbr:Manifestation ;
      rdfs:label "Bloomsbury (UK) paperback (adult cover)" ;
    ] ;
    frbr:embodiment [
      a frbr:Manifestation ;
      rdfs:label "Scholastic (US) hardcover" ;
    ] ;
    frbr:embodiment [
      a frbr:Manifestation ;
      rdfs:label "Raincoast (Canada) hardcover" ;
    ] ;
    frbr:embodiment [
      a frbr:Manifestation ;
      rdfs:label "Raincoast (Canada) paperback" ;
    ] ;
    frbr:embodiment [
      a frbr:Manifestation ;
      rdfs:label "Penguin (Canada) paperback (adult cover)" ;
    ] ;

    frbr:translation [
      a frbr:Expression ;
      rdfs:label "Arabic translation" ;
      frbr:embodiment [
        a frbr:Manifestation ;
        rdfs:label "Hardcover" ;
      ] ;
      frbr:embodiment [
        a frbr:Manifestation ;
        rdfs:label "Softcover" ;
      ] ;
    ] ;

    frbr:translation [
      a frbr:Expression ;
      rdfs:label "Finnish translation" ;
      frbr:embodiment [
        a frbr:Manifestation ;
        rdfs:label "Hardcover" ;
      ] ;
      frbr:embodiment [
        a frbr:Manifestation ;
        rdfs:label "Softcover" ;
      ] ;
    ] ;

    frbr:translation [
      a frbr:Expression ;
      rdfs:label "Italian translation" ;
      frbr:embodiment [
        a frbr:Manifestation ;
        rdfs:label "Hardcover" ;
      ] ;
      frbr:embodiment [
        a frbr:Manifestation ;
        rdfs:label "Softcover" ;
      ] ;
    ] ;

    frbr:translation [
      a frbr:Expression ;
      rdfs:label "Spanish translation" ;
      frbr:embodiment [
        a frbr:Manifestation ;
        rdfs:label "Hardcover" ;
      ] ;
      frbr:embodiment [
        a frbr:Manifestation ;
        rdfs:label "Softcover" ;
      ] ;
    ] ;

    frbr:translation [
      a frbr:Expression ;
      rdfs:label "Wolof translation" ;
      frbr:embodiment [
        a frbr:Manifestation ;
        rdfs:label "Hardcover" ;
      ] ;
      frbr:embodiment [
        a frbr:Manifestation ;
        rdfs:label "Softcover" ;
      ]
    ]
  ] ;

  frbr:realization [
    a frbr:Expression ;
    rdfs:label "Jim Dale's unabridged reading" ;
    frbr:embodiment [
      a frbr:Manifestation ;
      rdfs:label "Listening Libary: 17 CDs" ;
    ] ;
    frbr:embodiment [
      a frbr:Manifestation ;
      rdfs:label "Listening Libary: 12 cassettes" ;
    ] ;
  ] ;

  frbr:realization [
    a frbr:Expression ;
    rdfs:label "Stephen Fry's unabridged reading" ;
    frbr:embodiment [
      a frbr:Manifestation ;
      rdfs:label "BBC Audiobooks: 18 CDs" ;
    ] ;
    frbr:embodiment [
      a frbr:Manifestation ;
      rdfs:label "BBC Audiobooks: 14 cassettes" ;
    ] ;
  ] ;

  frbr:subject [
    a frbr:Subject ;
    rdfs:label "Wizards -- Fiction" ;
  ] ;

  frbr:subject [
    a frbr:Subject ;
    rdfs:label "Magic -- Fiction" ;
  ] ;

  frbr:subject [
    a frbr:Subject ;
    rdfs:label "Schools -- Fiction" ;
  ] ;

  frbr:subject [
    a frbr:Subject ;
    rdfs:label "Harry Potter -- Fictional character" ;
  ] ;

  frbr:relatedEndeavour [
    a frbr:Work ;
    dc:title "Quidditch Through the Ages"@en ;
    frbr:creator [
      a foaf:Person ;
      foaf:name "Kennilworthy Whisp" ;
    ] ;
  ] ;

  frbr:relatedEndeavour [
    a frbr:Work ;
    dc:title "Fantastic Beasts and Where to Find Them"@en ;
    frbr:creator [
      a foaf:Person ;
      foaf:name "Newt Scamander" ;
    ] ;
  ] ;

  frbr:relatedEndeavour [
    a frbr:Work ;
    dc:title "The Sorcerer's Companion: A Guide to the Magical World of Harry Potter"@en ;
    frbr:creator [
      a foaf:Person ;
      foaf:name "Allan Zola Kronzek" ;
    ] ;
    frbr:creator [
      a foaf:Person ;
      foaf:name "Elizabeth Kronzek" ;
    ] ;
  ]
.

