Once upon a time any advanced level technical or navigational calculation relied on the access to numerical tables of special functions. Those were calculated by hand and then, after some verifications, they were typeset by hand and finally printed, after some more verifications. This was a process where errors easily could be introduced in any of the steps. One way of minimising the errors was to fully mechanise the process of calculating and printing tables in order to remove the human interaction in the process. Johann Müller, a German engineer, was the first to propose this idea in 1784, and the machine was designed based on the method of differences but it was never built. Charles Babbage in England invented an engine (note that it is not a machine from this point) in 1821 which was based on the same ideas. This engine was under construction from 1823 to 1833 but it was never completed. Instead Georg and Edvard Scheutz, father and son, managed in 1843 to complete the first functioning prototype engine for table calculations. However, it turned out that there was not a big need for these kind of calculating engines, in particular when considering their immense cost, so only two regular Scheutz engines were built after the first prototype. Müller and the Scheutz family are mostly forgotten nowadays while Babbage has reached an almost mythical status, mostly thanks to his analytical engine, which he also did not complete, that is considered to be the first “modern” computer.
The book Glory and Failure by Michael Lindgren has the subtitle The difference engines of Johann Müller, Charles Babbage, and Georg and Edvard Scheutz and that describes its contents perfectly. It contains descriptions of the mathematical principles behind the machines, the tecnology and the history of the machines and their inventors. Lindgren has also tried to explain why Babbage failed in his very costly attempt to produce a complete difference engine and why the Scheutz engine, while significantly less expensive than the Babbage engine would have been, failed commercially.
Glory and Failure is an academic text, so it is not always an easy book to read. On the other hand, it also means that all facts are properly sourced and referenced which simplifies digging deeper into the subject. The last aspect is important because the book is actually very interesting to read, and it could inspire further studies. It was also interesting for me to read about how the Babbage and Scheutz engines were financed and supported because there are many similarities to how the financing of research infrastructure, and the corresponding political manouvering, works today. Having the right connections means everything!
The book was salvaged from a down-sizing of a library where a large number of books were heading for destruction. When a book is destroyed, some potential knowledge is also destroyed in the process, and the humanity in the future will be a little bit poorer. Read books, don’t burn them!