The relationship between the scientist for whom the theory of tides has always offered a fertile field for academic research, and the practical man whose operational need for tide tables has seldom coincided with the desire to seek a full understanding of tidal phenomena, have played a fundamental role in the development of accurate tidal predictions.
Until the last decade, without the sympathetic collaboration of the scientist, the intelligent mariner was rarely able to satisfy the more demanding of his own requirements for tidal forecasts; academic dilettantism was (and still is) of little use to him. In the United Kingdom, the earliest recorded efforts, those of John Wallingford, Abbott of St Albans in the thirteenth century, were only intermittently improved, as for example by John Flamstead during the seventeenth century, until the full potential of Isaac Newton's work on gravitational theory was utilised systematically by Sir Joshua Lubbock, Lord Kelvin and Sir George Darwin in the second half of the nineteenth century. The extension and refinement of the framework of applied theory so formed was then taken over by A. T. Doodson until the dawn of the digital computation revolution just over a decade ago.
The nature, scope and quality of tidal predictions over the centuries have seen many changes in response to demands for improvement, which are as pressing today as at any time in the past. Perhaps the most remarkable feature of this trend has been the way in which the development of techniques for observing tidal variations—the raw material from which predictions are derived—has failed to keep pace with advances in theory, computational techniques and practical usage.