Technologies for Accessing and Replicating Authorities
Nancy Morin:
I think this has come up in comments a couple of times. We've
been hearing from the library side that the Library of Congress plays a
really important role in developing the standards and processes. So I
was trying to think, what is our equivalent on the taxonomic side?
You're shaking your head, no, we don't have one. Well, maybe what Noel
and Sally just talked about [The International Plant Names Index]
should be turned into our Library of Congress-equivalent. Or perhaps I
should rephrase that as: "Could we turn it into our LC equivalent?"
Stan Blum:
There may be other opinions here, but mine would
be that, in a sense, this system has very different flavor from
an institutionally-based review process. Perhaps there are committees of
people in the Library of Congress, who approve subject headings, who
share similar training and perspectives, who act as filters on what
goes into the subject authority file. Perhaps the processes
surrounding the name authority file aren't so complicated and are more distributed and more
automated. But what this system [IPNI] does is to distribute the
editorial capability and authority out to exactly where the expertise
is. It doesn't require that they all be at one institution. They all
have access to, or membership in, a virtual institution. So this could
work very well in mimicking or facilitating the processes that are
going on naturally, that have been conducted on paper within our
community for all these years. This just accelerates the processes
because we won't have to wait for print publications to appear.
Eimear Nic Lughadha:
If I could give as an example a recent case where an author
published two alternative names for the same species, in the same
paper. In that situation, the botanical code deems that neither is
valid; each one invalidates the other. Now, when the system is up and
running, both names will be in there, with a comment to that effect.
In the mean time, because this was until now a paper-based system,
three people went into print saying "invalid", "stupid", whatever; but
lots of other people took up one of the new names because they had a
pressing need for it—it was for an interesting taxon that's in
cultivation. So we're applying rules that anybody can see and almost
99.99% of the time would come to the same conclusion, but some of the cases
are rare enough that people aren't familiar with those rules. This system
will get those decisions, which are in fact consensus decisions of
experts, out to the community faster, so we'll reduce the risk of names
being taken up when they shouldn't be.
Anon.(F):
You're making an assumption, though, that people have ready access
to the Web, aren't you?
Eimear Nic Lughadha:
Yes, I am, but in my experience more people have ready access to
the Web than have to the sort of funds that would buy the CD of Index
Kewensis or the hard copy, and I think that Web access is really
increasing dramatically. There are places in Brazil that have one
phone line in a village, and they can and do use it for connecting to the
Web. It really is increasing geometrically, and most places that don't have
access to a library of any size now are going to have access to the Web
in the next five years. Admittedly, that's an opinion.
Walter Berendsohn:
I believe the same thing, but if you have a bad connection, I think
that's much more of a problem. And that still might be a problem in
five years, unless something dramatic happens. I think this system, if
I understand it correctly, will replicate the entire database to other
sites. So the actual updating processes, the number of new names coming
in, is a comparatively small portion of the information that is coming—
and it can be easily transmitted. So you could have something like
the Library of Congress mechanism, doing something whenever is the best
time to load these things through your modem and phone line, and update
your local database in that way.
Eimear Nic Lughadha:
You're correct in your interpretation, and I would also add that
most working taxonomists, for instance, are actually interested in only a
small subset of this data. So you'll be able to download a subset that
you're interested in, and have a subscribe mechanism that will ask for
new items relevant to your subset. The total number of new names each
year is about 6,000. If you're looking at an order, you're only
looking at less that 1,000 new names. So you're not talking about big
volumes of data.
Randy Ballew:
Can you subscribe by geography as well as taxonomic classification?
Eimear Nic Lughadha:
Not yet.
Nancy Morin:
I was going to respond to something that Stan said in response
to my comment... which is: This mechanism for having community agreed-upon
standards for reviewing, and filtering, and so on... Well, part of the reason that
we've come as far as we have in plants is because of the Taxonomic
Databases Working Group. It's been our way of addressing standards on
a community-wide basis. I would think that the other disciplines
should be encouraged to get together and use that mechanism for
developing other standards that are needed. So that we keep moving
forward that way.
Paul Morris:
Looking at the last two talks, it strikes me from the point of view
of computing architecture, the way in which the Library of Congress
distributes authoritative records on subjects, is very very different
from this concept of distributing objects related to botanical names in
a system where there is no central server. But once you add in the
editorial control, from a perspective of sociology—how the system
works from outside—having those records tagged as having been viewed
and approved by an editor makes the system function in much the same
way. There is still a centralized control even though the computer
architecture is completely different.
