This page last revised 23 November 1999
The EPO presentation given by Ferdinand Rudolf is now available online.
The CAS presentation
by Robert Massie is now available online.
Welcome cocktail and buffet dinner sponsored by Chemical Abstracts Service
Patricia
Rougeau
MDL Information
Systems, California, USA
The impact of new technology
on chemical information management
This decade has seen major
advances in the technology of chemistry and information management.
Automation, miniaturisation and robotics techniques applied to
chemistry and screening have placed new demands on scientists
and information specialists alike. Chemists now plan for and manage
thousands of syntheses at a time, and record results for massive
experiments. Interpreting screening results is a challenge when
scientists are faced with new data on thousands of compounds.
There is an unprecedented need to help scientists plan and record
their work, and make decisions based on the data generated.
Information technology
is key to meeting these challenges. Today's object relational
database systems and rapid application development environments
enable developers to design tools that closely fit scientists'
work flow processes and provide ready access to personal, project,
enterprise and public data sources. With the help of modern information
technology, scientists can focus on the science, not on recording
and finding information.
Huge productivity improvements
can be made by supporting scientists' workflow, but the biggest
gains will be made by systems that help to turn data into actionable
information. Data visualisation and mining will be differentiating
technologies. Organisations that enhance their scientists' ability
to analyse data and zero in on new products more quickly and cost
effectively will be the stars of the future.
David
Evans
Claritech
Corporation, Pennsylvania, USA
Text mining for information
co-ordination and discovery
Decision support includes alerting and event
detection, trend analysis, and quantification of text information.
It is a product of advanced functionality (information retrieval,
information analysis, information organisation, and decision support).
Text in databases causes many analysis problems since it is essentially
unstructured, of varying quality, often using non-standard language
and can be of large volume. This presentation suggests that text
mining should be implemented in organisations, since its potential
value is great. A number of desiderata for text mining technology
are described, including graphical presentation of results, a
user interface to shield the user from underlying complex processes,
automatic document structure analysis, and integrated free-text
processing.
René
Deplanque and V. Schubert
FIZ Chemie,
Berlin, Germany
A network for chemistry
education
The Networks for Chemistry
Education project plans the development of interactive knowledge
modules and information tools, using VRML- and XML-technology.
This system will be made available using the internet as its distribution
medium. Under the project management of FIZ-Chemie, 16 professors
from 13 universities from eight German states aim to create an
electronic platform to enable engineers and scientists to access
worldwide chemical knowledge in a problem-oriented way. Current
state of the art is to search a database for information relating
to the task at hand, but still missing is to be able to search
and interlink all necessary basic and advanced knowledge (learn-modules)
tools to understand the underlying science and processes technologies
as well as business applications.
Using a modular approach,
the system could be used not only as a help in basic studies for
college students and postgraduates in chemistry, but also for
students in polytechnics, schools and for trainees in industry.
Whereas there are many approaches for multimedia teaching aids
for various subject areas, there is no combined effort to treat
all of chemistry and related areas such as patent law, accountancy,
etc in a combinable (network) and complete manner. This project
is one of the major strategy projects of the German government
and has a budget of more than €25 million. Its main strategic
goals are to shorten the time needed for getting a chemistry degree
by optimising the learning process. But equally important is to
shorten the product-to-market cycle in industry. This will be
done by shortening the time which is normally needed to transfer
new knowledge into the learning process in industry. By creating
a better understanding of the process technologies, a faster turnover
and better quality of product can be achieved. This presentation
explains the structure of the project and demonstrates already
existing multimedia tools and basic modular networks to show how
the above goals can be reached.
James
Myers
Environmental
Molecular Sciences Laboratory, Washington, USA
eResearch: The rise
of scientific virtual facilities
Collaboratories and virtual
facilities are a new way of organising and performing scientific
work that holds tremendous promise. Researchers accessing these
facilities remotely can securely control instruments, run analysis
and visualisation tools, store notes in a shared electronic notebook,
and converse with colleagues using videoconferencing, whiteboards
and shared applications, as easily as if they were onsite. Pacific
Northwest National Laboratory's Environmental Molecular Sciences
Laboratory (EMSL) is a new national user facility that is adopting
the Collaboratory as a primary means of supporting users and interacting
with collaborators and partners. The EMSL Virtual NMR Facility,
already being discussed as a national model for future NMR facilities,
provides a good example of the state-of-the-art, and of the specific
benefits that can be obtained. (Details of the EMSL Virtual Facilities
efforts are available at http://www.emsl.pnl.gov:2080/docs/collab/.)
