I'm not sure the FRBR model is as specific and particular as you say. It
leaves most things about exactly what will be defined as "work" up to
the implementation, ie RDA.
I still say we need SOME formalized domain model before we start coming
up with new ways to do cataloging. It is absolutely a pre-requisite. I
don't see anyone disagreeing with this, I just see people, like Karen
below, saying that the FRBR model may not be sufficient to capture
everything we want. Okay, fine, so we can add to it. Or we can even
scrap it and start out with something new -- but I still think that if
you want any kind of backwards compatibility with our legacy data, you
aren't likely to do any better than FRBR as a foundation. But, sure, it
needs to be added to.
But how are we going to know how to add to it until we start using it
and seeing what we need?
As Tim says, we can't afford to wait anymore. We've waited over a
decade on FRBR already. We can do things in parallel -- evidence
gathering to see what users want (more evaluative data?); strategic
architecting to see how to properly model that information in our domain
model; at the same time we need to keep moving forward, not waiting
another decade for more 'evidence' that we may or may not actually gather.
My hypothesis is that moving forward by twiddling with rules for
creating text, without instead using a formally defined domain model, is
foolish. My secondary hypothesis is that nothing better than FRBR
presents itself for this formally defined model, and we can't afford to
wait another ten years to come up with something better before getting a
move on it.
"My gut feeling is that the main importance of WEMI is that it provides
that underlying structure, but even more important is that we have
strongly defined data elements whose use is clearly understood. "
Yes, I agree. So let's start moving on it. Note well though that
"strongly defined data elements" implies you know what _entity_ those
data elements apply to, which implies that you've formally defined some
entities.
Jonathan
Karen Coyle wrote:
> Jonathan Rochkind wrote:
>
>> (And really, I think it all makes a LOT more sense if you consider
>> those 'things', work, expression, and manifestation as sets of items.
>> The FRBR document doesn't make that clear, but neither is it
>> incompatible with that conception).
>>
>> I think FRBR is both more and less than people seem to be assuming. It
>> is simply a domain model for the bibliographic universe, which
>> establishes that there IS a thing called a work, and it matters.
>>
>>
>
> Unfortunately, the FRBR definition of work may not be the one that users
> desire. And that's the main problem with WEMI. If a user goes looking
> for War and Peace, odds are seeing all of the various translations isn't
> going to be useful. But that's what W gets them. What they really need
> is E. They MAY, however, be interested to know that it's been made into
> a movie. FRBR doesn't include that in the W. Users have a concept of
> series that isn't included in library catalog, that is things like the
> Harry Potter books, or mysteries with the same detective. WEMI doesn't
> cover that.
>
> I think WEMI may be a model for the creation of individual records, but
> there are extended relationships between resources that are the ones of
> interest to users. The WEMI structure is kind of background, the
> scaffolding that helps catalogers create records for items. But it's a
> cataloging view, not a user view. The question is: what should we be
> offering to users?
>
> Recent studies conclude that users like "evaluative" information. I
> haven't seen a good definition of that, but I assume it includes things
> like reviews, rankings, and "users who liked this also liked...". I
> suspect that something like "other books with these characters..." "...
> books that cite this book..." and a number of other pointers and links
> would fall into this category.
>
> We need a better idea of what we want to do with the records before we
> accept or condemn FRBR, because it's only valuable if it helps us do
> what we want to do for the user. If WEMI ends up being a good scaffold
> for everything from scholarly investigate to summer reading groups, then
> great, let's go for it. But if it gets in our way... we have to change
> it. I think we should be focusing on the part of FRBR that is likely to
> be of more use to information seekers, and that is the rich set of
> relationships between resources. Because those aren't structural, we
> should be able to expand them as needed to serve our users.
>
> My gut feeling is that the main importance of WEMI is that it provides
> that underlying structure, but even more important is that we have
> strongly defined data elements whose use is clearly understood. Exactly
> where each element fits into WEMI may not be of the greatest importance.
> We know that some communities have stricter definitions of W than
> others. For this reason WEMI may need to be variable, but the data
> elements need to be understandable no matter where the community places
> them in WEMI. I think that the relationships between bibliographic
> elements and segments will tell us more than the WEMI structure. "Is a
> translation of" is a E->W relationship. "Is an adaptation of" is a W->W
> relationship. (OK, this doesn't always hold, but where it doesn't, maybe
> it doesn't matter to the user.)
>
> Sorry, this rambled. I need another cup of tea.
>
> kc
>
>
Received on Wed Oct 21 2009 - 11:59:38 EDT