With some time away from the keyboard, I had a chance to put my
thoughts together a little better.
There are basically two possibilities here:
1) People can find things
2) People can't find things
Let's start with #1. If they are able to find things, how are they
finding them? My guess is that a vast majority of users at OCLC
member institutions did not find whatever they were looking for
because of a catalog card, and yet it's almost certain that the bib
record for the item was cataloged using rules intended to aid the
display and printing of a catalog card. Even if the record was
created in 2009. I am not saying all of our painstaking rules go
unused, I just think we need to figure out what *is* used and what
isn't.
On to #2. There are, obviously, a lot of reasons we could get to #2,
the most obvious being, "the library does not have what you're looking
for".n Let's set that option aside for now, since it's a little out
of scope. There is also the possibility that the library actually
*does* hold the item that the user wants, but they, or the agent that
is performing their search, does not know the proper way to ask their
question to get the correct answer. How could our rules be modified
or improved to reduce the chance of false negatives? How much does
user behavior go into how we model our data or classify our materials?
-Ross.
On Wed, Feb 11, 2009 at 2:49 PM, Ross Singer <rossfsinger_at_gmail.com> wrote:
> Er, ok, obviously referring the bookstore was a mistake because people
> are focusing on the wrong point. I am not saying we adopt a bookstore
> model. I am not saying that bookstores don't shelve things in a model
> similar to libraries.
>
> I'm saying this building is chock full of books (by the way, bigger
> than a big box store/smaller than The Strand) with no catalog and
> people are able to find things.
>
> I would also wager in our much bigger collections, where a catalog
> /is/ necessary because 'browsing' would be too inefficient, people are
> still using extremely crude searching methods and finding things.
>
> I asking if (and betting probably yes) a vast majority of them are not
> using any aspect of the MARC record that couldn't easily be rendered
> in an unqualified Dublin Core and if this is true, is the effort of
> descriptive cataloging being expended most appropriately.
>
> -Ross.
>
> On Wed, Feb 11, 2009 at 2:32 PM, Kyle Banerjee <kyle.banerjee_at_gmail.com> wrote:
>>> After all, in my home town we have a very large used book store. It
>>> has no publicly accessible inventory of their 'collection', and yet it
>>> is always crowded with people that seem to have found what they want.
>>>
>>> Somehow, people have a knack for finding things even in the crudest of
>>> classification.
>>
>> Just out of curiosity, how big is "very large?" I often hear of using
>> bookstores as a model to emulate. However, the number of books in a
>> very large bookstore is minuscule compared to a decent library
>> collection. It's one thing to arrange tens of thousands of things,
>> mostly on popular topics. It's another to arrange millions of things
>> that come in different formats, publication patterns, etc.
>>
>> People may use libraries differently than they do bookstores. For
>> example, in an academic library, a huge percentage of the patrons are
>> trying to find numerous very specific materials (since they're working
>> from a bibliography). In a used bookstore, they're probably searching
>> for only a few specific things and might just be browsing to see
>> what's interesting. In this case, you can be well served by a much
>> looser system.
>>
>> I'm no cataloging fanboy since I think many of the most time consuming
>> procedures amount to fidgeting with no benefit, but the bookstore
>> experience is overrated. Besides, many bookstores use cataloging data
>> (publisher created, CIP, etc) whether they state it or not...
>>
>> kyle
>>
>
Received on Wed Feb 11 2009 - 16:31:29 EST