Pierre
Buffet
Questel-Orbit,
Paris, France
Does value added information
any longer have a strategic value?
In the main, the information
community is still committed to the old business model of 'pay-as-you-go'
that came into being in the 1970s. However, the history of online
information projects suggests that the strategic value of information
as perceived by the marketplace is a fragile one. On the one hand,
there has always been a strong demand for value added information
from users, as the history of the Markush search systems in the
mid- 1980s suggests. On the other hand, the revenue realised from
such high value-added projects suggests that, when faced with
the financial test, the real demand for such value adding is ephemeral.
The presentation examines these paradoxes and suggests possible
paths for evolution of services that are required by only a small
minority.
Mary
Berger, Elsevier Engineering Information, New Jersey, USA
Re-inventing the database
business. Getting from the 1990s to the 2000s
Elsevier Engineering (Ei)
acquired the API EnCompass service and its three databases from
the American Petroleum Institute in April of this year. As part
of the deal, Ei obtained the right to use the API name, under
licence, for the files APILIT, APIPAT, and API EnCompass News.
Ei came from the highly traditional Compendex database (Engineering
Index) and has steadily developed itself into an internet-based
engineering and technical information site.
The API files have always
been traditional online databases covering patent and industry
news for the petroleum industry. Ei is now investigating how to
leverage its new petroleum files to enhance its business profile
in the age of the internet. This presentation details the possible
factors and strategies that are propelling the database industries
in a race against time to adapt their traditional products to
the new information age.
Martin
White
Intranet
Focus, London, England
Changes in the legal
and regulatory environment for information: what users need to
know
The information community
has always required a familiarity with the basic rules concerning
copyright. Increasingly, however, these rules are being supplemented
with various EU directives and new national rules and regulations
concerning areas such as database protection, and privacy. In
addition, the fact that so many information workers are now active
in a networked environment such as the internet means there are
additional areas of laws and regulations that need to be taken
into account - rules dealing with matters such as libel, slander,
trademark abuse, defamation, etc. This presentation looks at the
legal and regulatory environment from the user point of view,
and concentrates mainly on new areas of regulation that are appearing
because of the environment of information that is delivered electronically.
Robert
Massie
Chemical
Abstracts Service, Ohio, USA
SciTech information
at the millennial boundary
At the end of the
nineteenth century, fin de siècle came to connote decadence
rather than progressiveness. Fin de siècle in today's terms
connotes a pace of technology change that induces vertigo in some
and instant riches in others. The chemical information sector
seems a trifle long on vertigo and short on instant riches, but
our fast-paced evolution is worth analysis. This presentation
reviews the most important influences on our sector and speculates
on their impact in the next five years. These influences include:a
movement in the academic sector to exert more control over the
distribution of their research findings and to rethink if not
overhaul the traditional role of journals;the role of government
entities, especially patent offices, in providing taxpayer subsidised
free information – most recently this includes an NIH initiative
for biomedical publishing on the Internet;the dramatic potential
of electronic journals and other primary information available
online, and new possibilities for an interlinked Web environment
(the five key impacts of the WEB, and their changing meaning for
the chemical information sector);the continued impact of capital
markets, scale economies, and for-profit business practices on
a sector that was academically oriented for much of the century.
Survival and success
in this fin de siècle world demand a strategy of continual
improvement and evolution, if not reinvention. CAS' own evolution
is perhaps more dramatic than appears on the surface. It mirrors
the changes within the chemical information sector and are discussed
briefly. Several new developments are profiled.
Georg
Schultheiss
FIZ Karlsruhe,
Germany
Classic online hosts
in the scientific and technical information area: the need for
new strategies for the new millennium
It can be claimed that
the classic online hosts will disappear within a few years because
of the development of the internet. The internet marketplace has
created a demand for new distribution and handling paths for all
types of goods, including information. But time is needed not
only for the development of these features, but also for the ability
of customers to adapt to them. Although a change in generations
means that the attitudes towards computers and digital material
will also change, this has not yet resulted in a dramatic break
in information requirements and habits within organisations. Using
the experience of the STN service centre in Europe, together with
years of experience of different customer needs and abilities,
an example is given as to how a classic but developing service
can survive as a reliable and useful information partner within
a web environment, and can also contribute to future developments
of the global business environment.
