Karen,
I take your points. I feel that catalogers [and even some developers :) ] do crave a certain amount of structure, and can be daunted by the a pile of Legos (not implying anything about my age) on the ground. If we have profiles of things that suggest what properties to use, we're more likely to get it right.
I'm not in favor of spinning out new subclasses every time a new mime-type appears on a web page, but we should be able to come up with some basic ones, and then we'll probably need to expand as technology adds to ability to express ourselves.
Nate
-----Original Message-----
From: Bibliographic Framework Transition Initiative Forum [mailto:[log in to unmask]] On Behalf Of Karen Coyle
Sent: Thursday, February 07, 2013 4:31 PM
To: [log in to unmask]
Subject: Re: [BIBFRAME] Newspeak
Nate,
I'm not sure what the "semantics of 'type'" means in this context, since you do say "defining a new 'type' of work." I think the question
becomes: will all combinations of additional properties require defining as a new type (or whatever it is called)?
As we learned with MARC, the number of combinations of attributes is, albeit not infinite, at least daunting. Which is why we did format integration which then allowed ANY MARC field to be used in any instance record. Previously there had been "Book" fields and "Map" fields and "Sound recording" fields that could only be used when that was the format coded in the Leader. This sometimes left catalogers unable to say what they wanted to say about something.
I'm convinced that we cannot "model our universe of data" as a metadata model that covers everything anyone would ever want to catalog, or how they want to catalog it. This is why I am highly skeptical of FRBR -- because it tries to fix one view of bibliographic data, as if the world isn't undergoing constant change. While there may well be a convenient core of elements, beyond that the main qualification, IMO, is re-usability -- give catalogers a whole host of elements that they can use wherever they want, even if no one has used that combination before.
The instance data then becomes the picture of the bibliographic universe, not a pre-defined structure. In other words, create the tinker toys (or Legos for those not old enough to remember tinker toys) and let the catalogers make things with them.
kc
On 2/7/13 10:20 AM, Trail, Nate wrote:
> I concur with "... a work that has some attributes that are different from other works...". This approach allows us to have a core "Work" set of properties, and when there's something that has properties that are not in common with the core, defining a new "type" of work that has all of the core plus the additional properties, for it. The semantics of "type" don't need to enter in to it.
>
> Nate
>
> ---------------------------------------------------------------
> Nate Trail
> ---------------------------------------------------------------
> Network Development and MARC Standards Office Technology Policy Mail
> stop 4402 Library Services Library of Congress
> 202-707-2193
> [log in to unmask]
>
>
>
>
> -----Original Message-----
> From: Bibliographic Framework Transition Initiative Forum
> [mailto:[log in to unmask]] On Behalf Of Ratliff, Louise
> Sent: Thursday, February 07, 2013 11:47 AM
> To: [log in to unmask]
> Subject: Re: [BIBFRAME] Newspeak
>
> Hi,
>
> I applaud Stephen for his analytical approach to BIBFRAME and the FRBR model -- thanks for helping us to sort this out!
>
> Overall, I have a general caveat to present with regard to defining "types of work," which is to beware of defining things in terms of their carriers rather than their content. That being said, I don't see a dissertation being a type of work, but rather a work that has some attributes that are different from other works. Dissertations have an association with a degree and an institution. Other than that, they don't differ from other works, be they textual, visual, or whatever.
>
> Cartographic works might be a valid "type," since they present information within a geospatial context.
>
> On balance, though, I wonder what value there is in specifying "type of work?" Aren't there just "works?" At the FRBR Expression level, some works may appear in different representations from the original "work" but still retain some identity of the original "work."
> As a user, I am not looking to find "types of works," but rather a set of resources with certain characteristics (author, format, form, etc.) I find "types of works" to be an artificial concept.
>
> Just my 2 cents based on trying to get my head around this!
>
> Louise Ratliff
>
> Louise Ratliff
> Social Sciences and Map Catalog Librarian UCLA Cataloging & Metadata
> Center Los Angeles, CA
>
> P.S. The Open Metadata Registry already has defined all of the MARC
> vocabularies. I don't see why you cannot use that. (I'm being naive
> here to make my point.)
>
> -----Original Message-----
> From: Bibliographic Framework Transition Initiative Forum
> [mailto:[log in to unmask]] On Behalf Of Ford, Kevin
> Sent: Wednesday, February 06, 2013 3:49 PM
> To: [log in to unmask]
> Subject: Re: [BIBFRAME] Newspeak
>
> Let me begin with the really easy ones. :)
>
>> Are these the sort of things that an RDA profile for BibFrame could
>> specify?
