Late summer/early fall 2024 books

Portraits of Learned Men by Paolo Giovio I Tatti collection. A collection of mini biographies of men of the Renaissance. This collection was supposed to be housed in a Museum on lake Como with physical portraits of the men in the descriptions: the idea being that people would wander from portrait to portrait and take in the mini biographics. Some of them are intensely funny poop flinging by Giovio. In my opinion they're an odd collection of figures, and their brevity doesn't allow one to develop much curiosity about any of the figures to inspire further reading. Still, as lists go it is probably a useful starting point for renaissance scholarship. Sort of Cornelius Nepos for renaissance nerds.

Detecting Regime Change in Computational Finance by Jun Chen, Edward Tsang. I got excited reading the PDF version of this thing in the Sauna because it mentions HMM and something I hadn't thought of: Naive Bayes for identifying regime change in time series. Unfortunately I didn't simply read it through: I thought that was enough to fork over my hard earned bucks. This was an error, probably produced by high carbon dioxide levels in the Sauna. This is something I've been monitoring lately; for the 1-man booths it reaches alarming submerged U-boat levels after 10 minutes or so, and I'll have to do something about it, if only to avoid buying shitty books while under CO2 intoxication in future. In any case the book is a sort of student project which was bulked out to a 100 page book. If you've never been exposed to these ideas it might be helpful: for me it's just annoying I forked over adult sized loot for a student project. Their indicators are silly: some of them backward looking, and naive Bayes is just rolling means and standard deviations, even though they wrote it down different.

The Idea Factory by Jon Gertner. The author is a NYT columnist with all the "point and sputter" flaws that mar his caste. Decent book though; many of the characters mentioned have been lost to time, and assembling a historical account of the sprawling phone company and its research and technological development is a considerable achievement. When people today think of Bell Labs, they think about crap like C, Unix, Information Theory; maybe the transistor. This is a research firm which invented .... useful telephone wires (huge material science problem), vacuum tube amplifiers, digital circuitry, information compression, relays, statistical process control, plugboards: stuff so basic we don't give it a second thought. Solar cells, electronic calculators, error correction, masers, cell phones, mosfets, ABM systems, microwave transmission of data, video conferencing, microphones, communication satellites, automated switching systems, nuclear reactors, radio astronomy, molecular beam epitaxy, growing crystals of exceptional purity: they pretty much invented the modern world. The period of productivity lasted almost 100 years depending where you put the beginning of it (IMO should be in Western Electric days rather than its formal creation in 1925), and even when it turned into a glorified physics department in its later years it was still cranking out important results. Upside to the book: interesting account with historical context. Downside; it had virtually nothing about the culture which produced all of this: he literally has more information about the interior design of the labs than the management practices. Some hints here and there: creatives were unleashed, but during the best years, people were required to do stuff which might conceivably help AT&T in its mission to provide communication technologies. In the early days researchers were required to spend time working with the actual equipment used at the time: climbing telephone poles, working plugboards and so on.  They also isolated research programs from upper management to prevent premature announcements; something present day labs should consider in our era of press release fake "progress." There were also a wide variety of technicians, machinists, metallurgists, electrical engineers, chemists; the physics guys smoking pipes at the chalkboard got a lot of the credit but were a small minority. For "early life" Millikan's grad students were important in the early years: tinkerers, not grandiose theorists. People were required to work on applied stuff.  The section on Shannon was very good: much better than the existing biography written by management consulting goblins, so there was that. One passage which stood out (page 122)

(Mervin) Kelly would often point out that the Labs workforce including PhDs, lab technicians, and clerical staff—by the early 1950s totaled around nine thousand. Only 20 percent of those nine thousand worked in basic and applied research, however. Another 20 percent worked on military matters. Meanwhile, the rest of the Labs’ scientists and engineers—the majority—toiled on the never-ending job of planning and developing the system. Their work was arguably less glamorous.

Early 50s Bell Labs was definitely peak Bell Labs: it's worth noticing most of the people there were working on very practical problems. Only a small minority worked on basic research, and even that was tied to practical problems. Shannon almost certainly wouldn't have thought up information theory or boolean algebra applied to circuitry (effectively inventing digital electronics) if he hadn't been working at the phone company. Another pattern: virtually all the noteworthy workers there were redneck kids from the midbest from lower to middle class families and generally humble schools: much like in the space program back when it was doing innovative stuff, rather than preventing innovative stuff (with more girl bosses) as it does today. No immigrants or other "diversity;" not even any Ellis Island Americans or married women were allowed for most of its history.  The author was amazed by this and brought it up constantly. Sorry fellow Ellis Island Americans; most of American physics and science was done by WASPs; you know, like the people everyone around the world thinks of as actual Americans.  Similarly, he was amazed that technicians were treated as peers rather than scumbags: something every experimental physicist knows is The Way to Get Things Done in the national labs system at least. Another thing recorded: open door policy. The technician could go bother the director if he thought it would solve a problem. People weren't even allowed to close their office doors; Shannon excepted. Author's explanations of complex scientific topics: occasionally strained, but over the target. Good analysis of anti-trust concerns leading to AT&T building all kinds of wacky things for the military; also interesting that a guy working on nuclear weapons chaff discriminators ended up being a key person in inventing the cell phone.