Ferdinand
Rudolf
European
Patent Office, Vienna, Austria
Patent information:
the EPO perspective
Good value CD-ROM
productsa and free internet services such as esp@cenet have changed
patent information for ever. In the past ten years, patent information
has become cheap, accessible and easy to use. Our task is no longer
to make patent information available, but to nurture awareness
of the information, to change the way people tackle problems,
to help them find their way through the information overflow,
to turn information into knowledge. the EPO sees its role
in this as being to ensure that it reacts to the needs of users.
It approach, in co-operation with its member states, is to: i)
offer reliable data in a timely fashion; ii) provide products,
especially on the internet, that are friendly and easy to use;
iii) offer competent and high quality advice.
Peter
Kallas
BASF, Ludwigshafen,
Germany
CD-ROM archives: will
the internet consign them to history?
The World Wide Web has
provided millions of people with more or less instantaneous access
to a huge number of patent specifications. They are generally
either free of charge or require only relatively small fees. After
the launch of big projects such as Espacenet by the EPO and the
initiatives of other offices, many people are convinced that this
is the end of the traditional CD-ROM archive such as those in
use in big companies. But is this really the beginning of the
end?
This presentation analyses
the demands of a big company which currently relies on CD-ROM
archives for printing specifications on demand and offering them
via internal networks. It looks at just how the internet developments
will upset the status quo.
Peter
Rusch
California,
USA
Transitions for value-added
patent information services
As more patent issuing
authorities provide patent information service at low or no cost,
the traditional distribution chains and service levels are subject
to greater scrutiny among customers. Aided by web technology,
these new services challenge the established services and compete
for customers. This paper reviews the role of value-added services
and their place in the emerging markets created by the new entrants.
The perspective of producers, providers and customers is examined.
Victoria
Veach
3M Information
Services, Minnesota, USA
Changes in patent searching
brought about by the availability of web-based resources
The coming of the World
Wide Web and 'free' information in the patent area have resulted
in growing changes in the traditional world of patent searching.
This presentation examines ongoing changes in the 3M technical
community in the search habits of both patent information professionals,
and end-user searching as 3M searchers adapt to the world of the
internet and the intranet.
Michael
Hehenberger,
IBM Global
Business Intelligence Solutions, NY, USA
Text Mining of chemical
and patent literature
Text mining tools, including
Advanced Search, Linguistic Analysis / Feature Extraction and
Clustering can be used to navigate databases of chemical and patent
literature. Web crawlers can add up-to-the-minute information
to competitive databases. Java visualisation is helping users
equipped with standard Web browsers to make sense of search results,
classify documents into topics characterised by lists of keywords,
and relationships between topics.
It is shown that text mining
adds value beyond conventional search engines, and that it enables
scientific end users to explore chemical and patent literature
with less assistance from information specialists. In addition,
text mining is helping patent portfolio managers and competitive
intelligence experts to provide crucial decision support to corporate
management. The presentation covers elements of text mining technology
and its application to chemical and patent literature, along with
brief demonstrations.
Ulrich
Kämper
Wind GmbH,
Cologne, Germany
Chemical business information,
competition from the internet
Business and financial
information in chemistry has changed since the internet allows
the sale or free donation of information that could not earn enough
money to be offered via traditional database hosts. Small, emerging
and risky businesses, especially in biotechnology and diagnostics,
are focussing the attention of investors, stockbrokers and bankers,
but well known business magazines will hesitate to write about
these companies, and business databases will not put any new newsletter
on their source list. From the internet a wide variety of additional
information is available, more or less reliable, but better than
no information as in the past. The reader is increasingly involved
in critical evaluation of the information published on the internet,
but this additional workload is counter-weighed by the gain in
scientific, technical and business information. From the macroeconomic
point of view, information on emerging companies is desirable
to provide venture capital according to their growth speed independent
from the rather slow and tactical activities of international
chemical companies. This paper presents examples of chemical business
information that could not be found in commercial chemical and
business databases.
Gabriele
Ilchmann
Beilstein
Informationssyteme, Frankfurt, Germany
A full-scale data compilation
of pharmacology, toxicology and ecological chemistry into a single
bibliographical source: CrossFire EcoPharm
Concepts and practical
approach to abstract information on pharmacology, toxicology and
ecological chemistry of organic substances from the original primary
literature into a factual database are introduced. The compound
data, be they pure substances, mixtures, polymers or bio-molecules,
are abstracted into a new hierarchical data structure, which is
described in detail. The combination of these valuable data with
other physicochemical properties such as Henry constant, n-octanol/water-partition
coefficient or solubility, as well as chemical reactions, permits
researchers to assess the ecological impact of xenobiotics on
various ecosystems or to evaluate the therapeutic potential of
new drug candidates.
