HTML | TXT

Are some intellectual environments better than others? Yes.

ETA! states that E, the evolution time of ideas, is the time T needed to test alterations of ideas, divided by the factorial of the number of ideas in the Assembly Pool A!.

Longer evolution times means worse ideas last longer before evolving into better ideas.

June 29, 2024 — A child draws.

You take his paper.

He screams.

His scream is just: you stole his property.

A famous celebrity passes away and wakes up on a beach.

"Welcome to the Afterplace", says a man in white.

He extends his hand and helps her to her feet.

"You must be hungry. Let me show you to the Omni Restaurant."

May 14, 2024 — In America, AIs have more freedom to learn than humans. This worries me.

Do you want learn at the same library as ChatGPT, Gemini, Grok, or Llama?

Then you must become a criminal.

You have no legal option^[1].

by Breck Yunits

Newton, Darwin, and a modern Scientist go to heaven.

February 14, 2024 — The color of the cup on my desk is black.

May 26, 2023 — What is copyright, from first principles? This essay introduces a mathematical model of a world with ideas, then adds a copyright system to that model, and finally analyzes the predicted effects of that system.

April 28, 2023 — Enchained symbols are strictly worse than free symbols. Enchained symbols serve their owner first, not the reader.

March 6, 2023 — I believe Minsky's theory of the brain as a Society of Mind is correct^[1]. His theory implies there is no "I" but instead a collection of neural agents living together in a single brain.

We all have agents capable of dishonesty—evolved, understandably, for survival—along with agents capable of acting with integrity.

Inside our brains competing agents jockey for control.

How can the honest agents win?

January 27, 2023 — Today the trade group Lawyers Also Build In America announced a new file system: SAFEFS. This breakthrough file system provides 4 key benefits:

October 15, 2022 — Today I'm announcing the release of the image above, which is sufficient training data to train a neural network to spot misinformation or fake news with near perfect accuracy.

These empirical results match the theory that the whole truth and nothing but the truth would not contain a (c).

October 4, 2022 — Every second your body makes 2.83 million new cells. If you studied just one of those cells from a single human—sequencing all the DNA, RNA, and proteins, you would generate more data than can fit in Google, Microsoft, and Amazon's datacenters combined. Cancer is an information problem.

August 30, 2022 — Public domain products are strictly superior to equivalent non-public domain alternatives by a significant margin on three dimensions: trust, speed, and cost to build. If enough capable people start building public domain products we can change the world.

May 12, 2021 — Are you anti-copyright and anti-patent? Then this post is for you.

(If you are new to this issues, you might be more interested in my other posts on Intellectual Freedom)

February 28, 2021 — I thought it unlikely that I'd actually cofound another startup, but here we are. Sometimes you gotta do what you gotta do.

We are starting the Public Domain Publishing Company. The name should be largely self-explanatory.

If I had to bet, I'd say I'll probably be actively working on this for a while. But there's a chance I go on sabbatical quick.

The team is coming together. Check out the homepage for a list of open positions.

February 21, 2020 — One of the most unpopular phrases I use is the phrase "Intellectual Slavery Laws".

I think perhaps the best term for copyright and patent laws is "Intellectual Monopoly Laws". When called by that name, it is obvious that there should be careful scrutiny of these kinds of laws.

However, the industry insists on using the false term "Intellectual Property Laws."

Instead of wasting my breath trying to pull them away from the property analogy, lately I've leaned into it and completed the analogy for them. So let me explain "Intellectual Slavery Laws".

January 23, 2020 — People make biased claims all the time. A decent response used to be "citation needed". But we should demand more. Anytime someone makes a claim that seems biased, call them out with: Dataset needed.

Whether it's an academic paper, news article, blog post, tweet, comment or ad, linking to analyses is not enough. If someone stops at that, demand a link to a clean dataset supporting the author's position. If they can't deliver, they should retract.

January 3, 2020 — Speling errors and errors grammar are nearly extinct in published content. Data errors, however, are prolific.