Jon,
I think I misunderstood your initial proposal to some extent and then
responded in a way that could be misunderstood, too. Is this what you
were proposing: that there be author browse screens that allow you to
click for more information on individual authors, and when you do,
there's a sort of "drop-down" display of titles by that person or
subjects she's written on, without leaving the index screen? I can see
how something like that would be an enhancement of what we currently
have. And in agreement with some of your comments and things I've seen
by others, "undifferentiated" headings would be excluded from that.
There might be a message when you click that this is an undifferentiated
name, so a list of titles or subjects would not be helpful. Is that
roughly what you were proposing?
I wasn't sure what you meant below: "In the current systems, using
connections made in authority files force a user to jump through many
hoops." While I agree that your proposal seems helpful, I don't see why
our current systems are that difficult for users. Maybe it's just
because I'm a cataloger and am used to "jumping through the hoops" all
the time myself. Ordinarily, if you search on a "see reference," you
only have to click, and it takes you to the established form. And if
you're browsing an author index screen, it doesn't seem that tough to
click on an individual name, go to the screen listing his works, find no
match to what you were looking for, and go back to the author index
(except in Horizon, where it annoyingly takes you back to the top of the
list instead of the place where you clicked, so you have to remember
where you were).
But at any rate, I think I see your proposal as a helpful idea, if I
understand it correctly now.
The way undifferentiated authorities like "Smith, John" were explained
to me is as places to store information about authors we don't know much
about. They identify various people who are distinct and give examples
of their work. Of course that means they're strictly for catalogers.
They don't do anything for users. But that's okay, because cataloging is
complex, and catalogers need help figuring authors out.
--Ted Gemberling
-----Original Message-----
From: Next generation catalogs for libraries
[mailto:NGC4LIB_at_LISTSERV.ND.EDU] On Behalf Of Jon Gorman
Sent: Tuesday, May 15, 2007 7:12 PM
To: NGC4LIB_at_LISTSERV.ND.EDU
Subject: Re: [NGC4LIB] Next Gen Catalog and FRBR
On 5/15/07, Ted P Gemberling <tgemberl_at_uab.edu> wrote:
> Jonathan,
> I realize that the undifferentiated practice is not exactly what Jon
was
> proposing. I was only saying there's a certain similarity between the
> two things, based on a stress on works and subjects associated with
> names.
>
I guess I'm just completely missing your point. I suggested a way to
improve the current author browse. I'm not suggesting any great
silver bullet or modification to current authorities practice. So
what about the problems you listed (mistakes, not knowing the proper
author is, etc) could be solved by any interface? In the current
systems, using connections made in authority files force a user to
jump through many hoops. So....if there's essential flaws in the
underlying authority records that are so bad any attempt to "hook" the
rest of our cataloging data to them, I see that as more an argument
not to do any authority work at all. What I'm proposing is to at
least help give more information from what work we already have. If
an author is in a author browse/listing and we have an authority
record for them, it seems to make sense to connect this to
bibliographic information that uses this authority form.
I consider authority work an excellent part of cataloging and
libraries. Given that, I have to go with "we'll assume what we have
is good, and authority records are correct as we can know.". In fact,
this system might be a good tool for others to quickly scan and see
possible mistakes as well. Similar techniques in combination with
data-mining could also yield some interesting results. As more
resources of knowledge are electronic, the more we can create tools
that might help researches identify possible problem records, etc.
I'm not hugely familiar with authority practice, it's possible that
some institutions create authority records where they purposely lump
together various "Smith, John"'s that they can't distinguish into a
single authority record. That would seem to me to defeat the point of
having authority records. I would hope they do something to indicate
that this authority record is actually a "I don't know" record.
Like I said, I'm still learning some of the authority practices
myself. Still, I'm not a complete stranger to cataloging. I can't
see how my suggestion is any worse the typical author browse that I've
seen in several online catalogs as they currently exist. Some
examples comparing it to a current system and how it would be worse
might help me understand what you're getting at.
Jon Gorman
Received on Wed May 16 2007 - 10:29:53 EDT