His analysis of what the special sauce was at the end wasn't very interesting, but as this is my special subject I have to rant about it. To his mind somehow AT&Ts regulatory situation was supposed to be part of it; google couldn't possibly do this according to the author. This despite Google having absurd free cash flows, about 150,000 too many employees nobody would miss, and a 14 year old moon shot department which burns multiple billions a year; maybe half what the NSF spends per year: zero Bell Labs tier innovations thus far. Of course Google-X isn't Bell Labs, it's a collection of unrelated random science projects run by people unaccustomed to managing large (or even small) scientific enterprises which succeed. They could afford 9000 talented people like peak Bell Labs working in a well stocked lab: that's not how they roll. Also none of the problems they attempt solve any business problems google actually has -none of them are even in the ballpark of problems Google have. When AT&T tried their hand at top down directed wacky ideas with stuff like the "picturephone" they came up with bupkiss as well (though unlike Google-X's problems, that actually functioned: just nobody wanted one at the market price). Google can afford it: they effectively have a monopoly on advertising and search in all Western regions. They just don't run their research division the way AT&T did. They top-down dictate the problems to be worked on rather than limiting the problems to the more broad but general kinds of things Shockley, Townes and Shannon worked on. The kinds of people who made Bell Labs great also basically don't exist outside of Mennonite apostates (I worked with one: genius tier talent and based tinkerer; I hope she has 400 kids like her relatives do), and if they did Google-X wouldn't hire them. Since they're not solving basic problems like how to make a goddamned wire or triode that functions reliably, you don't get the cool superstructure that can grow on top of such practical knowledge. Heck, no tech company would hire Shockley or any of the other myriads of uptight bow-tie wearing personality types that Got Things Done in those days. As a reminder, Bell Labs guys did research on rivets for the linesman's leather belt, flux and solder compounds, wire splicing techniques, compounds for making reliable relays and crossbars; absurdly basic things: Google wouldn't dream of hiring people to do serious thinking about rivets. Without such real-world research: no transistor and definitely no information theory. The other thing which seemed relevant: the management structure of the place was made up of people who actually achieved things and knew how it was done: Mervin Kelly led the team which perfected the vacuum tube triode.  He later led the team for wartime mass production of radar magnetrons. He also knew how to hire people and had the vision to invest in semiconductor physics. They made him president and chairman of the board during peak years. They did not make genius tier dudes managers randomly: they had to be good scientists and good scientific managers. They also weren't chosen for agreeableness or the basket of quirks, neuroses and psychopathologies that mar the character of Googly people: they were chosen for effectiveness in their assigned role.

The author then talks about some resort-like research institute for doing neuroscience as a future Bell Labs, rather than a normal research place. None of the ingredients are in place there: it doesn't even have permanent employees, let alone practical sub-problems to work on. And he goes on to quote Steve Chu, the worst lab director in LBNL history, mentions the possibility the fame of the people they hired as causative (rather than a result of being there): as if it were celebrity culture that got us .... wires and transistors. He does quote Chuck Elmendorf who was over the target: lots of talented no-names and technicians along with the later-famous dudes who often got credit for the unsung work of many. Anyway, decent thought provoking book, even if it wasn't all that I would have hoped for. Certainly tremendous scholarship went into it: I just wished he'd have shut the fuck up about the fact that Shockley predicted the Idiocracy future we live in (author kvetching over this was both irrelevant,  defamatory and annoying), the interior design of the labs and concentrated more on the sociology of the place. Upside it made me reread Shannon's old papers on mind-reading machines and cryptography, as well as the previous link by Mervin Kelly which actually does talk about the sociology of the place and what the secret sauce was. Lots of good summaries in Eric Gilliam's substack. He even noticed that machine shops are based.

