The workshop - Building and Owning Biotechnology Databases was
held at the Golden Tulip Hotel, Purmerend, The Netherlands on 22-23 September
1998. It was organised by The Biotechnology Information Strategic Forum,
with support from DGXII of the Commission of the European Communities.
Adding value, and delivering it -- Stella G. Dextre Clarke, Information Consultant
If the main financial support of a database is to come from users who pay, rather than from a funding agency, then the service must deliver value for money. Hence the design of a successful database requires knowing what the users want, and knowing it in all the varieties and flavours that will be demanded. Typically, different users want different things: some will want comprehensive coverage of the field while others prefer a selective service; different users have varying degrees of capability and may need different types of interface, perhaps using different languages or different distribution media; and even the same user is quite likely to want different things on different days. So a database which will stand the test of time must be structured with the flexibility to be re- packaged in a variety of ways, and to evolve as the user environment changes. At all times, the database must deliver quality, which is often defined as meeting the requirements of customers within time limits and at costs which are acceptable to them.
Among the community of online database users, quality is usually assessed under the following headings:
A quality service depends not just on the production system, but on the whole delivery chain. This extends from acquisition of raw data through implementing robust production systems and back-up systems, marketing the product/service, distributing it through all the required media, pricing it appropriately, implementing a manageable charging system, and providing prompt, friendly user support. Customers make just as much fuss about inaccurate invoices as they do about unreliable gene sequence data.
To ensure the overall service meets user requirements, the business-like database producer puts in place a quality management system. This means specifying standards for the level of quality required in each process and aspect of the service, monitoring and measuring compliance with the standards, and correcting every deviation.
This talk has dwelt so far on the need to be competitive in the market-place, but the collaborative aspects of database production should not be forgotten. The producer rarely works in a vacuum. There may be stakeholders in the shape of an employer, a funding agency, a board of academics or even the bank manager. Overseas sales support may need a network of trained agents around the world. Furthermore, the business may be reliant on receiving or extracting data from some other source before using it to add to the database. The source publisher might not be at all pleased to find his data being sold on or even given away free to third parties! Agreements with all parties are vital. Even when the prime objective of the venture is research rather than money- making, future frustration can be avoided by putting the right legal framework in place from the start. Copyright and ownership issues are so much easier to resolve before a suspected infringement has soured the atmosphere.
Would-be database producers may be excused for worrying
about the cost of putting in place all these agreements and arrangements
for quality systems. Excused but not exonerated. In today’s increasingly
litigious society, the cost of being sued for copyright infringement, or
for professional negligence in the supply of services, can be very much
higher. And disappointed customers who walk away from the product will
do little to defray the operating costs of the venture. On the other hand,
the rewards of designing and operating a successful database can be great.
To make a success of the enterprise, start with business-like planning
and then just concentrate on lining up those happy customers!