I have a couple of questions related to economic models, the
first is: given your organizational status, why do we consider
you commercial, when you're no different from the Philadelphia
Academy of Natural Sciences? You're a non-profit, but you just have
to pay your bills?
Michael Dadd:
That's what we are, undoubtedly. There are some parts of the
community that seem to have the impression that we're
commercial—perhaps it's because we have a marketing division,
which sells things, which goes out there and tries to persuade people
to pay up—but we are definitely are not commercial in the make a profit sense.
Scott Miller:
But then the second question is: how does the Federal government
putting its literature databases—namely the National Library of
Medicine that's on the World Wide Web now, and the AGRICOLA database
that's coming onto the World Wide Web—how does their becoming
available for free effect your marketing strategy for Biological
Abstracts?
Michael Dadd:
It stands a very good chance of destroying Zoological Record.
[SM: "Zoo. Record or Biological Abstracts?"] Both, because it could
destroy Biological Abstracts, who support Zoological Record, and not
only Biological Abstracts but a number of other secondary services.
We have very good relations with NLM, and understand and know why they did
what they did, but the fact is that the actions that they took have
had a very major impact on any secondary service handling life
science information. I would say also that it's very worrying that
when you talk to many people, many scientists, they use the MEDLINE
database because it's free, and they don't look outside it, and they
don't realize how little of the literature they're picking up that
way. But the impact is that it takes out a large part of the market
from which it's possible to get revenue, and makes it harder for any
secondary service working in the life sciences to survive, and
therefore makes it harder for those services to subsidize parts of
the operation, like Zoological Record, which recognize the
importance, but can't generate sufficient revenue on their own,
directly.
Gary Rosenberg:
I was wondering if you know anything about how GenBank works.
Would it be possible to set up a system in which the systematists or
the primary people enter the data into a system like Zoological
Record?
Michael Dadd:
I think that GenBank is a somewhat different situation. On the
whole it's an area which is much more strongly funded and where
people tend to work in a much more centralized manner. There are
people at Los Alamos who will go out on the weekend and watch birds,
or collect bugs, and write observations, and do some useful work.
There aren't many biologists who go out on the weekend and do
high-energy physics research. I think it would be difficult,
perhaps, to deal with it that way. It is the way Zoological Record
was compiled for a long while. It was being compiled in that way
when I joined it in 1966. Almost all the input was done by
individuals, and it was very difficult to get the work done on time,
and it was very difficult to get it done consistently, and it was
very difficult to have useful and efficient retrieval. That problem
goes right back to the first volume. If you look at the first volume
of Zoological Record, in the Introduction it says something to the
effect of "includes these sections, except for one particular group
where Mister X failed to honor his obligation to provide this
material on time." What we would like to do, and I've had one or two
discussions with people in this audience, is to see if we could find
a way of building on the TRITON-ION system we've got there, to
encourage people to improve and build that data, and to do that on a
non-cost-recovery basis. We have the Nomenclator Zoologicus
in machine-readable form. It's not perfect because it was
OCRed—same problem as Kew's Index Kewensis. We're talking about
putting that up, accepting that it's not perfect, and saying: "Will the
world community work with us—on a we're not charging and nobody's
getting any money out of this, but we're all getting a better tool -
basis—to try and improve that data?" We would like to do that for
other legacy data, but I think the basic compilation of the database
needs to be done as a coordinated, professional activity.
Stuart Nelson:
I was thinking, as you were talking about the comparison of
Biological Abstracts to Chemical Abstracts, because I'm in the middle
of negotiating with them over the use of their registry numbers. They
make good money and support a really important
activity, essentially by the use of these registry numbers, which are
used by the chemical and pharmaceutical industries. And I'm wondering,
where is the analogous market for that kind of biological
information?
Michael Dadd:
Well I suppose at the moment we don't have one. We have a system
of registry numbers in addition to the scientific nomenclature of
chemicals and it's the registry numbers that Chemical Abstracts have
effectively patented in use. We don't have that sort of system in
Zoology and we certainly don't in Biology. We certainly don't want
to control the names, but I think it is through funding something
like Species 2000 as a central register, and getting some of the
governments and agencies to fund that, which is perhaps the best way
of doing it. In some sense that's the equivalent.
Mary Mickevich:
Could you please tell us how much of the Zoo. Record budget is
self covering. What's the proportion?
Michael Dadd:
Perhaps 25 percent. It's a big enough sum. We're
talking about hundreds of thousands of dollars, and not
small sums.
Stephanie Haas:
I remember hearing that, if you're doing DNA work, before your
paper can be published, you have to submit the sequence to GenBank.
So they have a built in mechanism for having people contribute,
because your paper doesn't get published unless you put your sequence
in.
Michael Dadd:
I think the simile in biology is registration of names.
Stephanie Haas:
OK, but it would be also interesting if somehow that could be
incorporated into the literature aspect of it, as far as into the
indexing.
Michael Dadd:
Well, yes. We tried it, but it didn't work.
Stephanie Haas:
The other comment I have, just from listening as a member of the
library world, is that it seems to me that a lot of the income that
supports the taxonomic indexing that Zoological Record and Biological
Abstracts provides the taxonomists, comes from the academic library
world.
Michael Dadd:
It comes from the users as a whole, including academics,
pharmaceutical companies. Obviously, we are cross-subsidizing within what
BIOSIS does as a whole, but the academic...
Stephanie Haas:
But from the subscription base for your products...
Michael Dadd:
Our subscription base is about 50% academic, in cash terms. In
numbers it may be a bit different, but in cash it's about
50%.
