These repeated acts of public description adds each idea to a supersaturated, subconscious solution of fragmentary elements that have the potential to become something bigger. Every now and again, a few of these fragments will stick to each other and nucleate, crystallizing a substantial, synthetic analysis out of all of those bits and pieces I’ve salted into that solution of potential sources of inspiration.

The act of making your log-file public requires a rigor that keeping personal notes does not. Writing for a notional audience — particularly an audience of strangers — demands a comprehensive account that I rarely muster when I’m taking notes for myself. I am much better at kidding myself my ability to interpret my notes at a later date than I am at convincing myself that anyone else will be able to make heads or tails of them. Writing for an audience keeps me honest.

There’s a version of the “why writers should blog” story that is tawdry and mercenary: “Blog,” the story goes, “and you will build a brand and a platform that you can use to promote your work.” Virtually every sentence that contains the word “brand” is bullshit, and that one is no exception.

"The Curse of Recursion: Training on Generated Data Makes Models Forget," a recent paper, goes beyond the ick factor of AI that is fed on botshit and delves into the mathematical consequences of AI coprophagia: https://arxiv.org/abs/2305.17493 Co-author Ross Anderson summarizes the finding neatly: "using model-generated content in training causes irreversible defects": https://www.lightbluetouchpaper.org/2023/06/06/will-gpt-models-choke-on-their-own-exhaust/ Which is all to say: even if you accept the mystical proposition that more training data "solves" the AI problems that constitute total unsuitability for high-value applications that justify the trillions in valuation analysts are touting, that training data is going to be ever more elusive.

For people inflating the current AI hype bubble, this idea that making the AI "more powerful" will correct its defects is key. Whenever an AI "hallucinates" in a way that seems to disqualify it from the high-value applications that justify the torrent of investment in the field, boosters say, "Sure, the AI isn't good enough…yet. But once we shovel an order of magnitude more training data into the hopper, we'll solve that, because (as everyone knows) making the computer 'more powerful' solves the AI problem"

As the lawyers say, this "cites facts not in evidence." But let's stipulate that it's true for a moment. If all we need to make the AI better is more training data, is that something we can count on? Consider the problem of "botshit," Andre Spicer and co's very useful coinage describing "inaccurate or fabricated content" shat out at scale by AIs: https://papers.ssrn.com/sol3/papers.cfm?abstract_id=4678265 "Botshit" was coined last December, but the internet is already drowning in it. Desperate people, confronted with an economy modeled on a high-speed game of musical chairs in which the opportunities for a decent livelihood grow ever scarcer, are being scammed into generating mountains of botshit in the hopes of securing the elusive "passive income": https://pluralistic.net/2024/01/15/passive-income-brainworms/#four-hour-work-week

Censorship and Information Control During Information RevolutionsExploring how new information technologies from the printing press to the digital age have stimulated new forms of censorship and information control.

But there's another side to this playlistification of feeds: playlists and other recommendation algorithms are chokepoints: they are a way to durably interpose a company between a creator and their audience. Where you have chokepoints, you get chokepoint capitalism: https://chokepointcapitalism.com/

using rome as a almost a tool to convey information to your future self

I go through my old posts every day. I know that much – most? – of them are not for the ages. But some of them are good. Some, I think, are great. They define who I am. They're my outboard brain.

First and foremost, I do it for me. The memex I've created by thinking about and then describing every interesting thing I've encountered is hugely important for how I understand the world. It's the raw material of every novel, article, story and speech I write.

Those questions are hard, but they’re not wicked.

they totally broke the FCC’s regulatory model