Salve!
Cataloguing in general has been stewing in my brain for many years, and
cataloguing on Koha in particular. I started on a DRA system years back,
and I could hear a game show buzzer in my head when I saw the interface.
Then I remember colleagues getting all excited to look at Dynix Horizon
new cataloguing module. We just kind of looked at one another after the
sales guys had gone away. That wasn't it either, but it wasn't game show
buzzer bad. III was kind of nice, and as I became a dork and would ask
people I didn't know to view their cataloguing systems as the years swam
by I came to the conclusion that some were better than others but none
were particularly well liked. I've long hated the one that's come
bundled with Koha, and I've long though "Oh come on, I know we can do
this in a sexy way!" If we put together a module that was better than
everyone else's, that would be the proprietary nail in the coffin, and I
don't think it's terribly difficult to do since everyone else's bar is
so low. I arrived at the inevitable conclusion that this was one of many
areas where it was going to be hard or nearly impossible to get fellow
Librarians to be very specific in terms of what they would like to use
and not just what was in front of them. I feel like it's the aesthetics
and interface of things more than anything else. Big warning: I'm not
properly a cataloguer, but gosh over the years my cataloguing disguise
seems to be getting very good.
The time to think about a major overhaul is *now* since the RDA proposal
is on the table.
>Hopefully starting on a new thread. Just in the last few weeks we've
had
>a couple of clients who would really benefit from an easier way to do
>their own cataloging.
>
>They don't (Sadly) find their specialist materials being well serviced
>via the various z39.50 servers etc, often because they are doing
>cataloging of materials they are producing themselves, or from other
>government departmens/private publishers essentially.
>
Yeah. There are an absolute tonne of special Libraries that simply
*can't* find records that would match their own since they're drummed up
in house or are so rarely held that there aren't records to be found to
match. Engineering, transportation, medical, and legal come to my mind,
just to name a few. I shudder to think about folks not having someone
around that could do original cataloguing, but after a number of years
on this project, I know from some of the cataloguing questions that I
get that folks can't afford someone or just don't want someone that can
do original cataloguing.
Many times in this sort of environment, folks want a simple text box for
Author, Title, and I think they ought to have subject just to round
things off for the big three. I'd argue that a URL needs be in there for
electronic organisation of data.
>As alluded to we (Katipo) did have an interface to do this in the very
>early versions of Koha, and I still have essentially the spec for it in
>the form of a working version, and screen shots/manual type stuff - we
>called it simple acquisitions, and in the current version of Koha you
>can see that there is a preference for either simple or normal
>acquisitions - of which only normal is really there.
>
Find another chump for acquisitions, cause I've not got mah head around
that. ;)
>I've been toying with the idea of seeing what's involved in essentially
>adding a simple acquisitions module back in again - but saw that Galen
I
>think is planning something similar sounding as a simple MARC module.
>
>So I'm wondering if the two ideas are close enough to be the same
thing?
No, at least not from where I think. Again, I'm not expecting this any
time soon, but I'd like to see it as a road map goal, even if it's a
couple of releases out. (As in 4 or 5.)
I think that we have several different approaches to cataloguing itself,
and I think this is the root cause of why all interfaces suck. Something
that programmers might want to consider is having permissions tied in to
the different approaches, so that an amateur would be able to use the
simple module, but not the pro version.
I think there ought to be about 4 ways to handle a record in Koha, 2 of
them are close enough that they can live under the same roof. We should
have what Rachel is asking after.
Simple Cataloguing (since it's simple and she's got screenshots of what
it used to look like, there's not really much more work in bringing this
one back.)
then I think we definitely need a new version of
Pro Cataloguing
I still don't totally like what I see with other products in terms of
interface, but both Biblios and current cataloguing aren't it. I've seen
people whip up an original MARC record in minutes tops. They want a
blank playbox (that's a big old textbox, as for email) that will
recognise MARC fields.
I think we can go one better than that simple of an approach by making
that textbox operate more like Dreamweaver. Why not have those fields
change colour, and then have the tooltip display the name of the MARC
field in question and perhaps a R for repeatable and a N for non
repeatable. I think that would be a neat way of catching the cataloguer
before they've had their java, since you'd be able to just hover over a
blue 110 and think "Oh! Of course that's not Additional Author! Duh!"
So, mostly graphical. Mostly blank. Mostly leave the professional
cataloguer the heck alone and let them do their thing. The comment I
hear the most on any ILS cataloguing interface is something on the order
of "Maybe this is because I'm a little older than you, but I just want
my typewriter and card back. It was easier. I'm not opposed to
technology, but man, I could just do that in a few minutes."
Copy Cataloguing
Same approach as above, only a z39.50 button someplace on that same page
that would let you very simply import a record by ISBN or Title. Here's
the twist. eXtensible Catalog right now has The Thing that I was talking
about with authorities. It goes through and finds Evanovich, Janet and
Evanovich Janet. and Evanovich, Janet. and possibly even Evanogich,
Janit, and asks record by record if they need to be deduplicated. But
here's how The Thing would be applied to copy cataloguing.
Disagree with me here if I'm getting out on a limb, cataloguers, but in
general, when I was copy cataloguing with style, the longer the record I
found via z39.50, the better the record was. The exception to this was
the odd occasions where someone would apparently have copied and pasted
into a record multiple times. We should be able to tell how long that
record is before we import it. I don't know whether that's a constraint
of z39.50 or no. The system should check internally how many characters
are in a given record in the databases we point to and report a little
summary back.
I'd even love to have ratings creep into this, a la digg, though that
may well ruffle some feathers. Is MaineInfoNet good on local stuff,
cool, comment and mark em up. Does AccessPenn not do so great a job with
children's records, comment and mark em down.
A lot of people in rural Libraries end up using Delicious Library.
That's the sort of ease to strive for on the first page of copy
cataloguing. We could always just have a prompt if someone wanted to
edit further after the initial ISBN scan.
Guided Cataloguing
A long time ago when animals could talk, I had an AIM discussion with
Nicole about cataloguing. Why is all of this still a one way street that
starts with my brain? Aren't I essentially taking different paths down a
big flowchart in my head? When I hold an item in my hand and sit down to
catalogue, aren't I going through pretty much the same thought process
time in and time out? Yes, it's very complex. Yes, it varies per item.
BUT there are questions in common. This is why those veteran cataloguers
can whip up a very high quality record in no time flat. They've mental
muscle memory for cataloguing. And we have them. >:)
Why can't a programme ask the questions that I have in my head in a
series and formulate a MARC record from it? Why can't it show me what
the verso is in the process? Why aren't we leveraging linking when
there's a world wide web?
This is the hardest part since no one's tried doing it before that I
know of.
>Anyways, if anyone has strong views one way or the other let me know.
>I'm a bit biased against the current MARC interface - we get people
>being intimidated by it (me included :-) but I'm prepared to work
>through it.
>
I think that's from too much text and boxes on the page, plus all of
those tabs, like my long winded emails. You're not alone in being
intimidated. I wasn't but I know a bunch of people that saw that and
said "You're joking! That's a lot of fields!"
Cheers,
Brooke
