It sounds like people are being paid to push Erlang for agentic stuff. Cool story, but if I end up feeling like I need to “return to the 20th century” for abstractions, I’ll just use Lisp.

This is the first thing that comes to mind. However I wonder if not only the “general” vocabulary can be anonymized but also the underlying concepts and references, because they point to a particular place too.

The worst: ads.

Noooo. Makes me wonder how much money do you need to buy up all the ad slots in the world and replace them with blanks.

So much money that only running your mega ad operation would allow you to cover the costs.

This is a plain admission that the administration has lost on the merits of their arguments.

I guess if you get downvote for saying something ironical, then that's the loss of audience, not you.

One can argue that a thoughtful irony which gets downvoted might be more interesting than thoughtful irony which gets upvoted because of the points you mention.

Irony shouldn't be in a bubble of all upvotes. Funnily enough I had searched up some irony quote websites a few days back and going back on them was fun to find a relevant quote:

Irony is just the honesty with the volume cracked up

- George Saunders.

Damn that’s good quote. And I would even go as far as saying that the apparent cynicism of the “honesty with the volume cracked up” is just a mirroring of the inversion that symbols suffer while transiting Culture.

In fact now I’m a bit worried that you can’t really have any sort of accurate feeling of the magnitude of the ironic statement that went over your head. You hear something that’s not true, you internalize it, you figure the character of it’s falseness, and when it’s your time to subvert it you invert it and amp it up creating something more absurd, which in turn will be internalized by even more people without its apparent falseness that you tried so hard to convey.

This makes me think that speaking absurd is the only way to convey truth. Not immediately, but eventually.

There’s also the other way around. Semantic AI is a good chunk of meat, but it can only be useful as it’s harnessed properly with a nice set of bones. I think that symbolic AI will make a come back eventually. Not as an accelerator of what’s already been done in the Industry, but as the actual revolution.

And don’t be naive to think that there aren’t sophisticated symbolic handling mechanisms being implemented in the training of the models by Big Tech. Not even baby soap is truly neutral.

As a possible example of this, I was kind of baffled how quickly we're all now throwing the sophisticated AST/program analysis and refactoring methods over board we already had before AI. Just look at the refactoring methods of Eclipse or IntelliJ.

I think those should be very useful, especially with AI: Either as a tool for the agents themselves - why spend heaps of tokens completely rewriting a code file, if you could do most of it by calling some global refactoring operations on the IDE's AST/symbol database?

Or side-by-side with it, to give human users better insight what the AI did.

Instead it seems to be all VSCode (if at all) + grep + AI agents, and nothing else.

> especially with AI

Yeah, the middle path sounds promising.

"Code Mode", where the AI writes a little program or script to do the AST/symbol transformations sounds like the win. As you point out, less tokens, and gives the humans insight.

This isn't exactly the same application of a "code mode" as before, but in my view it's a broad philosophy. AI for building machines, instead of doing the work directly. It also allows for easier updates/retries too. https://blog.cloudflare.com/code-mode/ https://news.ycombinator.com/item?id=45399204 https://blog.cloudflare.com/code-mode-mcp/ https://news.ycombinator.com/item?id=47089505

This is an interesting idea! I searched around and it looks like there's [ast-grep](https://ast-grep.github.io/), an AST-aware CLI that can search and refactor code -- and you can expose it to your AI agent using a skill (https://github.com/ast-grep/agent-skill).

Not exactly symbolic AI, but pretty cool nonetheless.

Very nice piece of text. For me it’s the fact that the book is always there and no amount of pages can be too much for me to gather something from rereading a part of the book that leads to where I left off. In a way I think we’re also past the point of bookmarking anything in our lives. Culture after all is just the reinforcement of the same idea.

It is impossible to enforce a world free of heuristics, but this is certainly very cool.

Reminds me of that Black Mirror episode with the circular QR code.

completely agree on the heuristics (someone else mentioned the MIT Koan comment about this). And yeah Plaything is a little too close to home... no QR codes from werld agents yet though. Will keep you posted.

I just wanted to add: there’s no single piece of machinery that can void the human experience. A large collection of machinery can only delay the inevitable. Please have fun with your project.

Agreed — and I wouldn't want to simulate the human experience. What I'm more curious about is what experience agents would create for themselves, with no concept of ours.

It looks to me as a consumer that the worst option for my consumption is asserted as the inevitable decision of purchase only by coercion.