Catching a benign virus through mixed browsing

Publication of the Encyclopedia of Virology three years ago was a bold move since the book (The THES January 20, 1995) was pitched squarely between an introductory text for undergraduates and a standard reference work for research scientists. But the high quality of the editing and production made it an up-to-date and valuable resource for students and teachers of virology. What, if any, are the extra merits of a CD-Rom version of the book?

It is a truth acknowledged by all but the most hard-bitten computer hacks that it is both easier and more pleasant to browse through printed pages than through text on a computer screen. During the infancy of CD-Rom books, there has been a wide divergence between those that are simply electronic versions of the printed book, and those that make proper use of the information-processing power of a computer.

The usefulness of many reference books is limited by the ease of finding not only large subject headings but minor references in the text. A computer cannot do the creative thinking for you (at least, not yet), but it can do the boring, repetitive tasks of searching, listing and cross-referencing much faster than you could ever do with a paper book.

The ideal sort of reference book to publish on a CD-Rom is a general encyclopedia, as this type of work will make the greatest use of the ability of a computer to search for common features among disparate subjects. The CD-Rom format is less well-suited to a one-subject book such as this Encyclopedia of Virology. There is necessarily a large element of repetitive structure in the book: that is, under each virus heading there is a more or less predictable list of subheadings. This predictability partly obviates the need for a sophisticated search mechanism such as that existing on this CD-Rom.

However, if the power of a reference book is limited by its index, then good software should always be able to increase this power. Fortunately, Academic Press have used the excellent EBT Dynatext software, which makes it simple to search the text in a surprisingly sophisticated way.

A particular virtue is the well-written reader's guide, which gives comprehensible instructions on the use of the software - unlike the usual manual which normally weighs 30 times as much as the CD and appears to be written in hexadecimal code.

Three types of search are possible: proximity, boolean and context searches. Suppose you wish to search for text on (viral) polymerase mutations. You can specify in the search "polymerase mutations" or more complex criteria such as "polymerase" within four words before "mutations", or "polymerase" within six words of "mutations".

Boolean searches use the standard boolean operators "and", "or" and "not", in a way that all readers will be familiar with. The context mode allows you to confine your search to specific sections of the encyclopedia.

The other standard features of computer searches, such as wild cards, are also here. Simple instructions are given under the headings "advanced searching" and "advanced browsing" - which sounds more like the activity of giraffes than of virologists. Frequent icons and text hyperlinks can be used to lead to tables and high-quality colour or monochrome pictures.

The tables are especially useful; but the graphics require a fast processor and a good deal of random access memory in the computer. Any of the contents of the book may be printed, but again, this is likely to be limited by the quality and speed of your computer equipment. And while several views of the book can be opened simultaneously, readers will not have a large enough screen to make this practical. A facility called a "journal" allows you to record your "reading path" through the book - that is, the sequence of sections that you read. I imagine that single users of the Encyclopedia are unlikely to make use of this, but it could be invaluable for preparing reading material for students, in relation to a special topic.

If you use the book intensely or repetitively for any purpose, such as for research or for preparing lectures, and if you are adept with the keyboard as well as the mouse, you will gain from a very civilised feature of CD-Roms - the ability to add your own bookmarks and annotations. Of course, you can add these to a book by writing in the margin, but the CD-Rom allows you to make your notes in a more socially acceptable way.

Lastly, the very simple, two-click method of making hyperlinks, such as your own cross-references, will again be handy for teaching and revision purposes.

Putting this book on to CD-Rom will not increase the scope of the readership. But it will make it very much easier for students and teachers of virology to find and assemble the material they need. What I'd like now is the standard texts in virology and immunology in a similar format, and with the same software.

Charles Bangham is professor of immunology, Imperial College School of Medicine, London.