archive
REX·DAILY
Wednesday August 19 Anno 2026 · Ibaraki, Osaka
I Daily Wisdom
Robert Greene The Daily Laws August 19 · Let Them Win on the Minor Points

In 1782 the French playwright Pierre-Augustin Caron de Beaumarchais put the finishing touches on his great masterpiece

The Marriage of Figaro

. The approval of King Louis XVI was required, and when he read the manuscript, he was furious. Such a play would lead to a revolution, he said: “This man mocks everything that must be respected in a government.” After much pressure, he agreed to have it privately performed in a theater at Versailles. The aristocratic audience loved it. The king allowed more performances, but he directed his censors to get their hands on the script and alter its worst passages before it was presented to the public, To bypass this, Beaumarchais commissioned a tribunal of academics, intellectuals, courtiers, and government ministers to go over the play with him. A man who attended the meeting wrote, “M. de Beaumarchais announced that he would submit unreservedly to every cut and change that the gentlemen and even the ladies present might deem appropriate. . . . Everyone wanted to add something of his own. . . . M. de Breteuil suggested a witticism, Beaumarchais accepted it and thanked him. . . . ‘It will save the fourth act.’ Mme de Matignon contributed the color of the little page’s ribbon. The color was adopted and became fashionable.” Beaumarchais was indeed very clever. By allowing others to make even the smallest changes to his masterpiece, he greatly flattered their egos and their intelligence. Of course, on the larger changes later requested by Louis’s censors, Beaumarchais did not relent. By then he had so won over the members of his own tribunal that they stridently defended him, and Louis had to back down.

Daily Law: Learn to lower people’s defenses by agreeing to matters that are not so important. This will give you great latitude to move them in the direction you desire and get them to concede to your desires on more important matters.

The Laws of Human Nature

, 7: Soften People’s Resistance by Confirming Their Self-Opinion—The Law of Defensiveness

Steven Pressfield The Daily Pressfield Day 231

GO BACK TO GENRE What saved me was going back to genre. Specifically: One convention of a supernatural thriller/Devil story is that the Evil One must be conducted into the material world by a female. This was nowhere in my original story. I had no scene like that, no sequence, not even a shred of a glimmer. Yet that convention was present, big as life, in The Ex orcist. It could not have been more obvious in Rosemary’s Baby. Why hadn’t I seen it? Why hadn’t I implemented it? Black Irish JAB #13, “Starting Over, Pa rt Two, There are rules, Shawn was (gently) telling me, that you, the writer, cannot break. Learn those rules and adhere to them.

Leo Tolstoy A Calendar of Wisdom August 19

Life is constant movement, and therefore goodness in life is not a certain state, but the direction of movement.

This direction is not in serving yourself, but God.

Some people seek goodness in power, others find it in science, others in dissipation.

Those people who are really close to goodness understand that real goodness for one is when all people have this goodness and share it among themselves.

—BLAISE PASCAL

Happiness is a thing which a person wishes only for himself; goodness is a thing which a person wishes for himself and for others.

Happiness can be achieved by struggle; goodness, on the contrary, by being humble.

Real goodness is to serve God.

You should do goodness without choosing to whom.

Good things, once done, will never disappear, even if you forget about them.

There is only one way to be happy, and this is a sure way: to do goodness and to share this goodness with others.

II The Day Ahead
  • Today · all day Jakarta trip with Minori and Eisuke (until Sep 2) Google Travel itinerary
  • Today · 15:00 ⭐ 基礎研全体Mtg (Group A) · Presentation 基礎研全体Mtg 2026
  • Tomorrow · 18:30 サークル(木) 2026年間予定
III Signals & Dispatches

AI / ML

  • @claudeaisource ↗ Claude can now send Gmail replies and manage Google Drive files, with approval controls when needed.
  • @samasource ↗ OpenAI has paused some frontier reinforcement learning training to meet higher alignment, security, and monitoring standards.

Drug Discovery

  • @julian_englertsource ↗ A wet lab benchmark of Claude protein designs found a 26.8% binding hit rate and binders for 14 of 15 evaluable targets, though these are not yet therapeutics.
  • @adaptyvbiosource ↗ Adaptyv Bio is publishing the wet lab case study behind Anthropic's protein design benchmark.

Biotech

  • @brizbrainspinesource ↗ A Phase 1 T cell therapy trial targeting aggressive childhood brain tumors reported prolonged disease control in some patients.
  • @Banana_Oncologysource ↗ Newer CRISPR knockout data suggest much stronger PAX8 dependency than the older Project Drive RNAi results indicated.

Tech

  • @trq212source ↗ A proposed agent economy would make SaaS products headless, expose them to agents, and charge enterprises per interaction.
  • @svpinosource ↗ After a 90 day Linux experiment, one developer says Apple's hardware and ecosystem remain difficult to beat.

Japanese Politics

  • @toroneisource ↗ 自民党の自衛隊OB枠の参院比例候補が全滅した場合、自衛隊の待遇改善をめぐる政治の示唆があるという見方。
  • @masaru_kanekosource ↗ 米中間選挙をめぐるReuters Ipsos調査とイラン情勢を引き合いに、高市外交への批判を展開。
IV Sky over Ibaraki
☁️ +26°C
humidity 67% · wind ←8 km/h
cloudy, light wind
V What Rex Can Do Today
  1. Draft a one page internal note on what autonomous protein design changes for StemRim's discovery workflow.
  2. Prepare a concise follow up for the Basic Research Division presentation while the details are fresh.
  3. Review the Jakarta itinerary and flag any travel or family logistics that need attention before departure.
VI Toward the Long Game
  1. Read the Anthropic and Adaptyv Bio protein design benchmark, then write five technical takeaways for your own workflow.
  2. Reproduce a small open source protein or peptide design experiment and document the full computational path from target to ranked candidates.
  3. Find one Japanese or Asian AI drug discovery researcher whose work overlaps with StemRim and prepare one precise networking question.