--- title: "New vs. better, 2026" type: essay status: draft date: 2026-10-14 topics: "AI,El Niño,social media,Bitcoin,privacy" pages: 2 related: [../DERIVATION.md, ../BIO_ROBERT_PIRSIG.md, ../QUOTES.md, ../notes/202610140004-whats-new-and-what-is-best.md, ../.claude/skills/essay-montaigne/SKILL.md] --- # New vs. better, 2026 *A draft. Claude derived it from the owner's notes ([`DERIVATION.md`](../DERIVATION.md)). Each "I" below stands on a note that the owner gave; where a note is missing, a marker says what the owner has to write. It is written in October 2026 and it says how things look this month.* In 2017 I learned a language because a tool failed. Three of us had a risk model in R, Spark would not run it, and so I wrote a part of it again in Scala. I did not choose Scala because it was new. I chose it because it worked. In the same months I was taking Latin classes, and nobody understood why, not even I. A man laughed when I told him. It took me years to see that he was laughing with me. I begin there because of a book that I go back to. Robert Pirsig opens his talks with two questions. "What's new?", he writes, "is an interesting and broadening eternal question, but one which, if pursued exclusively, results only in an endless parade of trivia and fashion, the silt of tomorrow." He would like to ask another one: "What is best?", "a question which cuts deeply rather than broadly". He wrote that in 1974. I say *better* where he says *best*, because I can compare two things and I cannot rank them all. I try his two questions on five things of 2026. **AI.** This month everybody talks about one subject, reverse engineering with agents, and I ran to write about it. That was the first question, and I followed it. In the same days an agent and I moved the tools of this book from Python to Gleam: 8,222 lines added, 1,647 removed. Gleam is the newer language of the two. Is it the better one? What I can show is small. Each new tool gave the same output as the script it replaced. The same output is "not worse". Better is what I will know in a year, when I have had to maintain it. **El Niño.** Nothing in this list is older. On 8 October 2026 the Climate Prediction Center of NOAA wrote that "El Niño continues to strengthen", and that a strong one, or a stronger one, is likely to last through January to March 2027. It comes back, and each time it is told as news. For a forecast that people act on, I wrote a rule down: the forecast in operations is the one that wins the benchmark, and a new model is tried in simulation, where a bad pick costs nothing. That rule is Pirsig's second question made into a procedure. I still feel the pull of the new model. **Social media.** A feed asks only the first question. It is the "endless parade" in a literal form. My own rule goes the other way: I only need to be seen by the right people, so I choose the words that the right reader would type. I do not know if that rule is better or only quieter. **[AUTHOR ANECDOTE: what social media gave you or cost you in 2026, in one case.]** **Bitcoin.** Its paper proposed payments "sent directly from one party to another without going through a financial institution". That was new. I have not done the work to say if it is better, or for whom. **[AUTHOR ANECDOTE: your position on Bitcoin in 2026, and the one fact that moved it.]** **Privacy.** Here my own book seems to stand against me. I argue that closed source failed. Yet in my notes my friends have no names: they are friend1, friend4, prof1. One man has his name there, because I asked for it. I think that this old thing is the better one, and not only the older one: open work, private people. A program can be opened and nothing is lost. A person who was named cannot be un-named. Where the line goes between the two, I cannot say exactly. What do I know? Five things, and of none can I say "this is best". I notice one thing only. In each of them the new announced itself, and the better had to be checked later, by somebody who stayed to maintain it. I will read this page again in October 2027, and I may contradict it. ## Sources - The two questions: Pirsig, *Zen and the Art of Motorcycle Maintenance* (1974), chapter 1, as quoted on Wikiquote; checked in [`QUOTES.md`](../QUOTES.md) and in [the note](../notes/202610140004-whats-new-and-what-is-best.md). Status: partial. - 2017, Scala, Latin, the man who laughed: [the first Scala](../notes/202610132352-the-first-scala-a-risk-project-at-ufrj.md), [Marcus Baffa laughed with me](../notes/202610132353-marcus-baffa-laughed-with-me.md). The owner's memory. - AI: [why I wrote fast](../notes/202610132130-why-i-wrote-fast-from-preprint-to-book.md); [the port to Gleam](../notes/202610132355-gleam-principles-and-the-port-to-pure-gleam.md), pull request [#27](https://github.com/h0ffmann/closed-source/pull/27). - El Niño: NOAA Climate Prediction Center, "ENSO Diagnostic Discussion", 8 October 2026, , read on 14 October 2026 (the page is replaced each month). The rule: [exploit in operations, explore in simulation](../notes/202610122359-exploit-in-operations-explore-in-simulation.md). - Social media: [seen by the right people](../notes/202610130845-seen-by-the-right-people.md). - Bitcoin: Nakamoto, "Bitcoin: A Peer-to-Peer Electronic Cash System", abstract, , read on 14 October 2026. - Privacy: [`PERSONAS.md`](../PERSONAS.md) and rule 8 of [`SMART_NOTE_POLICY.md`](../SMART_NOTE_POLICY.md); [not against property](../notes/202610132128-not-against-property-copy-professionally.md). - "What do I know?": Montaigne, *Essays*, "Apology for Raimond Sebond", in Cotton's translation (Project Gutenberg, eBook 3600).