Although the intent of media as a tool to educate the public is sometimes realized, just as often mass media is driven more by market forces than by accurate presentation of information. Recently, NBC broadcast "Mysterious Origins," narrated by Charleton Heston, which blatantly presented false information regarding evolution and humanorigins. The show was recently rebroadcast, because controversy is more important than accuracy when it comes to moneymaking. New media like the Internet promise to be driven less by commercial interests and have many advantages over television, radio or newspaper. One of the greatest potentials of the Internet, still mostly unfulfilled, is to bridge the gap between research and education in the sciences. Currently, research is disseminated broadly in only a few major journals like Science and Nature. Most research is of special interest only to other researchers. Although the Internet can provide a means to allow communication between researchers (what I call server bounded research), more importantly it can allow access to and contextualization of material in a broader context (client bounded research). If research is properly handled on the Internet, I believe that institutions can make their products more valuable to society. I will discuss the notion of hierarchy of information in a laterally directed medium like the Web and present examples of bridging the gap between research and education taken from my work on the University of California Museum of Paleontology Web site.
The topic at hand is the use of the World Wide Web for disseminating museum
research. This topic immediately raises important questions: What qualifies
as research or scholarly pursuits? What kind of person is looking for this
information? Are we talking about the kind of person who is a professional in
a field or can a high school student do ìscholarly researchî?
Can such research exist without a context? Is a database of a set of images
something that should be considered ìresearchî ? Is
presentation of research different than presenting something that can
be researched by an end-user? What kind of professional uses do Web sites
have in the sciences and for science museums? I want to address some of these
questions in the context of my own work on the University of California
Museum of Paleontology Web site over the course of the last three of four
CLIENT AND SERVER BOUNDED RESEARCH
I have read a number of articles in the local newspaper recently about how people are researching questions of importance to them on the World Wide Web. One title was ìHow to research loan rates on-lineî. The gist of these articles is that people are using the search engines available on the Web to find out questions of importance to them. Let me give an example. I have recently bought a freshwater aquarium in which I would like to grow snails. This aquarium is a vital part of my research, since I need living snails to breed and deposit eggs. These eggs will be used for my biological research on the development of snails. The first thing I did after reading the relatively uninformative directions that came with the tank was sit down at my computer terminal and search for information about both aquarium maintenance and about the golden apple snail (the kind I bought).
I found out a lot of important information on the Web about both these topics; not only information that could be summarized in a book on aquarium maintenance, but also information archived in email digests from people who have dealt with similar issues. Some of the advice that I read would never make it into a book, and books on raising freshwater snails are extremely rare to begin with.
This kind of research is what I call client bounded research. In this case, a client is trying to compile a bunch of information together from disparate sources, possibly disjointed sets of different kinds of information; email archives, Frequently Asked Questions, Web Pages, whatever. I call this client bounded research, because in this case it is the person using Netscape or Internet Explorer that has an explicit question that he or she wants answered, and is going to spend the time compiling and weeding through many different places on the Internet to grab this information. The client is responsible for the synthetic kind of work that qualifies as research.
A separate kind of research is server bounded research. Here, the people responsible for putting information on-line are doing the synthesis and presenting the results of this work on the Web; this is presentation of research through the Web. Server bounded research is research that is being presented via the Web as research. The best example of this kind of server bounded research is the Los Alamos E-prints server. Here, physicists and mathematicians submit completed papers to the server, the abstracts are text based and actual reprints are usually in postscript format. The people most likely interested in the preprints are other mathematicians and physicists, but the e-print server is open to any who are interested. A friend of mine relates a similar story from computer science. Most computer scientists no longer submit journal articles to publications geared towards the field because the turn-around time for publication is too long. By the time the results come out in a journal, the article is obsolete because the field moves so fast. The Web has provided a way to quickly and effectively disseminate research in computer science. Preprint archives in computer science are now the source for up to the minute results in the field. Even fields where results are probably still important a year from the time the article is submitted are moving towards electronic publication because of the flexibility of the new format. With the Web, results are not limited to static pictures but can include movies, three dimensional projections and sound.