Valeri
Kulkov
Advanced
Chemistry Development, Ontario, Canada
A corporate solution
for structure-based chromatography and spectroscopy management
One of the major directives
of corporate Research and Development management is to track and
manage data and information. Analytical laboratories provide measurements,
information and analytical support and solve problems. During
this effort they process many samples and requests and produce
large numbers of test results and reports. In an attempt to increase
efficiency, many analytical laboratories have computerised their
logbooks, focussing their efforts on tracking jobs, samples, tests
and results. With this information accessible via a computer,
it has become possible to provide functions such as management
of test results, archival, calculation, comparison to specifications
and control charting. The data and knowledge generated as a result
of chemistry and pharmaceutical R & D efforts today can commonly
be defined as laboratory notebook. However, these efforts have
not given rise to the expected breakthroughs in a lot of areas.
Since chemists have to deal not only with text or numbers but
also chemical structures, spectra, diagrams and charts and other
types of data that require visual user interface, the number of
computer systems capable of handling chemical objects remains
limited. In particular, there is a distinct need for both spectroscopic
and chromatographic database management systems since any analytical
spectroscopy laboratory can generate in a single year thousands
of spectra or chromatograms which are associated with individual
samples, and commonly with known or suggested molecular structures.
Corporate-wide solutions
to address the need for a spectroscopy and chromatography management
system with linking to molecular structures and individual sample
details are presented. These solutions unite chemical structure
and spectra databases with chemical property predictions and spectra
processing utilities into one, cross-platform, client-independent
computer information system taking full advantage of modern global
networking technology.
Wolf-Dietrich
Ihlenfeldt
University
of Erlangen-Nürnberg, Germany
The NCI database: bringing
the largest open structure database to the www
Within the Developmental
Therapeutics Program (DTP) of the US National Cancer Institute
(NCI) structure and cell-plate test data on anti-tumour and AIDS
anti-viral activity of submitted compounds have been collected
for many years. This effort ultimately resulted in the largest
public structure data collection with more than 250,000 open non-confidential
compounds plus numerous associated data items such as names, CAS
numbers and activity screening result tables. This data collection
has been available to the public for some time. Basic substructure
and other search capabilities on selected subsets of the open
NCI database have recently been made available by DTP on their
Web site. In collaboration with the Laboratory of Medicinal Chemistry,
NCI, we have used the publicly available bulk structure and data
files to implement a state-of-the-art WWW interface to these data,
with advanced search capabilities by numerous criteria in the
entire open collection.
Several design goals of
this project set it apart from other data sources on the Web.
First, due to its public nature, this database was intended to
be usable on any platform, with any browser. This constraint dictated
the exclusive use of portable technology for structure input and
output, allowing the use of PC-centric software only for auxiliary
features. Instead, Java applets for structure input, dynamically
generated GIF images and report tables, VRML for 3D-visualisation,
import facilities for structure files or SMWES strings and other
generally usable techniques for structure handling form the core
of the service, which provides access to all the usual chemical
query functionality such as full-structure and substructure search
plus queries on associated textual and numerical data.
Second, this database was
designed to link to as may other public data repositories and
public services as possible. A wide variety of approaches was
taken depending on the nature of the external information source.
These include: inclusion of computed data in the local database
files (computed logP), full cross-referencing with other databases
(liquid crystal properties), transfer of structure data to other
WWW form-based databases (Cambridge Soft ChemFinder), and linking
to external WWW-based computational services (IR simulation).
We also provide extensive export functionality for molecules and
datasets in many different formats, making this data service one
of the few sources for structures usable for further examination
with computational chemistry methods.
We have found that it is
indeed possible to provide a fully functional chemical database
service usable in a cross-platform manner on all major Web browsers,
although sometimes unexpected obstacles and problems were encountered.
We report on the implementation and user experience aspects of
this project.