Eimear Nic Ludghadha:
Just as we don't get amateurs doing high-energy physics, we don't
get amateurs doing DNA systematics. So the
sort of people who are doing DNA sequencing would have access to the
technology where they can easily submit their sequence. Whereas we
do get people in very out-of-the-way places publishing new names, and
we (Kew and at least some of the other participants in the Plant
Names Project) would like to maintain that. I think you implied, in
your talk, that there was an increasing level of comfort, with
respect to registration, on the part of botanists, but I don't think
I'd like to leave people with the impression that the vote on
registration is by any means a foregone conclusion. I think,
particularly in North America and definitely in South America, there
is very strong opposition to the concept of registration.
Michael Dadd:
I'm not the person to speak to that; I'm not a botanist. There
are probably other people here who are better qualified, but I think
that we need some system along those lines. My understanding is that
what is being questioned is the mechanism, rather than the principle,
and there are a number of methods that have been tried out and that
this is one of them. I think if you look at the reports of the Tokyo
congress, it's perhaps a little bit more positive than you're
suggesting.
Eimear Nic Lughadha:
With respect, I think we can say that the report of the Tokyo
congress was compiled by one of the major proponents of registration,
and that it's not a question of various methods having been tried. A
single method is now being trialed, and there is a lot of opposition
to that particular method.
Michael Dadd:
As a zoologist, I don't agree with the mechanism which is being
used anyway, but that's a different issue. I think the important
thing is that we need some sort of principle, some sort of place
where you can go and look these things up.
Chris Thompson:
I would say registration is a frustration for a lot of us. It's
worked for bacteria. Am I correct? [MD: "Yes."] But in the
zoological community our feedback—when we put a mechanism into
our code, which wasn't formal registration, but sort of notification
that papers containing new species would have to be indexed in the
Zoo. Record for those names to be valid—the community input was
about 40/60 against it. The fear was that the known record [past
performance] of Zoo Record to be comprehensive was not good enough to
trust them with this registration of papers concept. There was no
recognition, that if you incorporate it [the requirement for
notifying Zoological Record into the International Code of Zoological
Nomenclature]—this is a chicken and egg phenomenon—that it
would increase Zoo. Record's ability to be more comprehensive,
etcetera. Nevertheless, there is a fear that the amateur somewhere
in Mongolia would have his new species not recognized because they
were not indexed. So they rejected it.
Michael Dadd:
I think I do really need to answer that point, because I think it
is very relevant—I think there are two things I would say.
Firstly, we were trying to demonstrate that it was possible to
provide this information freely and easily to everybody, and
secondly, that some of the comments that were received about coverage
of Zoological Record and its inadequacy in that respect, were in our view based on
what was done about 50 or 70 years ago. We have a very different
system in place these days, a much better system. I think that we
could do a reasonable job.
Jim Beach:
I'd like to ask a broader question, and maybe put Michael and Bob
on the spot somewhat, but also ask the speakers for the next session
if they could address it as well, and that is, I get a little
confused about what an authority file is. We've been talking, for
the past several talks, about authority files, and I'm trying to
understand the ecology of authority files. I'm wondering if a single
person, or a single company, or a single museum, creates a species
checklist or a database, is that an authority file? If a product
like that is created, but not applied to any other database, say for
controlling the quality of data in a catalog, is that a requirement
or a required function of an authority file. Is there a difference
between an authority file and a species checklist, in terms of
community infrastructure? Does the community have to be involved in
an authority file? Getting back to my first point, can a single
entity create something and declare it unilaterally as an authority
file for names or accepted species? I guess what I'm trying to get
at here is not the semantics of the phrase "authority file", as much
as I am the essential functional attributes or features of authority
control and authority control within a community. Are we talking
about two different kinds of things here?
Michael Dadd:
Jim, can I try and quickly answer that, and then give Bob a
chance, and perhaps leave it to other people after the break.
I think what we're
doing and what I've been talking about is not, in fact, an authority
file. What I'm talking about is a name list, a nomenclator, exactly
the same comment that Kew were making earlier on this morning. We
have lists of names that are put together in a form that are usable
by people. They're a bit more than a spell-checker, but perhaps not
a great deal more than that. I think that the authority files are
things that derive from activities like Species 2000 and the global
species database that Frank has within that. The work that ILDIS has
done is undoubtedly a taxonomic authority file, to me. But that's a
very different category of information than what I put
together.
Stan Blum:
I would like, in some sense, to disagree. I'm not sure it's
useful to make those distinctions. In fact, you see that the
nomenclators are authorities that need to be used in the compilation
of taxonomic authority files, those that include checklists and
taxonomic decisions. We need authority files about people —the authors of names, we need... Well, there's a whole network of values...
Michael Dadd:
There's a hierarchy of authority files.
Stan Blum:
Yes, exactly. What I'm trying to generate out of this meeting is
a greater awareness that there are different users, perhaps, for
different pieces. But the basic idea is to create consistency across
a distributed set of servers, or information resources.
Frank Bisby:
Like Jim, I'm concerned about that word "authority" in there. I
think that it would be great if, rather than calling them taxonomic
authority files, we called them taxonomic resource files. There have
been several moves at TDWG to try and stop people talking about
authorities, and talk instead about authors of names, authors of
resources. So again, when you use the word taxonomic authority file,
what does the word "authority" mean? Do you mean some special
quality or just that it's a file.
Stan Blum:
Well that might be a good thing for the library people to
address. I think the term authority file comes from the library
community. We just borrowed the term, and it describes a functional
role. So we'll hear a little more about the library community's
definition of authority files later this afternoon, and perhaps what
imparts authoritative status to a record. We [the taxonomic
community] probably are going to have those same functions at all
those different levels.
Chris Thompson:
But we have heard about authority files. ITIS, whether we want
to say it or not, was derived from an authority file called NODC,
which is a set of numbers used by bureaucrats to index, and so
forth.
Stan Blum:
So that's a functional definition of an authority file.