The Christian Recovery of Spain Henry Edward Watts. I've read numerous histories of Iberia at this point; large and small alike. This one purports to be a complete, albeit abbreviated history of the  reconquista.  The author appears to have made his living as a sort of spaniardist and early translator of Don Quixote, though this is hilariously filled with British jingo black legend poop flinging at the poor benighted Spaniards (according to Watts anyway). The author takes great joy  in dispensing with legends of Pelayo, Roland and Charlemagne.  I suppose the "British" contemporary with these legendary heroes were illiterate Saxons living in contemporary Denmark, or, if you're of celtic kidney, the even more fake and gay King Arthur, whereas Spain was a thriving and civilized country before (and after) the Moor came. This is typical British fedora-atheist midwittery, he even opines that St. James wasn't present at the battle of Covadonga with flaming cross and sword: this is completely ridiculous (of course St James was present with flaming cross and sword). While some may find this entertaining, and his historical arc is good, Mary Platt Parmele's Short History of Spain might be a better choice if you care for Spaniards and don't like sniggering fedora-protestant halfwit takes. That said, the book has OK engravings in it, and once the Spanish royals took a few English wives, he seems much more sympathetic towards them. FWIIW I was reading this in Asturias sitting 20 yards opposite a red-headed English neanderthal woman bellowing half-baked received pronunciation about things nobody could  conceivably care about, including her husband who is mere inches as opposed to yards away. The Spanish ladies in the restaurant don't bellow or don't look as cannibalistic as she does. While I can't see from where I was sitting: I get the distinct impression she's the type of woman to ostentatiously scratch her vagina in public, and the Senioritas absolutely ain't.  While the Englishman had his day, as of today, the average Englishman is far more debased than the average Spaniard who at least eats decent food and lives his life with some semblance of dignity even if he works as a plumber. Not to mention nobody in Spain gets tossed in the pokey for saying shit on facebook.  Anyway fuck this book, but if you can get over the fact that the author is a proto-redditard who absolutely loathes Spain, it has its moments.

The Charterhouse of Parma Stendhal. The Red and the Black is one of my favorite books (it will never not be hilarious it was Al Gores favorite as well), and Bappy goes into transports over this one so I gave it a look. Galloping good fun from start; lots of action, skullduggery and 'mires of the Emperor. It bogs down in the middle in the scheming of women in court life and is absolute insipid twaddle by the end with a ridiculous love affair a la "Sorrows of Young Werther." I guess it's a clever book; the narrator is a kind of unreliable thing especially towards the end -might have been on purpose, might have been because it was written in two months in a slipshod way. It took me three days to read 3/4 of it and two more months to get through the bilge at the end. He should have asked Dumas (who I love) or Victor Hugo (who I dislike) to finish it for him. Apparently was inspired by a semi-fictionalized account of Pope Paul III's dissolute youth as well as Stendhal's time in Italy under Napoleon. Kind of weird it is about Italian court life, but told by a Frenchie; there are probably subtle ethnic jokes in this.

Information-theoretic Signal Processing and its Applications by Stephen M Kay. I liked Kay's earlier statistical signal processing textbooks which had some strong information theoretic components to them. This one, which I have only skimmed is more or less more of the same, with some updates. Decent text if you need stuff like rolling GMMs.

Sex and the Japanese by  Boye Lafayette De Mente. My excuse for reading this: I was looking for books on Shinto for a pending visit to Japan and this one came up. The author is an interesting fellow who ended up in Japan after WW-2 and more or less made a career out of it after retiring from his work in the early NSA, having various other adventures you can read about on his wiki page, including crossing the Pacific in an amphibious jeep. The book itself appears to be 60s era lurid trash; I have no indications he was wrong about anything, it's just not very edifying and probably is meaningless excepting as a historical document of what postwar American expats got up to in Japan. The latter half of the book is a dirty phrasebook; sort of like Making Out in Korean. Anyway, 60s version of a pre-conversion RooshV kind of thing. He also wrote a book reviewing the various strengths and weaknesses of Asian women in general; hence the comparison to contemporary pick up dudes (I skimmed it; both more retarded and more insightful). Makes me a little sad as it is somewhat a document of the Japanese being a conquered people; some of the "social innovations" he speaks of probably directly came from the US embassy.

Shinto: the Way of the Gods and Shinto the Ancient Religion of Japan both by W.G. Aston. Thinking about a visit to the Land of the Rising Sun, I realized I knew next to nothing about Shinto; more or less the software (along with Zen and of course Confucianism, which I do know something about) that Japanese society runs on. The latter seems to be a pamphlet sized abbreviation of the former, and is completely useless. The former is a bit marred by 19th century sociological views of the function of religion in the human experience, but at least it paints a picture of the subject at hand. This book is tough to rate. It's bad; I think the author is an ignoramus (though he may have been well informed for his time) who doesn't understand Japan, Japanese religion or much of anything. It's good in that it explains quite a lot of Shinto custom and folklore and like most things Japanese, Shinto custom and folklore is elegiac and beautiful. After getting through it, I'm not sure I learned anything about Shinto, but at least it made me feel something of the outlines.