The problem is less with the Babbage Engines than with measurements of physical properties to go into (and come out of) the Babbage Engines. Much of what modern highschool science students take for granted -- e.g., cheap and replaceable, well-calibrated pipettes -- simply does not exist in 1884, even for the "professionals."
As a specific example, consider the Michelson-Morley experiment (Professor Fowler's explanation includes an animation of the actual experiment setup (http://galileoandeinstein.physics.virginia.edu/lectures/michelson.html)). Every single piece of that equipment was custom-built. In turn, that means that there was a significant possibility of mechanical, as opposed to scientific, failure... in the data-gathering process or, if the Engine was turned into a supergiant magical "loom", as others have proposed, the data-exploitation process. And yes, this should have giant, neon "Plot Complication!" signs flashing by now.
Then, too, nineteenth-century mathematics and scientific practice did not consider scale as a problem. At one end, we've got the problem of not having a model of the atom, let alone of molecules and crystals (which are a lot more interesting than atoms; consider, for example, trying to figure out why the emerald cut can be made in diamonds and carborundum-based stones, but not in, say, water ice or onyx). At the other end, we've got problems with impurities, side reactions, etc. (grinding up enough peach pits to make commercial quantities of "cancer cures" also releases quantities of cyanide compounds large enough to create hazards). Even without going to that extent, "temperature" outside of ordinary human experience was, at best, an unreliable matter: Just take a look at the construction of the initial Bessemer furnaces and their complete lack of verifiable temperature control!
Too, the differences between stochastic and statistical results had not yet been assimilated. One of the best places to exploit this is to look at the pure-classical view of thermodynamics that prevailed in 1878; Fermi's little book of lectures -- which includes all the hairy math, but also includes the historical context that you're interested in -- leads to some truly profound Plot Complications.
In short, I think the problem is less with the Babbage Engines than with the reliability of the data in and data out. Perhaps that's where the magic comes in...
![[personal profile]](https://www.dreamwidth.org/img/silk/identity/user.png)
yati
Expanding on Drydem's comment from a bioinorganic chemist's perspective:
Date: 2010-06-04 03:37 pm (UTC)As a specific example, consider the Michelson-Morley experiment (Professor Fowler's explanation includes an animation of the actual experiment setup (http://galileoandeinstein.physics.virginia.edu/lectures/michelson.html)). Every single piece of that equipment was custom-built. In turn, that means that there was a significant possibility of mechanical, as opposed to scientific, failure... in the data-gathering process or, if the Engine was turned into a supergiant magical "loom", as others have proposed, the data-exploitation process. And yes, this should have giant, neon "Plot Complication!" signs flashing by now.
Then, too, nineteenth-century mathematics and scientific practice did not consider scale as a problem. At one end, we've got the problem of not having a model of the atom, let alone of molecules and crystals (which are a lot more interesting than atoms; consider, for example, trying to figure out why the emerald cut can be made in diamonds and carborundum-based stones, but not in, say, water ice or onyx). At the other end, we've got problems with impurities, side reactions, etc. (grinding up enough peach pits to make commercial quantities of "cancer cures" also releases quantities of cyanide compounds large enough to create hazards). Even without going to that extent, "temperature" outside of ordinary human experience was, at best, an unreliable matter: Just take a look at the construction of the initial Bessemer furnaces and their complete lack of verifiable temperature control!
Too, the differences between stochastic and statistical results had not yet been assimilated. One of the best places to exploit this is to look at the pure-classical view of thermodynamics that prevailed in 1878; Fermi's little book of lectures -- which includes all the hairy math, but also includes the historical context that you're interested in -- leads to some truly profound Plot Complications.
In short, I think the problem is less with the Babbage Engines than with the reliability of the data in and data out. Perhaps that's where the magic comes in...