Please Book Early

Appeared in Volume 9/1, February 1996

'We live in interesting times' is a rallying call of people who have little to do with computers. Two such categories are politicians and business identities, who employ administrators, secretaries, and boffins to mine the information coal face. For the rest of us, 'interesting' means an ever increasing pace of change.

I was reminded of this when a friend bought an important biological sciences text, printed in the 1930's. I could hardly believe that something that old could still be relevant. Of course, some work on the foundations of computer science dates back this far, but the average computer scientist thinks the 1970's are the Dark Ages. A sure sign of this is the way that publishers advertise books dating from those halcyon days as 'classics', or as valuable sources of historical information.

In fact, the computing sections of most book shops are loaded up with 'How to' manuals on riveting topics such as Excel 6, Word 6, and MacCheeseMaker 6.5. However, the relentless upgrade cycle almost immediately force these out the door, into the dumpster, as the latest 'How to be a dummy' in Excel 7, Word 7, and MacCheeseMaker 6.6 come storming in, taking no hostages.

It is instructive to compare an earlier version of a 'How to' book with a later one. Aside from the changed digit, it often appears that only the screen dumps and the amusing artwork have been altered.

A frightening new development is the way that books are becoming obsolete before they even reach the shelves. I experienced this recently when looking for texts on the WWW programming language Java. Sun Microsystems did a naughty thing and changed most of the Java I/O mechanisms just as a slew of Java books hit the stands.

These books were not affected because the author's indulged themselves in years of agonising ruminations about pedagogy. On the contrary, most of the material looks like it was directly dumped form the Java WWW pages (with the addition of spelling mistakes and grammatically challenged explanatory expansions). The books were made obsolete because the publication process (which takes a few months at best) is too slow to keep up with rate of change of the technical content.

I wonder if this signals the beginning of the end for computing books (in the traditional paper and ink sense)? At least for Java, the WWW is a timely alternative.

Andrew Davison