_:jk
  a foaf:Person ;
  foaf:name "J.K. Rowling" ;
Technorati Tags: ,

3 thoughts on “Harry Potter in FRBR

  1. Jonathan says:

    Hmm. As far as your questions about translations/audio versions, here is my understanding of the way the people who wrote the FRBR report thought about things, which I don’t think always quite matches the way you are doing it. (There is not neccesarily a ‘right’ answer, of course).Either a translation is the SAME work, or a translation results in a DIFFERENT work. This is a matter of judgement, although the FRBR report suggests that normally translations are the same work, unless they are especially creative/free translations. In any event, if the translation is the SAME work, the relationship is simply that the translation is a particular expression realizing that same work—there is no ‘translation’ relationship, I don’t think! You just know it’s a translation because it’s got a different language (and a translator), I guess. [Is this sufficient, does it make it easy enough for the system to pick out translations, an important task? Not sure.]But, you seem to have decided that the translated expression IS the same work, but there is a ‘translation’ relationship between the expression and it’s work, instead of a ‘realized’ relation. But I think the intent of the FRBR model (which again isnt’ neccesarily right) is that an expression ‘of’ a work _always_ has a ‘realized’ relation to that work–that’s the indicator that an expression is of the work! Or am I mis-reading the RDF, and there is still a ‘realized’ relation between the (translated) expression and it’s work, but ALSO a ‘translated’ relation between two expressions of the same work? That seems justifiable, although the extra ‘translated’ relation I dont’ think was contemplated by the original FRBR model.Alternately, if the translation is in fact decided to be a whole new work, then there is of course a ‘translated’ relation between the two works. Or, alternately, if you know the exact expression that served as the source of the translation, between that expression of the original work and the new translated work. But then you’ve got to give the translation it’s own new work entity, which I don’t think is what you are doing.Likewise, for the audio recording. Either the audio recording (“audio book”) is a _new_ expression, or it is just a manifestation of an existing expression. Which of these should be so is debatable. You seem to have decided it’s a new expression, a realization of the work just like any other expression. I think that’s probably the right decision, becuase it allows you to group together the CD vs. the cassette versions of the exact same audio book together in an expression. It would be arguable that an audio book is just a manifestation of an existing expression, if the complete and exact text of an existing expression is read—but that wouldn’t allow you to group the CD and cassette like that. In fact, this is probably as good a demonstration as any of why the ‘expression’ level is neccesary at all! Since some argue it isn’t.But note that here, you didn’t decide to make a relationship from the audio book expression to some prior (print) source expression—‘reading of’ or something. Perhaps simply because you don’t have a suitable relationship in your vocabularly, but note that this is not parallel with what you did with the translation, where you DID relate it to a prior source expression.Hmm. Hope those ramblings are helpful to someone. I think it’s great that you guys are making these FRBR examples. Not enough of that. It allows us to explicate what’s going on with a concrete example; I don’t know that there is neccesarily a ‘right’ answer, but I think it’s important to figure out where there are alternate possibilities.

  2. iand says:

    Thanks for the thoughtful comment Jonathan. I take your point on the translations. My RDF relates the written english expression to each translation expression using the translation property (in accordance with the relationship as shown in table 5.3 in the FRBR report). However I don’t state that those translations are explicit realizations of the same work.I could fix this in two ways, which are semantically equivilent. I could add a realizationOf property to each translation, e.g.: frbr:translation [ a frbr:Expression ; frbr:realizationOf _:hpgf ; rdfs:label “Arabic translation” ; frbr:embodiment [ a frbr:Manifestation ; rdfs:label "Hardcover" ; ] ; frbr:embodiment [ a frbr:Manifestation ; rdfs:label "Softcover" ; ] ; ] ;or I could label the translations in the rdf and cross-reference them:_:hpgf a frbr:Work ; dc:title “Harry Potter and the Goblet of Fire”@en ; frbr:realization _:arabic ; …_:arabic a frbr:Expression ; rdfs:label “Arabic translation” ; …Adding these relationships makes the data more useful.

  3. Jonathan says:

    Thanks for the pointer to table 5.3. That certainly convinces me that this the ‘translated’ relatoinship from expression-to-expression within the same work IS in fact intended by the original FRBR model. Which is good, because the more I thought about it, the more I concluded it was important to have that relationship!I do still definitely think it’s important to explicictly have a ‘realized’ relation between the expression and it’s work entity–all expressions should have a ‘realized’ to a work, right? You’ve provided a couple solutions for that, cool.An interesting dificult case for this translation stuff: A work (w1) of which multiple expressions (E1, E2, E3) exist in it’s original language, and which a translation (E4) exists—but it’s unclear _which_ expression served as the source of that translation. So it’s impossible to make an expression-to-expression ‘translated’ relation. E4 is definitely a realization of W1. It’s also definitely a translation—but there’s no way to make a translation relation, true? How would we want the system to display such a situation (as per William Denton’s mockup); and what data _would_ be required to support that display?That said, I realize getting caught up in these rare dificult cases is not always productive. I can’t even neccesarily think of an actual example for my hypothetical case, so maybe it’s not too likely, and not too important how it is handled.Also: A colleague who knows many of the people involved in FRBR suggests to me that the list of relationships, despite being so large, was not neccesarily meant to be exhaustive, or fully thought out. It was just meant to be a first stab. So when you aren’t sure what relationship to use for something, such as the possible relating of the audio book to the expression it was based on (where known)—it may very well be that there is no one right answer. But ‘transformation’ sounds as good to me as any. Is an audio book of a text in a similar position as a dramatization? To me, not quite, but close enough, and I coudln’t at all predict how ‘the user’ would think about it (surely, different users would think about it differently).Anyway, as you can see, I love to talk about this stuff, thanks to the both of you for your experimentation.

Leave a Reply

Please log in using one of these methods to post your comment:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s

Follow

Get every new post delivered to your Inbox.

Join 27 other followers

%d bloggers like this: