Miksa, Shawne wrote:
> I think this takes us back to ongoing argument of what are the parameters of library catalogs and the differences from the type of "information systems" you describe that go beyond the library. Are you speaking of a system "out there" somewhere or of an interconnected system of library systems? We're talking about redefining library collections if that is the direction you are going.
No. I don't think it redefines library collections. I do think we need to re-define library data, however. And that's hard, because the data and the catalog have always been the same thing. (We call it "cataloging" -- creating the catalog.) But I think we have to change that because the data that we put in our catalog isn't unique, and isn't just used by libraries and library users. There's a heck of a lot of bibliographic data on the Web -- and it is mostly unrelated to libraries. So the bibliographic world has opened up, and the place where most people encounter bib data today is NOT in the library catalog -- it's on Amazon or Wikipedia or Google Book Search.
What I think it comes down to is this: does it make sense to put the time and effort into creating library catalog records if the information in those records will not be used outside of the library catalog? And does it make sense to continue to create elaborate and expensive catalogs when we know that in fact users do the vast majority of their information seeking elsewhere? I'm not questioning whether the library needs an inventory of its holdings, because it obviously does. I AM questioning whether the effort that goes into "cataloging" as it is done today is a good use of our time and money, and suggesting that at least we could get more bang for our cataloging buck if we create data that can be re-used in other environments. We also need to be able to take in data more easily from other sources so that we don't have to create it anew. There just isn't enough time and money to justify that.
> Karen said:
>
>> You have no place for library data in a "stumble-upon" situation. No role for linking from other resources or between resources. No following links and paths. >And therefore your bibliographic data probably doesn't support those functions, even though those are the key way of finding information on the Web, which is >the primary information space today.
>>
>
> As for following what people do with the information once they have it---this is not something we do as it far as organizing information. Our objective has always been limited to connecting people to the information they want and need. To follow them and observe what they do with it---how is this in our purview? Invasion of privacy, more like. Perhaps you mean what they do with the information after obtaining and are still in the system. Now, if they want to share how they remix, annotate, compare, tag, within the catalog---great! Let's do it. Let's add citation analysis data as well. Jonathan Furner talked of this a few years back at a Classification Research workshop (Philadelphia, i think )
>
I didn't mean anything relating to following users. I was referring to
creating both data and systems that lend themselves to linking, as in
hyperlinking, but even thinking more about the richer linking that we'll
get with some of the semantic web concepts. Imagine that the user can
send a search query to the Web that asks: "What books did Herman
Melville write?" Not a keyword search on Melville and books, but a
specific query. And what if, where appropriate, the answers to that
query could come from the vast store of bibliographic data in library
catalogs. Not a reply made up of a bunch of bib records, nor links to
library catalogs (although those could be offered), but a simple answer
to a simple query:
Billy Budd
Confidence-Man
Isle of the Cross
... etc. (and which I had to go to Wikipedia to get, because the
library catalog data couldn't answer this question for me)
We have that data, but we don't have a way to answer users' questions.
Without dismantling our catalogs we could offer up this data to the Web.
To do so, however, we've got to get it out of MARC records and we've got
to make use of certain notions of what data looks like on the Web. We
also have to allow the data to be harvested from our catalogs for use in
other applications.
Alternatively, if our data is Web-friendly and linkable, then we can
allow users to move more easily from the catalog to Web sites and back
again. Libraries could become "of the Web" -- interlinked with other
information resources.
It's the insularity and boundedness of the library catalog that doesn't
work for me as a modern information resource.
> All this reminds of Ranganathan's view of information ---multi-dimensional universe, information piece by piece, subject by subject, infinite, but bogged down by the tiresome variability in language.
>
Which is one of the things that the Semantic Web attempts to address:
variability in language. It's not unlike what is being done with the
VIAF -- bringing together variants of names so that you can do switching
between them in some kind of automated way.
> Karen said:
>
>> You have users looking for documents, not information. That already is a real limitation.
>>
>
> Not a limitation, just a different approach to finding and accessing. (and here I borrow a bit from my long ago dissertation research) It is a question of the use of structure when organizing and how we think of structured information. Much of the early work in information retrieval narrowed the notion of document retrieval from the intersection of a person with a structure of knowledge (a classification system, for example) to an intersection of a person's single question with a collection of documents and in this view seemed to conclude that a structure of knowledge was not necessary in order to retrieve documents that would answer a question. In essence, while some of us may ask "where in the structure (such as in a library classified by DDC or LCC) is the answer?" others, such as yourself, may counter with "No, where in the document is the answer?" where the document/resource/collection of information is out there in your "primary information space".
>
I don't think we're on the same page. I think there's a big gap between
"How big is a right whale?" and a search for a book about whales. Do
libraries have information, or just documents with information? And what
is it that users want? Sometimes they ARE looking for a book, but we
have reference desks because sometimes they are looking for specific
information, not a document. I suspect that reference desks are a bit
lonely these days, because there are sources of information that work
better than the library. I think that libraries tried to get into the
info biz in the 20th century but never really succeeded. We could argue
that's not the library's role, that the library is a store of documents
and that, information-wise, the library is an organizer of that store.
That would be rather sad, in my opinion. A lively public library has all
kinds of non-document information and is a pro-active creator of
information resources (local history, social resources) plus a social
site for literacy classes, reading clubs, and all kinds of other things.
Interestingly, these information resources do NOT interact well with the
catalog, and have to be communicated separately to the community. So
it's not just how we define the catalog, it's how we define the library.
And if we define the library as a lively information space, what should
the catalog look like in that space? Should it be an inventory of
library holdings? Or could it be more?
> Forgive me, this is a listserv about the future of catalogs for libraries, yes? Or am I in the wrong place? ;-)
>
>
It's the right place, but it may turn out that the future of the catalog
looks nothing like the catalog today, if we are talking about the
catalog that was created with the users in mind (as opposed to the data
that is necessary for library management functions like inventory
control, lending, acquisitions, etc.).
kc
--
-----------------------------------
Karen Coyle / Digital Library Consultant
kcoyle@kcoyle.net http://www.kcoyle.net
ph.: 510-540-7596 skype: kcoylenet
fx.: 510-848-3913
mo.: 510-435-8234
------------------------------------
Received on Mon Oct 19 2009 - 22:01:11 EDT