Monday 17 October 2005

Over the last decade, the Web Revolution has obviously transformed scientific publishing. Companies and professional societies have invested heavily and enthusiastically in technology to dramatically expand the distribution of scientific information and increase the speed of its dissemination. These changes have benefited every stakeholder in scholarly communication. Yet the Web Revolution and other technological developments have created numerous challenges for publishers and those who purchase and use their products. Controversies and conflicts abound over such issues as funding, pricing, cost, access, and functionality. The questions are many: Where will the money come from to fund this publishing system? Who should pay for it? What are fair and reasonable ways to charge for scientific publications? How can costs be contained as the amount of scientific literature grows rapidly? Who should have access to this information and under what terms and conditions? How can and should product quality be measured? How are competitors responding to these challenges? When are competitors also collaborators? How will new players such as Google affect the behaviour of users and hence the publishing system? What new capabilities and tools do authors and readers want and demand, and are they affordable? What is the proper role of governments in funding and regulating scientific information? Who is responsible for preserving the scientific record for generations to come? A tour of today's scientific publishing landscape will explore-and possibly elucidate-these and other complex issues.
 

The age of the internet has brought with it easy and affordable technology for the creation, distribution and access to scientific papers; commercial publishers, learned societies and libraries are having to re-define their roles in a system that has been built up over hundreds of years. Taxpayers want value for money, and scientists want to disseminate their research results whilst enhancing their reputation. In the life sciences, Open Access has become a popular alternative to subscription-based publishing and is giving freedom to information for the reader and larger audience and impact for authors. In chemistry change is slow; no new business models have taken hold; publishers are still trying to maintain the status quo. With the launch of the Beilstein Journal of Organic Chemistry, the Beilstein-Institut has taken a major step into moving chemistry into this new environment. The BJOC will make high quality original research information available online to a global audience at no charge. The current Open Access discussion is not only bringing transparency into the financial aspects of scholarly publishing but also changing forever how the scientific society views and values information and information services. The chemistry publishing establishment is being called on to define the “value” in “value added” and adopt fair solutions regarding intellectual property rights, libraries are being called on to play an active role in electronic archiving and scientists to put quality before quantity when publishing. After philosophy there must be action.
 

Scholarly information and communication are increasingly dominated by the Internet, with well-known challenges for established publishers and information providers. FIZ Karlsruhe is working with its STN partner CAS on strengthening STN's competitiveness; a step towards these goals was the recent release of STN AnaVist, a jointly developed interactive analysis and visualisation software offering insights into trends in scientific and patent information.

The Internet's influence on scientists' work goes beyond scientific publishing and could hardly have been imagined a decade ago. In the near future, ways of scientific working will change considerably. Therefore, FIZ Karlsruhe is co-operating with the Max Planck Society (MPS) in the project eSciDoc, aimed at creating an integrated information, communication and publishing platform for multi-disciplinary scientific cooperation within the MPS. After its final release, eSciDoc will be available to other national and international scientific societies as well as to researchers in industry. eSciDoc is part of the German eScience initiative and sponsored by the German Federal Ministry of Education and Research.

Information is most valuable when used in context. Intellectual property information, however, is often used independently of other equally important sources, sometimes resulting in expensive legal disputes. Earlier this year LexisNexis, the global legal publishing arm of Reed Elsevier, acquired Univentio Information Services a company in the production of national and international patent databases. LexisNexis, known principally for its legal, news and business information, is releasing a suite of integrated information solutions that will incorporate Univentio’s full text, bibliographic, translations and image patent archives with LexisNexis content. This presentation outlines LexisNexis’ vision of how this integration will occur and what benefits it will bring to various information user communities, as well as looking at the scope for integrating STM and other content from other Reed Elsevier divisions.

A number of factors have greatly raised the importance of data and knowledge management issues within The Dow Chemical Company Research & Development community. In particular, the acceleration of the adoption of High Throughput Research (HTR) methods has greatly increased the need to manage large quantities of data, share those data throughout a global R&D community, and preserve the interpretability of those data as the membership of the R&D community changes. However, it is often difficult to explain the science and art of informatics to non-practitioners, and it is often difficult for them to understand the complexity of informatics development and deployment efforts. This presentation discusses these challenges and some of the approaches that Dow's Research Computing group is taking to address them.

Workflow is not a new concept. It has been widely used in business process management (BPM) since the middle 90s. However, recently, workflow has become a key technology for discovery informatics in the pharmaceutical industry. Workflow technology enables scientists to dynamically construct their own research protocols for scientific analytics and decision making by connecting together various information resources and software applications in an intuitive manner. With the first generation workflow used in computational chemistry, known as pipelining, scientists enjoyed the capacity to flow data through various application functions without writing complex scripts. However, this is just the beginning. Workflow is not just a visual way of writing scripts for a vertical application. Workflow can form the foundation of an enterprise informatics environment enabling organisations to optimise their important data-driven processes and deliver information, expertise and derived knowledge effectively and quickly to decision-makers throughout the organisation from scientist to the board. With the wide acceptance of web services for application integration, workflow is the integration technology where applications, published uniformly as services, are composed via a workflow as an integrated service; moreover, workflows provide the means to represent the “best practice” of research. With the new technology of workflow warehousing, a company now has a way to manage its process knowledge as valuable intellectual property. Using two case studies, we will present how Open Discovery Workflow technology is used to create a pharma service-based informatics infrastructure and a public knowledge grid.

Recent surveys of literature use by both academic and non-academic scientists indicate that the average scientist spends between 5% and 10% of the week reading the scientific literature. Although the amount of scientific knowledge doubles every 15 to 17 years, with an estimated 80,000 to 100,000 journals now published worldwide, scientists today sample only a small subset. The average scientist reads from only 18 journals in one year, and most of those journals are read sparsely (few articles are read). Scientists have no more time available today than they did a century ago for literature study, yet the amount of information has increased manifold. One potential ameliorative is to bring the literature, and cross-database literature mining, closer to the practising scientist at the bench. Pipelining is already an established paradigm for chemical and bioinformatics research, and is becoming ever more recognised as an essential tool in the scientist's toolbox. The presentation discusses an integrated package of text analytic and search capabilities that enable scientists to readily incorporate literature evidence directly within their traditional scientific protocols. A number of protocols are illustrated, spanning AIDS, Arthritis, and fuel cell research, to illustrate the bench-level interaction between traditional scientific activity and the literature.

Collecting and analysing bioactivity data of compounds published in the literature is a frequent and cumbersome operation for Life Science researchers. Literature and patent databases together with integrated tools help to make collecting and evaluating this information much easier by providing detailed substance dossiers.

Companies need to use information technology to fully exploit their data. All these data must be made available, in one place, on a need-to-know basis across the organisation and all along the drug development process from Discovery to Clinical Trials and possibly beyond. One of the most critical factors is the effective delivery of information to people who must be able to browse, explore and search across it without requiring advanced IT skills.

Current integration solutions tend to be restricted in scope, typically to general areas such as bioinformatics or chemoinformatics, and the search facilities are either limited to forms and text searches or are technically complicated, typically requiring knowledge of a 'query language' and database schemas. Simpler alternatives restrict search across different databases and domains rather than enabling ad-hoc cross-domain queries.

One solution to this accessibility problem is to present the integrated data in a collated manner which allows users to browse and explore them and also search them in a scientific context without the need for IT skills. Additionally, the solution should be maintainable by 'in-house' administrators rather than requiring expensive consultancy. This presentation examines the background to this problem, investigates the requirements for effective exploitation of corporate data, and presents a solution strategy.