Re: [libdb-develop] W "Forms" and the use of matching "Roles"

[Morbus I. <mo...@di...>]

See my responses in the previous email, but the following is a
helpful bit of text from PAT...@bn... on the FRBR list:

 >The WG on FRBR/CRM Harmonization, while striving to "translate" FRBR into
 >OO formalism, has been thinking about the "form" attribute, assigned to the
 >Work entity in the FRBR Final Report. The reasoning was as follows: a Work
 >is "an abstract  entity, there is no single material object one can point
 >to as the work" (FRBR Final Report p. 16); a work can therefore definitely
 >*not* have a "form". In order to "grasp" the work, you must have access to
 >at least an Expression, which in turn can be grasped IFF (no pun
 >intended...) a Manifestation thereof is available. The Manifestation can be
 >perceived by the senses, and the information received by our senses allows
 >us to analyze and typify or categorize the Expression embodied in the
 >Manifestation; among the typification possibilities, there is a
 >typification by "form": "I recognize this as being text" (at a very general
 >level), "I recognize this as being a theatre script, a prose poem, a haiku,
 >a short novel (etc.)" (at a more specific level). Among all the possible
 >expressions for a given work, you have to choose one as being best
 >"representative" for the work, and the "form" of that "representative
 >expression" is then supposed to be the "constraining super-type of the
 >work", that is: "if an expression does not fit in this category, then it is
 >an expression of another work". For instance: in the FRBR conventions, an
 >adaptation of a novel into a children's book is supposed to another work
 >(p. 17; I've always found that questionable, but that's another issue).
 >This means that the novel has a "representative expression" which has as
 >constraining super-type: "text - novel"; then you compare the expression
 >embodied in the children's book (which has as form: "text - novel for
 >children") to that representative expression and you find out that they are
 >different. Of course, an authoritative list for "constraining super-types"
 >of works is dramatically important in that process, because this is what
 >will determine your cataloging conventions and demarcating lines between
 >works. If I do not declare "text - novel for children" as a distinct
 >"constraining super-type", but only "text - novel", then a novel adapted
 >for children will remain just an expression of the novel as a work, not a
 >new work. This flexibility was intended in the FRBR model, as no one in the
 >world can positively say "This is not the same work" (it's basically just a
 >matter of taste...).

-- 
Morbus Iff ( you, me, eropuri? aawwwwwWWWw yYeahahhHHAhhh )
Culture: http://www.disobey.com/ and http://www.gamegrene.com/
Spidering Hacks: http://amazon.com/exec/obidos/ASIN/0596005776/disobeycom
icq: 2927491 / aim: akaMorbus / yahoo: morbus_iff / jabber.org: morbus