As scientists begin to couple the low costs of electronic publication with the advantages of the new format, scientific publication will move toward more web based publications. Recently in Paleontology, the first completely electronic publication started; Paleontologica Electronica (http://www.ucmp.berkeley.edu/Paleonet/pe/glines.html), sponsored by major Paleontological Societies. The sciences potentially have the most to gain from electronic publication because the data is often difficult to represent in a static medium like journal publications. For example, the ability to present three dimensional views of objects through virtual reality engines may be particularly important for structural chemists.
Above I mention mostly journal articles as server bounded research because in the sciences, with the possible exception of computer science, the currency of research is peer reviewed publication. Important information is archived most effectively on the Web and can be extraordinarily important in research questions. For example, GenBank is a database of different gene sequences for many different organisms that can be used by researchers all over the world for answering comparative questions. GenBank is an important source of information for a researcher and an important research aide but the question to be answered still drives the show and the place where the question is answered is often a peer reviewed journal.
The distinction I am making between client bounded and server bounded
research blurs at the edges, because a client can certainly use server
bounded research (research that is being presented as such) in the context of
a larger or different question; the two are not mutually exclusive.
Research is one of the most important currencies by which the Museum of Paleontology is judged by the campus. Although affiliated with the University, the Museum does not have direct ties to any one department but is answerable to the vice chancellor of research. In a setting such as this, making research widely available is an extremely important part of our mission. Web publication removes many of the barriers to disseminating research far and wide, and in new and different contexts.
What are some examples of research applications at the Museum of Paleontology (UCMP)? PaleoBios is a UCMP journal that primarily publishes research done on the collections or by researchers at the museum (http://www.ucmp.berkeley.edu/museum/PBSI.html). Although PaleoBios has a fairly wide distribution to over 200 institutions worldwide, the Internet has opened a new door for both promoting the journal and presenting research published in the journal. We have recently begun putting one article per volume on the Web as an electronic publication. We eventually hope to move more towards concurrent electronic and traditional publication. We also put abstracts and titles of all papers on the Web, so that people interested in particular issues may contact us. When a new issue is published, we contact the community through established listservers and newsgroups so that people in Paleontology know a new issue has been published. Since we began using the Web to present and promote research at UCMP, we have seen a significant increase in number of subscriptions. At the moment, the translation from PageMaker to HTML is crude. A filter exists that will convert PageMaker files to HTML, but the conversion is sloppy and what you see in PageMaker is not what you get in HTML. This will change in the near future and filters should be able to make nearly exact copies of PageMaker documents represented in HTML.
Other bodies of research produced by UCMP exist only as Web documents right now, although they may eventually be published as hard copy. For example, UCMP has just developed a datamodel that can be used for any Paleontological collection, and is probably extensible to any biological collection (http://www.ucmp.berkeley.edu/museum/datamodel/datamodel.html). The datamodel represents two years of research into database design and the needs of modern paleontological collections. This information is obviously of use to the museum community. The fastest way to make this model available to the rest of the community is conversion to HTML and presentation via the Web.
Another example is three dimensional reconstruction of microscopic anatomy (http://www.ucmp.berkeley.edu/people/rpg/radula.html). These three dimensional reconstructions and VRML only exist on the Web and supplement published work. Three dimensional reconstructions and VRML are valuable because they show the power of the Web to represent and allow manipulation of scientific objects in ways that journals or books cannot. Other museums besides UCMP have begun exploring these possibilities. The National History Museum in London has an exhibit of VRML trilobites that Web users can spin around and examine from all angles.
Another area where research can be presented is in the context of education.
Although I do not intend to discuss education at any depth, research can be
contextualized within an educational mission using the Web. We call this
infinite hierarchical depth. We can start with very general
information about a topic and move deeper and deeper into the meat of the
topic culminating in presentation of research findings. If people are
interested, this kind of integration of research and education has the
potential to demystify the research that scientists do. For example, our
Tyrannosaurus rex exhibit starts off with very general information
about the lifestyle of the mighty predator. As we move deeper in the
hierarchy by clicking on hypertext links, we would learn that T. rex
had huge, banana shaped serrated teeth. Finally, we might move to some
research done at UCMP about the bite force of T. rex
(http://www.ucmp.berkeley.edu/diapsids/locomotion.html is an example).