David
Lide
CRC Handbook,
Maryland, USA
Chemical property data
sources in the internet era
Delivery of chemical property
data by electronic means has tended to lag that of abstracts,
patents and full texts of journal articles. However, the last
two or three years have seen a rapid growth in the availability
of numerical data on physical and chemical properties on the internet
and in the form of optical and magnetic media. This talk surveys
the major data sources currently available through electronic
channels. The design requirements for converting widely used print
handbooks to electronic form are discussed. The problem of quality
control, which is becoming increasingly important as web sites
with chemical data proliferate, is addressed, as well as issues
related to protection of intellectual property.
Stephen
E Arnold
AIT, Kentucky,
USA
The software revolution:
a state of the art survey
New developments in the
software universe include XML and cXML that increase search and
display options, the movement into the mainstream of computationally
intensive technologies, and the increased momentum generated by
visualisation software. Current trends considered in this presentation
include developments in intelligent software that make information
retrieval easier, the growth of clustering and tiered services,
the advent of meaty abstracts, synopses and extracts, and the
development of portals.
This presentation concludes
that pure search engines will continue to have financial challenges;
natural language searching and clustering will become commonplace;
embedded searching will be found in many application areas; visualisation
developments will continue to be important, but 3D representation
will be the major development area; and software currently under
development will move to the mass market rapidly.
Jane
Whittall
SmithKline
Beecham Pharmaceuticals R&D, Harlow, England
Leveraging knowledge
across continents
Intranet, Lotus Notes and
other technologies have enabled knowledge management programmes,
but have not guaranteed their success. A key factor for success
is to change behaviour from 'knowledge is power' to a knowledge-sharing
behaviour. This paper around knowledge management discusses the
elements that ensure a successful knowledge management programme,
and how intranet and other technologies have enabled this success.
Examples are drawn from practical experience within a multinational
pharmaceutical company, highlighting the issues encountered and
benefits gained from a successful knowledge management programme.
Daniel
Champlon, C Corlay, E Nocera
Institut
Français du Pétrole, Rueil-Malmaison, France
Knowledge management
implementation in an R&D centre
The IFP is a research centre
with a staff of 1,800 working within the domain of petroleum science;
the centre is evolving towards an environment that needs to be
more and more competitive.
The knowledge management
project that is being implemented aims to meet the end of millennium
challenges faced by a quasi-public body: the growth of in-house
resources and the rapid integration of new partners. This project
integrates information exchange, the existing knowledge base,
and the exploitation of in-house expertise, and will evolve towards
identification and capitalisation of both in-house knowledge and
expertise. This presentation details results obtained to-date
from this innovative project.
Louis
Culot
CambridgeSoft
Corporation, Massachusetts, USA
The Open Chemistry project:
An electronic paradigm for community publishing and peer review
on the WWW
The ChemFinder web site
(http://www.chemfinder.com) is the most popular public chemistry
resource on the Internet. It currently enjoys over 80,000 unique
visitors (eyeballs) each month, and over 500,000 searches. The
site is used extensively by researchers, university professors,
students, safety and regulatory officers and a variety of allied-chemistry
professionals and researchers.
Two of the site's key features
are its list of physical properties and its hyperlinks to other
information on the WWW. To date, accumulation and presentation
of these data have been a manual process, handled entirely by
CambridgeSoft personnel. Because of this, there is certainly information
available on the WWW which has not been indexed, and there is
little record of the origin of physical property data. In addition,
our current data model limits our ability to present spectra,
or effectively handle substances which are mixtures or natural
products, or record syntheses. The inability to track the source
of physical property information has been a criticism of ChemFinder,
as has the frequency and quality of the hyperlink information.
The Open Chemisty Project
will address these shortcomings by turning ChemFinder into a chemical-community
project. It is highly analogous to Netscape's Open Directory project,
which is a horizontal index that competes with Yahoo!. Netscape
has been successful in its approach, and the Open Directory is
currently the engine behind Netcenter and Lycos. The Open Chemistry
Project involves scientists from all reaches of industry and academia,
and combines resources and talents to produce a well checked and
comprehensive resource. We shall present the project, its mission
and discuss technical, business and social challenges encountered
throughout the project life.
Irmgard
Fischli
Novartis
Pharma, Basel, Switzerland
Re-creating corporate
information departments following mergers and changes: some lessons
and opportunities for the new century
Novartis was created by
the merger of Ciba and Sandoz. With its focus on healthcare, agribusiness
and consumer health, Novartis is committed to improving
health and well-being through innovative products and services.
For any such organisation, reliable information services are key.
The merger challenged this issue of re-creating corporate information
departments aiming towards delivering integrated information to
its employees on a global basis.