Adam Schiff:
I just wanted to say again that while it is true that the subject
proposals go through an editorial process, name headings do not.
They're created independently by individual catalogers who have been
trained with our code, which looks like this [holds up book]—and this
is just a part of it. In some institutions they are reviewed regularly
by a principal cataloger or someone assigned specifically to work with
authorities. In other institutions, after a certain period of time of
review, they are released and are basically independent. They are
assumed to follow the rules and basically function as individual
editors. And in this case we're talking about thousands of people.
Karen Calhoun:
I'd just like to make a comment that the system for exchanging
authority records for names is almost fifteen... well, it's ten or
eleven years old anyway ... [Gap in recording due to tape change; --
sorry Karen.]
Laurel Jizba:
Somebody brought up the idea, which we do employ in the library
community... There are people who use these databases who are searchers
only. They can't contribute to it, and what they do with it we have no
idea, but they are charged for doing it. People who interface with it
as catalogers are functioning as if they were an editor. So there are
two different functions, and the ability to be a cataloger/editor, just
so you know, is a more complex interaction with the database than it is
if you're just searching only. If you [the taxonomic community] build
something that the library community could just sit and search, which is what we do
with a lot of databases anyway... (We are often charged either by
transaction or by connect time, and we're used to that.) ...it
wouldn't mean that we want to have any editorial role. We would just
be using it.
Lori Starr:
I guess I want to make a plea for help. We have a thesaurus; or rather we
work cooperatively with CAB International in maintaining theirs.
The CAB thesaurus is about 60,000 terms, so it's no small matter.
It's not as big as LCSH, but
it's still quite large and a job to manage. Thirty percent of that is taxonomic, and
I am not a taxonomist. I do come from a science background—kind
of turned librarian—and have a lot of respect for taxonomists and what
they do... But this is really a plea for help. We would love to have access
to the information you have, and to be able to use it in our
indexing work. We do the indexing and abstracting for the AGRICOLA
database—that's our agriculture and bio-sciences database for those
of you who aren't familiar with it—and our indexers have
to do a wide variety of things. We're not only
doing the high-powered, peer-reviewed, science journals, but we're
also doing things like extension publications. And with that we get a
lot of use of common names. Indexers have to cross reference terms, and if, for
example, the article is about the Colorado Potato Beetle, well one of our
rules is that we need to use the Latin name. Our indexers come from
scientific backgrounds (thank God), so they usually know what to use.
However, it's very helpful to put an entry in a file that might say: "Colorado
Potato Beetle USE Leptinotarsa lineata." (I'll defer to
the entomologists.) I think that from our perspective, we would
really like to have access to what you build, so
that we are not responsible for that 50%, because we are not
taxonomists. Also with our work with... and I guess I can speak for CAB International as well,
because he (Peter Whiteman) was very interested: "Are you going to that meeting?" He's a soil
scientist, and so when it comes to soil classification, well he's right in his field,
but when it comes to taxonomy... Well, it's something that we really rely on others for. Not
that CAB International doesn't have taxonomic experts—they certainly
do—but they can only do so much. It would be just wonderful to have access to these files.
So that's why I'm making this plea for help—We want your data.
John Mitchell:
And if I could piggy-back on that and add to your plea Lori,
I would say the Library of Congress would be very grateful as well,
because Library of Congress looks to CAB for many of the agricultural
headings we need to set up.
Michael Dadd:
Since there's nobody from CABI here today, as far as I'm aware,
perhaps I should speak on their behalf, because BIOSIS and CABI have
just signed a letter of agreement to cooperate and do a lot of things
together in future. This is not a merger, this is not a take-over,
it's a partnership, but I think that you may well find that there are a
lot of things in that area we will be able to enhance—taking our
taxonomic expertise, our taxonomic files, CABI's thesaurus, an ours—and
I think you will see a lot more coming from that area in the
future. [From audience: "What does CABI stand for?"] It was originally
the Commonwealth Agriculture Bureau Information Services; now it stands
for "C-A-B International"...[laughter]... It's a non-governmental international
agency which happens to be based in the UK because it originated with
the old British Commonwealth, but it's considerably broader than that
these days. It deals with agricultural biology as a whole. Fair
summary?