Monday, September 25, 2017

Book Review: A Mind at Play

A Mind a Play: How Claude Shannon Invented the Information Age
Jimmy Soni, Rob Goodman
Biography, computers

Claude Shannon was the Isaac Newton of information theory--and that's not an exaggeration; he reified and measured the concept of "information" much as Newton made sense of force and acceleration. Unlike Newton, he seems to have been a genuinely playful and sweet-natured man. After revolutionizing communication, he rode unicycles, taught himself to juggle, and built whimsical machines--like the box with the switch on top; when the switch was turned on, a mechanical arm emerged, turned off the switch, and retracted.


A Mind at Play is not a super-dense book, either as biology or as mathematics. Its core is a very nice summary, very light on mathematics, of just what it was that Shannon did. I think the authors missed a couple of tricks for the more knowledgeable reader--the deep connections between information entropy and physical entropy go unacknowledged--but the book is well-written and provides a good, sympathetic character portrait.

Among the good books that overlap with A Mind at Play are:

  • The Innovators, Walter Isaacson
  • The Idea Factory, Jon Gertner
  • The Information, James Gleick (much more technically rich)

Sunday, September 17, 2017

Book Review: A Death by Any Other Name

A Death by Any Other Name
Tessa Arlen
Mystery

I sometimes feel as though I ought to give new authors a chance, and not just stick to my old favorites. In this case, I was swayed by the back-cover quote: "pitch-perfect classic whodunit."


I should have remembered the rules.

A Death by Any Other Name is just plain badly written. It's not entirely Tessa Arlen's fault; there are a lot of problems that would have been caught by any competent copy editor. Consider:
  • "Fraulein" is several times misspelled as "Frauline."
  • A chafing dish is used to warm food. A chaffing dish, if there were such a thing, would have to either convert food to inedible husks, mock it in a good-natured fashion, or confuse its radar.
  • One character--an educated man--says that France has been England's foe "since time in memoriam". The phrase Arlen is looking for is "since time immemorial."
  • Only other servants would have referred to the butler as "Mr. Evans." The family and guests would have called him "Evans." 
Then there's the wooden dialogue. This is supposed to be a man speaking with "simple conviction", having forgotten his "showy manner":
She brought refinement to the dishes she prepared that far outshone anything I have had from French chefs in more prestigious establishments.
Try saying that out loud and sounding natural. There are plenty of other solecisms as well--run-on sentences, misused commas, paragraphs that have unattributed dialog from two different characters, and an implacable devotion to telling what the characters are feeling rather than showing. Nor does Arlen understand that the "action" of a whodunit takes the form of "information is revealed" (either to obscure or to enlighten), so the pacing is nonexistent too.

At this point, one of my regular readers is already mouthing the question: "So why did you keep reading it, then?" The answer is that I still hoped that A Death by Any Other Name might prove to be interesting as a whodunit--that is, as a puzzle, a technical challenge, an ingenious piece of misdirection. Alas, I was disappointed in this as well. The whole of the deductive process displayed boils down to this (SPOILER, if anyone cares):
  1. This piece of paper, which we found quite by accident, has a 7 written in the continental fashion, with a bar through it: 7.
  2. X is French.
  3. Therefore, X is the murderer!
Not only is this feeble, it's not even original.

I'm not happy to have to pan this book. It's the kind of thing I'd like to like, and the publisher is one I'd like to support. But the author didn't do her job, and the editor didn't do theirs.

My usual suspects for whodunits are Steve Hockensmith, Aaron and Charlotte Elkins, and sometimes Anthony Horowitz. If you find any others, let me know.

Thursday, September 7, 2017

Book Review: Empire of Things

Empire of Things: How We Became a World of Consumers, from the Fifteenth Century to the Twenty-First
Frank Trentmann
Sociology, economics

At the end of Empire of Things you'll find 107 pages of densely-packed, small-print end notes. You'll also find an apologetic note from the author:

. . . these are only the tip of the research iceberg on which this book rests. Readers who wish to delve deeper . . .can browse my 260-page working bibliography at: http://www.bbk.ac.uk/frank-trentmann/empire-of-things/.
This, mind you, after 692 pages of text. These aren't light, fluffy pages, either. It's like you're walking into Dr. Trentmann's Famous Museum of Consumer Facts. Imagine a long room, lined with glass cases, each of which is crammed with exhibits, and where each exhibit has an explanatory plaque which you're expected to read. A semi-random trawl through the book's first half furnishes some examples. Page 175:
In 1800, Paris and London made do with a few thousand oil lamps . . . By 1867 . . . Paris was lit by around 20,0000 gas lamps. By 1907, it had 54,000; London had as many as 77,000 lights . . . each burnt 140 litres of gas a night. When the First World War broke out in 1914, Paris was seventy times brighter than during the 1848 revolution.
Page 212:
A typical local Varieté cinema in 1904 showed moving pictures to around 70,000 viewers a week . . . Lexington, Kentucky had two cinemas for its 25,000 inhabitants . . . By 1914, Britain had 3,800 cinemas. London alone had almost five hundred, with seating for 400,000, more than five times that in music halls . . . 250,000 Londoners went to the cinema, every day. In New York City, the weekly attendance was closer to a million . . .
Page 239:
. . . a factory worker typically earned $590 in 1890 . . . almost a million new homes were constructed in 1925 alone . . . In New York and Philadelphia, 87 per cent and 61 per cent were renting in 1920. In 1930, this was down to 80 per cent and 42 per cent.

Page 326-327:
. . . in the early 1960s, public expenditure was 36 per cent of GDP in France (33 per cent in the UK; 35 per cent in West Germany); by the late 1970s it had reached 46 per cent in all three . . . In the USSR, consumer durables grew at a rate of 8 per cent a year . . . the Hungarian government promised its people 610,000 TVs, 600,000 washing machines and 128,000 fridges within the next three years . . .
It's not that the facts aren't interesting; they are. It's not that Empire of Things is badly written, either, although someone should let Frank Trentmann know that it's no longer a flogging offense to use a contraction now and then. It's just that there's so . . . damn . . . much of it. Even the most dedicated reader isn't going to retain more than a tiny fraction of this information. Empire of Things would have been so much more memorable if it had only concentrated on telling a story. (The second half, which is organized by concept rather than chronologically, is a bit better than the first.) But if there's a theme running through the book, it's one that only really becomes clear in the final 20 pages or so. The book is intriguing in spots, and enlightening in spots, but as a whole it's something of a blur.