> -- Yes.
>
>> Are these useful questions to be raising?
> -- Yes. Yes.
>
> This may be a lackluster response but...: There are so many different avenues one can pursue on the resource type issue that suffice it to say it is easily one of the reasons we wanted to take a very deliberate approach. The various opinions in the few posts on this topic already underscore how variable the possibilities are.
>
>> Could the Types of Work be defined on a different level--less format,
>> more FRBR?
> -- "Could" is the short answer, but we're thinking that a profile approach or some kind of inference-based approach will be better for identifying FRBR Works and FRBR Expressions from BIBFRAME Resources.
>
>> must all
>> BibFrame Works include the property Language, regardless of type?
> -- No. Nothing is mandatory presently. I say "presently" only because I'm hedging not because I have special knowledge. Something *could* become mandatory in the future but that's a detail that hasn't come up. Well, that's not true - I remember saying at some point, probably to myself, "something - something! - has to be mandatory."
>
>> Could FRBR Work and FRBR Expression be types of the BibFrame Work,
>> with the latter taking over the Language property,
> -- I think that language (along with the arranged statement for music) is a reasonable way to infer a FRBR Expression from a BIBFRAME Work.
>
>> Could
>> Integral Work and Aggregated Work (another distinction from FRBR) be
>> defined in BibFrame as Types of Work?
> -- Dunno. I can see how it could solve some issues surrounding the distinction between the two types of works. What I can say is that, so far, the resource type discussion has revolved around actual types of resources (Book, Newspaper, Painting) versus slightly abstracted concepts (language material, aggregated work, etc). It may turn out that the two approaches are married, with the super-types being more abstract than their sub-types. But that is speculation on my part.
>
>> Can a BibFrame Work description
>> specify more than one Type? I think we learned with MARC format
>> integration that exclusive format definitions can be problematic.
> -- Technically, at this time, there is no restriction on this. However, community practice might want to enforce such a rule that there is to be one and only one resource type per resource. I can recall precisely one conversation that touched on this and opinion was immediately divided.
>
> Personally, in past projects, I have found the restriction of one resource per resource type limiting and prone to cumbersome solutions. That said, I can also see how the data world would be a much cleaner, more hospitable place with such a rule.
>
> You're right to note that exclusive format definitions can be problematic, but the mixing of the two can be equally problematic. I'm comfortable with seeing if the answer presents itself as the resource type discussion continues.
>
> As to your first paragraph about your interpretation of the vocab website all I can say is "fantastic." You've interpreted it correctly and it's reassuring to know that the presentation is comprehensible in just the way it was intended. Credit for that goes to Zepheira, which provided the design for the vocab pages.
>
> Yours,
> Kevin
>
>
>
>> -----Original Message-----
>> From: Bibliographic Framework Transition Initiative Forum
>> [mailto:[log in to unmask]] On Behalf Of Stephen Hearn
>> Sent: Wednesday, February 06, 2013 2:41 PM
>> To: [log in to unmask]
>> Subject: Re: [BIBFRAME] Newspeak
>>
>> I'm trying to understand how much flexibility there is or will be in
>> the BibFrame definitions of Work and Types of Work. Drilling down
>> into the explanations of Work and Types of Work I can see that in
>> BibFrame Works are Resources, and therefore can have the properties
>> of Resources as well as properties specific to Works, and that Types
>> of Work
>> (examples: Dissertation, Cartographic) include more specific
>> properties as well, in addition to the properties of Resource and Work.
>>
>> Could the Types of Work be defined on a different level--less format,
>> more FRBR? Could FRBR Work and FRBR Expression be types of the
>> BibFrame Work, with the latter taking over the Language property, or
>> must all BibFrame Works include the property Language, regardless of
>> type? Could Integral Work and Aggregated Work (another distinction
>> from FRBR) be defined in BibFrame as Types of Work? Can a BibFrame
>> Work description specify more than one Type? I think we learned with
>> MARC format integration that exclusive format definitions can be problematic.
>>
>> Are these the sort of things that an RDA profile for BibFrame could
>> specify? Are these useful questions to be raising?
>>
>> Stephen
>>
>> --
>> Stephen Hearn, Metadata Strategist
>> Technical Services, University Libraries University of Minnesota
>> 160 Wilson Library
>> 309 19th Avenue South
>> Minneapolis, MN 55455
>> Ph: 612-625-2328
>> Fx: 612-625-3428
--
Karen Coyle
[log in to unmask] http://kcoyle.net
ph: 1-510-540-7596
m: 1-510-435-8234
skype: kcoylenet
|