Lightning Studios

The "Personality" of Each Genre in the Jewish Library : A Synthesis, informed by Quantitative Analysis conducted on the @joinlightning knowledge graph of 900 texts, 2M entities, and 16M interconnections. Tanakh is a world of people (highest person density: 9.3%), places, and themes. It presents narrative populated by characters in geography, organized by recurring motifs. It virtually never interprets itself. Everything else is a response. Mishnah is a teaching machine. 37.2% of its entities are teachings — the highest proportion of any genre. It is the most assertive, least questioning, and least interpretive genre in the canon. The Mishnah legislates. The Talmud Bavli is the great questioner. 14.4% of its entities are questions (the highest), and it holds the highest Q:T ratio (0.688) by a factor of two, with the third-highest tension density (31.6%). Its cognitive mode: interrogate, sustain tension, eventually resolve. Commentary is the great interpreter. 31.6% of its entities are interpretive claims — the highest proportion. It has the highest tension density (32.8%) and the largest absolute entity count (1.6 million). Commentary's primary act is interpretation, not original teaching. Mussar is the value teacher. It has the highest value density (12.3%) and a strong teaching orientation (28.9%). Its mode: name the virtue, teach the practice, shape the character. Kabbalah is the concept revealer. 20.6% of its entities are concepts — the highest proportion — and its Q:T ratio (0.114) is among the lowest. Kabbalah discloses the architecture of reality. It teaches and reveals; it rarely questions. Midrash is the people-rich questioner. It has the highest person density after Tanakh (4.9%), a strong Q:T ratio (0.312), and balanced engagement across all entity types. Midrash is the most humanistic genre — populated by characters, driven by questions, anchored in narrative. Liturgy is a genre of concepts and teachings with a low questioning rate. It affirms and praises rather than interrogates. Its entity profile reflects a genre whose purpose is worship, not inquiry.

Have been running data analysis on our @joinlightning knowledge graph of Jewish library (900 books, 16M entity interrelationships), and the #1 value by frequency across the corpus is humility.

Waking up to read a little bit of the Bible, Isaacson on Da Vinci, Menci philosophy via @joinlightning and a touch of “Humor” from Stanley Donwood (Radiohead’s visual artist). Nobody does it like I do it. That’s a fact.

In the end, a teacher's only alpha is their passion for learning.

Everything in its place. @joinlightning secondvoice.substack.com/p/everything-i…

"I imagine Slavoj Žižek following Peter Sloterdijk, would say Carol’s relationship to Zosia is a classic example of cynical reason: despite knowing Zosia is a lie, she still cannot resist her charm. Or as Woody Allen’s character, Alvey Singer says at the end of Annie Hall, he can’t turn in his crazy brother who thinks he’s a chicken, because Alvey himself needs the eggs!)"

I wrote a philosophical essay on Pluribus. The show is about Safetyism. It demonstrates that helpfulness can be harmful and harmfulness helpful. Failure, sin, imperfection, are features, rather than bugs, of a good and human life.

Make the difficult desirable (again).

Before AI, many thought human dignity was a function of homo faber, man the creator. But now that AI can execute creative tasks, I argue, we should locate human dignity in homo religiosus, man the being who makes commitments and sacrifices. secondvoice.substack.com/p/what-is-the-…

Beautiful essay by @bobbyjgeorge on the past and future of meaningful work and its relationship to thinking. open.substack.com/pub/secondvoic…

Lightning: everyone is a philosopher

AI Poetry is Mid, But AI Poetry Translation is Next Level Can AI write great poetry? While I argued recently that it cannot, I must clarify that AI is a phenomenal translator of great poetry. I have come to this conclusion because I regard translation to be about rule following, and because the unit of translation is not simply the word, but the vectors of relationships between the words in a poem and their relationship to language across many dimensions. While poets develop an ear for these things, AI actually has the data to plot these relationships and take poetic license to new levels. secondvoice.substack.com/p/ai-poetry-is…

“Should we write for the AIs?” asks @tylercowen The Jewish commentary tradition know as Oral Torah was written for LLMs all along; we just didn’t know it. Until now. open.substack.com/pub/whatiscall…

Alex Jakubowski (@alexjakubowski), Sara Wolkenfeld (@RabbanitSara), and Ari Lamm (@AriLamm) ask: What are the goals of Jewish education today, what role should AI play, and why is character becoming a centerpiece? Watch here: 18forty.org/videos/torah-s…

The great walkers, Kant, Thoreau, Nietzsche, understood that thinking unfolds step by step. Their philosophies were shaped by movement, rhythm, repetition. Walking wasn’t a break from thought: it was the condition for it.

Our eight-year-old son was unfolding a paper lantern pattern on the kitchen table when I was reminded of a line by William Morris: "Have nothing in your houses that you do not know to be useful, or believe to be beautiful."

You can read one of the great works of Averroes here. Such an amazing title: The Incoherence of the Incoherence alexandria.wiki/texts/The-Inco…

"We’ve allowed “truth” to be colonized by one specialized meaning—correspondence to facts—and in doing so, we’ve impoverished our language, weakened our defenses of the humanities, and disabled ourselves from articulating why meaning matters."

We at @joinlightning got a grant to create Yochai, an AI-powered Torah chevruta or tutor that supports Jewish learning and teaching by creating personalized, dialectical engagement with our textual tradition. Woohoo! If you're a Torah teacher and excited (or daunted) by what we're doing please take 5 minutes to fill out this short survey which, I hope, will also be a meaningful exercise in making the implicit explicit. docs.google.com/forms/d/e/1FAI…

Someone once threw a match into an 'empty' gasoline drum. The word 'empty' suggests safety. But an empty drum—thick with vapor—is lethally dangerous. The words give you a 180-degree misreading of reality. This is what we train AI on: the words, not the vapor.

New essay by @bobbyjgeorge on architecture and language as living systems and the implications for AI. secondvoice.substack.com/p/what-would-a…