Computer Handling of Chemical Structure Information

Ah yes, Computer Handling of Chemical Structure Information (1971) ? a title so magnificently unapologetic that it manages to sound both wildly futuristic and crushingly administrative at the same time. Written by M. F. Lynch, J. M. Harrison, W. G. Town and J. E. Ash, and published by the sternly dependable alliance of MacDonald and American Elsevier, this is a book from the glorious early age of computing, when the idea of getting a machine to understand chemistry still had the thrilling whiff of science fiction about it. Today, of course, we take it for granted that a computer can store, search, sort and juggle chemical structures while also distracting us with emails and weather alerts. But in 1971 this was serious frontier work. The challenge was not merely one of chemistry, nor merely one of computing, but of persuading these two extremely complicated realms to sit down together and behave themselves. How, after all, do you teach a machine to cope with molecules, rings, chains, bonds, substitutions and all the rest of the gloriously fiddly architecture of chemistry? It is one thing for a chemist to glance at a structural formula and nod wisely. It is quite another to reduce that same formula to something a large humming box can process without existential complaint. That is the territory this book inhabits: the fascinating, slightly forbidding borderland where punch-card logic meets molecular elegance. It is concerned with systems of representation, indexing, retrieval and classification ? in other words, the hidden bureaucracy of scientific knowledge. Not the dramatic bubbling flask end of chemistry, but the equally vital question of how one actually organises chemical information so that it can be found again before the heat death of the universe. There is something wonderfully ambitious about a book like this. It comes from a period when computers were still enormous, temperamental, and far less interested in helping ordinary people than in quietly revolutionising specialist disciplines from the inside. Here we see that revolution in progress. The authors are grappling with problems that now underpin huge areas of cheminformatics, database science and pharmaceutical research, but they do so in the measured, deliberate tone of people who know they are building part of the future with graphs, coding schemes and a deep faith in orderly systems. And what a future it promised. This is a world where the proper encoding of a benzene ring is not a niche concern but the sort of thing upon which the whole modern edifice of searchable chemical knowledge may eventually depend. It is, in its own dry and noble way, a book about translation: taking the exquisite visual language of chemistry and converting it into forms that computers can store, compare and retrieve. It may not sound romantic, but there is a peculiar beauty in the effort. Humanity had spent centuries discovering substances, naming compounds and drawing structures, and then along came computing to say: splendid, now please put all of that into a system. Naturally, being a technical volume from 1971, it does not exactly bounce along like a thriller. It is dense, precise and entirely untroubled by the modern need to be ?accessible? It assumes, with admirable confidence, that the reader has turned up because they genuinely want to know about machine methods for handling chemical structure information, and not because they were hoping for a light diversion before bed. In that sense, it is refreshingly honest. No gimmicks, no simplification, no brightly coloured sidebars pretending data structures are fun for the whole family. Just proper intellectual machinery, laid out with seriousness and care. Still, that is part of the charm. For collectors of vintage scientific and computing texts, this is exactly the sort of book that radiates period authority. It belongs to that lovely moment when computers were beginning to colonise every corner of specialist knowledge, and each ne. …