by Richard Osman

This clever confection is a mystery about murderers who don’t kill people and murders that mysteriously do not occur. It’s spun sugar, but in the days after the murder of Alex Pretti, perhaps confectionery is what we require.

by Becky Chambers

Winner of a Hugo for Best Novella, this is the story of a tea monk — a sort of itinerant psychotherapist — who meets a robot on the road. Robots were emancipated long ago, and long ago they left and absented themselves in the wilderness. Now, a volunteer has been sent to check up on how the humans are doing. The result is very heavy on exposition, and indeed nearly everything here is exposition. The robot, whose name is Splendid Speckled Mosscap, owes something to Iain M. Banks but is exceptionally well drawn.

by Fania Oz-Salzberger and Amos Oz

For some time, I’ve needed a sounder foundation for my claim (in Uncle Fred’s Book) that Jews are inclined to argue. It’s true, as my cousin reminds me, that this goes without saying and really needs no justification. I want to have the footnote, just in case.

This is perfect. I’ve been reading a lot of Twitter from Fania Oz-Salzberger about October 7 and the ensuing war; she, along with Simon Schama and Simon Sebag-Montefiori, has made consistent good sense.

This volume a wonderful essay on strong women in (and out of) the Bible, some terrific notes on humor, and demonstrates a wonderful (and I think sound) approach to the problem that the ultra-orthodox pose to secular Jews.

Also, this was a delightful read.

Speaking of tech support: when you need help or advice on a project, it often helps to describe just what you’re doing. Yes, some aspects may be confidential, but a general outline saves lots of time and potential confusion. Having trouble with footnotes? You might have seven, or you might have thousands, and it makes a difference. Michael Bywater once said, “A man has no secrets from his valet, or his software developer.”

This week’s Tinderbox meetup was a deep dive into the meaning of names in Tinderbox. Each Tinderbox note has a name and a text space; the name is typically a short descriptive title and the text can as long as you like. What are the names for?

In the beginning, in Tinderbox’s Storyspace prehistory, the name was a way to identify which note was which on a small, low-resolution screen. In Storyspace 1, the title was a Str32 — a 32-character buffer — because there wasn’t a proper string library for Pascal in the 80s. (It wasn’t my fault, though most of the design errors since those days are.). I’m not sure it was done with much thought; I didn’t think much about it for a long time, beyond wanting to get rid of those fixed buffers.

The duality between title and text, though, has always been interesting. In afternoon, the titles are often in tension with the text. The note named “poetry” begins with a question: “ she asks, ”. Is this poetry? Is poetry more exact? Exactly what is the Name telling us?

But Names in Tinderbox let actions refer to other notes and use them as prototypes or as sources of data. If you want to do this, the naming of notes can be a difficult matter. For once thing, it becomes desirable for notes to have distinct names. For another, names that look like actions can cause trouble, just like little Bobby Tables.

In the forum, this prompted Andreas Grimm to raise the concept of Wittgenstein’s Ladder. Tinderbox tech support sometimes requires the philosophy of language.

(Speaking of early Tinderbox, I recently scrapped support for #operators, which were used in queries in Tinderbox 1-2 and have been deprecated for 20 years. “Surely,” I told myself, “no one is still using them!” Dear reader, this weblog was still using them, and the change broke my booklist for a time.)

If we are to free ourselves from subservience to ignorance and cruelty, it will not be through the new civil war that the Silicon Sovereigns sometimes imagine, the hordes of brown-hued rioters spreading out from ruined cities to loot their Fatherland. We will not have a second Gettysburg; armies don't work like that anymore.

What it will be—what it will have to be—is a notionally amicable separation in which a notional Union is preserved for old time’s sake, while we keep ourselves separate and avoid meeting, talking, or having anything to do with the others.

That’s already well advanced. When I was a kid, I knew Republicans. Today, I believe I know one actual Republican, a guy who runs my fantasy baseball league. I am acquainted with two or three more, kids I knew at school, a college classmate who got a little but famous, a second cousin once removed who has a Facebook habit. There was a Republican in hypertext research, back in the 90s, but I haven’t heard from him in decades.

But that’s not enough, not when the Red States are sending ICE to shoot people in the face for fun. We need a restraining order.

The murder of Renee Nichole Good by ICE officer Jonathan Ross, who remains at large, has upset me deeply. It has been hard to work. Objectively, I’ve shipped a Tinderbox backstage release with some fairly ambitious new fixes, straightened out a tricky support issue on Tinderbox actions, made some progress on AI fiction and paid the mid-month bills. Still, it feels like I can do nothing but Twitter, Facebook, and Chess.com.

Half the country thinks this is just fine. The President, the Vice President, Kristi Noem, Robbie George: she had it coming. The local Dems: a collective shrug, not even a meeting.

I think that, if you believe this is OK, I do not want to break bread with you, I do not want to meet with you, I do not want to know you, and I do not want to be governed by you—not even if you have the votes.

This happened before, in 1775-1776 and again 1859-1860. I think disunion may be the only resolution.

I read David Owen’s recent New Yorker piece on dyslexia with special interest, because I am dyslexic. The New Yorker piece is pretty good, though it’s also dismaying that the Gillingham approach, an experimental reading program that saved my bacon back in the early 60s, is still experimental. It does appear that we know a bit more about the neuroanatomical basis for dyslexia thanks to MRI studies, and there may even be some evolutionary reasons that is makes sense for social hunting animals to keep around neurodivergent strains; we might not be great at reading but, it seems, there are compensations that stem from a differing approach to exploration.

by Steve Wolfram

A fascinating research report from a fascinating researcher. Wolfram studies cellular automata, which are very small and simple models of abstract computation in which each “cell” adjusts its state according to the state of some number of neighbors. These automata are even simpler than neural networks, but they, too, can learn. It appears that machine learning, in this case at least and perhaps in all cases, including animals and people, may be a matter of sampling lots of abstract computations and refining those that seem to have some slight utility for addressing the current situation.