All the morning at home, and Mr. Berkenshaw (whom I have not seen a great while, came to see me), who staid with me a great while talking of musique, and I am resolved to begin to learn of him to compose, and to begin to-morrow, he giving of me so great hopes that I shall soon do it. Before twelve o’clock comes, by appointment, Mr. Peter and the Dean, and Collonel Honiwood, brothers, to dine with me; but so soon that I was troubled at it. But, however, I entertained them with talk and oyster till one o’clock, and then we sat down to dinner, not staying for my uncle and aunt Wight, at which I was troubled, but they came by and by, and so we dined very merry, at least I seemed so, but the dinner does not please me, and less the Dean and Collonel, whom I found to be pitiful sorry gentlemen, though good-natured, but Mr. Peter above them both, who after dinner did show us the experiment (which I had heard talk of) of the chymicall glasses, which break all to dust by breaking off a little small end; which is a great mystery to me. They being gone, my aunt Wight and my wife and I to cards, she teaching of us how to play at gleeke, which is a pretty game; but I have not my head so free as to be troubled with it. By and by comes my uncle Wight back, and so to supper and talk, and then again to cards, when my wife and I beat them two games and they us one, and so good night and to bed.
A curious entry on the blackboard and not what you might expect in a physics class. It turns out this wasn't a standard class - an applied physics class that looked at failure modes in materials and structures. Sort of a physical analysis of how things break... The professor was very well read and a big fan of history - England and Scotland during the early years of the Industrial Revolution.
He walked in and gave us a few minutes to ponder the words and particularly those that were written in with red chalk. It seemed a bit familiar and turned out to be an entry from The Diary of Samuel Pepys.1 Smiling he took out a curious looking blog of glass with a tail and proceeded to strike it with a hammer. Not terribly hard, but hard enough that a glob of glass should have shattered.
Now watch this carefully..
He flicked the end of the very thin tail with his thumb nail and the entire blog exploded in a shower of tiny pieces of glass. He has us come up and look at the remains - it had mostly turned into powder.
As we walked back to our desks another blackboard came down and he launched into a discussion of how these globs - he called them Rupert's Balls, but they are also known as Prince Rupert's Drops - were made. A couple of stresses are involved and a lot of potential energy is stored in the glass. Breaking the thin tail starts a shock front that moves through the glass at something over Mach 5. Not much of the structure survives. Then it was time to begin to sketch out a model of the physics and the board quickly filled with drawings and equations while a discussion with the class was guided towards something that made sense.
It was an entertaining moment in a great class. I won't go into any detail of the physics as this video gives a nice discussion.
The video points out some important things. First you get some insight into how the blog is made. Then comes some high speed photography which adds some information on the dynamics of the failure mode. And finally, taking advantage of the fact that glass is transparent, a look at the structure in polarized light. Just the information you need to solve the puzzle.
Although it is physics, this is how many problems can be approached. You observe carefully and in enough ways to sort out missteps that might lead you to the wrong model (it is very easy to build sort of descriptive, but wrong models). And you try to make the problem as simple as possible focusing on the important points and describing just enough.
Last week I was doing a bit of work for someone who wanted me to look at a "big data" problem they intended to solve.2 I saw a different way that could be done with a bit of thinking and some simple math giving an answer to more than enough accuracy for his decision. Dealing with the data would have taken quite a bit of time and may have been too noisy to have found a clear signal. There are domains where hunting through boatloads of information works well and domains where it is inappropriate as well as those where you really need both. For the record I'm as impressed by the big data free analysis that goes on in Trader Joe's as I am by some of the very sophisticated techniques that places like Google users. I suspect the Trader Joe's analysis is often more applicable to their well constrained problem.
I end with a beautiful triumph of 19th century physics. How do you describe the motions of the air molecules in the room you are in? It turns out the study of thermodynamics - a field that became hot (sic) as people were working out how to make better steam engines - led to a statistical description of what was going on at a molecular level. An average room has several orders of magnitude more information in the motion of its air molecules as all of the information we and our machines have produced throughout history. It turns out the description is simple and elegant and a nice proxy for the average kinetic energy of a molecule is easily measured.3 All you have to do is measure the temperature.
1 Specifically January 13 1662. I had read the diary as part of a history class in high school, but detail like this sailed over my head at the time. I've re-read the book since and strongly recommend it is you are unfamiliar. He lived during an interesting period, knew everyone who was anyone and had a good eye for detail - plus he kept that dairy.
2 Big data is such an awful phrase. "Data" is a terribly loaded and misused word and big is approximately meaningless.
3 At least a good approximation, but I don't wish to get into the deeper physics.
A short recipe shard and a bit of hacking around I'm doing these days. First the hacking around.
I'm a vegetarian, but will get implicit eggs although I usually avoid it. Finding a good substitute is difficult and I'm off on a different direction. An egg gives some fat, protein and liquid, adds structure, helps with emulsifying and has flavor - and probably some other things I'm missing. Flaxseeds are really great at providing viscosity and that might give structure. In baking they have little taste and some baked goods effectively mask the eggs. I boiled 50 grams of flaxseed in about 500 grams of water. When it got thick I strained it a few times through cheesecloth (somewhat messy) and, not knowing what I was doing, make a batch of chocolate chip cookies using vegetable shortening for the fat and a one to one substitute of glop 'o flax for eggs. They came out reasonably well and were tasty (I give away most of what I bake if sugar is involved so I don't OD) I didn't detect a strange taste. There are other directions this can go, but it is an interesting start.
The recipe is really just a quick throw together inspired by a pound of dry roasted cashews from Trader Joe's. These are really addictive!
° 1 tbl extra-virgin olive oil
° 1-1/2 tablespoons ground turmeric
° 1 tsp Kosher salt (you may want more if you are a salt fiend)
° 1 pound unsalted, dry-roasted cashews (omit the salt if the cashews are presalted)
° Heat a wok large pan over medium-high heat.
° Add the oil, turmeric and salt and quickly stir to make a past.
° Add the cashews and stir-fry for a minute. Turn the cashews in the pan to coat with the spices and don't let them
° Pour into a colander and shake shake shake it to get rid of the extra spices. I had to brush off very dense regions of turmeric.