This contextualization makes the research done at a museum more valuable by
providing information that is impossible to get in a real museum due to space
constraints. The physical space of a museum can never provide as much
information as the virtual space because physical space is expensive and
virtual space is cheap. Therefore museums ignore one of the greatest
advantages of the Web by not using the medium to present general and
specific, educational and research orientated material as a cohesive,
The Web can be an important resource for researchers looking for crucial pieces of information. Organized information like databases are extremely important in this regard. GenBank and other gene and protein on-line databases has been an incredibly important way to share information vital for research questions among everyone involved in the field. The rise of bioinformatics in molecular biology has not been followed by a concommitant rise of museum informatics, although many signs point towards growth in this field. Museum informatics has the potential to revolutionize some aspects of museum business and to make obsolete other aspects. In museums that have collections, like many science museums, the Web will become the standard interface to collections information.
At the UCMP, we are implementing an ambitious database project that will provide a model for other museums with collections material. The actual datamodel is relatively specific to biological collections, but the implementation of this model is one that could be used for any data model for any collection. The first part of the implementation is the database. For the moment we use Paradox 7.0. The data model is relational and therefore the database we use has the capacity to establish and check relationships between tables. We use the same piece of software to build front ends for museum curators and as front ends for Web users. Delphi2.0 provides immense flexibility for designing modules that allow for searching and formatting database tables into a usable format. Delphi also has Web components that automatically passes information from the database to the Web and also on the fly constructs a table of results. The forms based Web interface can also be used to post accessions directly into the database. Workers with portable computers circulating through a collection can both accession the material into the database and put the physical object in the right spot in the collection. People accessing the collection never have to print out records before walking into the collections. They can simply roam the collections with the portable computer.
Remote use is also an important advantage to having a Web interface. People from other institutions can instantaneously determine if UCMP has material they may need for research by searching the on-line database. Ultimately, we would like to have pictures of all our specimens available on line, but this is a daunting challenge with huge collections. Images of science collections may not be as valuable as most museum people interested in computerization think. The task of collecting the images is labor intensive and the rewards are often small. Descriptions, even a thousand word description, is smaller to store the pixels used to create one image and potentially more valuable especially with context. However, our datamodel is designed to accommodate specimen images should we want to pursue that tack.
Since the conception of our Web site, UCMP has had a collections interface
based on a WAIS index generated from an ASCII database
(http://www.ucmp.berkeley.edu/collections/catalogs.html). The WAIS index
solution has many disadvantages; slow speed of searching and huge files with
lots of redundancy generated from the ASCII database. Although not elegant,
this solution was pragmatic in 1993 when the kinds of interfaces necessary
had yet to be built. For those museums who want to put databases on-line, as
long as the database exists it can be put on-line in a matter of a couple
hours using WAIS. This solution has served us for over three years and we
have found that just as many non-professionals are interested in our holdings
as professionals. Even though collections serve mostly a research purpose,
museums will need to justify their collections beyond the research value.
Partially, collections are important because the material has some value to
the culture. By showing as many people as possible both the specimens and
the data associated with specimens, museums can better justify their
collections and the data associated with those collections.
INTO THE FUTURE
The Web promises to remove distinctions between research and education and web designers may facilitate compromises between research and education departments in stand-alone and university affiliated museums. Many people argue that the Web will create problems like an information glut or a shifting sea of information that will never be citable because the location of the information changes or disappears. These problems are also some of the greatest advantages of the medium. The Web, both the software that makes it work and the information that is presented, continues to grow and change. As an evolutionary biologist, change is the rule not exception, so the Web seems vibrant and alive not musty and stale. Although the information is not constant, those workers who are presenting research will be the ones most likely to take great care to archive information and try to keep pointers when the information changes location. It is the useless tripe that usually vanishes without trace and good riddance.
Museum exhibition is often an attempt to build an immersive virtual environment replete with sounds and sights that attempt to evoke visceral responses. The future of the Web, especially with regards to museums, is to capture that aspect of the museum experience. Exhibition often only exists because we understand those environments, and understanding gets back to the research. As the technology and integration of technologies (separate issues!) grows and becomes more rich, the research and education that we can present over the Web will also become rich. Museums are not only storage places for objects, but are places where that information is made meaningful partially through research done on those objects. The Web allows us to send that message out millions of times each week!.
Last modified: February 28, 1997
This file can be found below http://www.archimuse.com
Send questions and comments to email@example.com