For the curious, there is an official YouTube channel with the previous editions https://www.youtube.com/@racketlang/playlists (It usually take a few months until the videos are edited and uploaded.)

I've been messing around with a language that I summarize as:

ALOE = Scheme + Smalltalk + Types

https://github.com/dharmatech/2026-09-02-aloe-racket

Prototype is in Racket.

Cool project, thank you for sharing. I've been thinking about something similar lately.

However. "Send, not apply" - sigh.. how shall I put it.. let's try this - eyes_rolling_so_hard_they_fall_out_of_sockets.png

you lost me at the receiver selector order.

It's a common way to implement OO in Lisp-y languages, and it fits if you're taking Smalltalk as an inspiration. SICP has an example in chapter 3 (IIRC) of doing the same, but the selector had to be quoted as in:

  (my-account 'balance)
  (my-account 'withdraw 10)

quite elegant, feels like a great balance idiomatic scheme and OO

RacketCon is Saturday https://con.racket-lang.org

> With AI, we have converged on a set of languages. For the web, it's JavaScript or TypeScript. For backend, it's Go and Rust.

what a disaster we have ahead of us not because AI but because this kind of thinking

My case for why Elixir is especially suited for use with AI: https://news.ycombinator.com/item?id=49835781 (summary: it's mostly about the BEAM, which is more of a k8s killer than a rust killer).

I don't think this really applies to any of the benefits I've described there:

> AI can replicate a lot of its features in Rust for one-tenth of the cost and 10 times the speed

The force of convergence should almost make us care more about other languages, since the explosion of software is a multiplier on good/bad designs, and we should consider what we want our AIs to be using.

Aye--even if you've reached the point of thinking that the only role of a human is to feed good ideas into the machine, language is a thing that shapes thought--Lisp as a family is a thing different enough from a lang wanting to be a better C/C++ or a better Java that getting your brain thinking in terms of it can help you produce a different set of ideas (some of which may even be good ones). For me personally, it's really opened my mind to functional programming and intelligent use of recursion, in a way that I probably wouldn't have ever had cause to consider if I only ever concerned myself with the minimally convergent set of standard business langs, in the way that all our legacy code had been using them.

Of course, Lisps also seem to have the property of attracting people who genuinely like the practice of programming and, if it clicks with you, it has the property of generally leaving you really excited to write more of it. If you're hauling up and spending real American dollars to spend a weekend packed in a hall full of nerds to chat about a niche programming language, you're probably in that group. Consequent to that, it'd be a mighty small-minded thing to wonder what the point of joy is, when you could just be managing an agent swarm to write more enterprise Go code.

Industry takeoff would certainly be a nice-to-have. But the racket community is largely computer scientists, who program for research. Will computer scientists turn to Rust to write papers because it's fast ? I don't think so, because their needs seems to be a language that's easy to pick up and experiment with, and writepapers with. People will argue over sexp syntax, but that's why Rhombus exists.

> Pretty much nothing else can get picked by AI if you start a new project.

You can pick literally almost anything, and current models will be able to write it. I don't know how you arrived at that statement.

Hygenic macros and the DSL they enable are fantastic ways to prototype things that then inspired stuff like LINQ.

There are enough AI stories on Hacker News. We don't need AI advocates coming to the stories that aren't about AI and saying they don't matter because they're not AI.

Things that aren't AI still matter and are still interesting.

This. For real I can't take it anymore. What to do?

> What to do?

Flag the twits that bring it up in every thread. They're trolling at this point, not actually trying to have a conversation.

I think I’m ready for an internet break

Oh thats such a killjoy attitude. There is more to programming then "production", "shipping" and the overall corporate lala. You said it yourself, you get a lot of joy from designing your own programming language. And you know, even BEFORE AI it would have been unrealistic to see your langauge actually take off big time. So you answered your own question. Racket is likely not interesting to one of the big coprs. But you know, frankly, who cares? Some people use AI and have it write only 4 languages? Sure. Go ahead, and you know? Who cares? Let them have their monoculture. Such a sad sad sad comment.

Well, taking it to extreme why care about any other than machine code?

As AI programming tools improve, we will have the second coming from what 4 GLs, RUP, CASE tools, and Japan's Fifth Generation Computer Systems were trying to achieve, each on its own way as the next evolution on programming abstractions.

LLMs are able to write racket. I have a fairly large project I've been vibecoding for the last 6 months. Models predating deepseek flash 0731 used to make paren mistakes and had a hard time. But now Opus 5.5 for example does not make any mistake writing clear racket code.

I wouldn't say we've settled on a specific set of languages

To add to this, I definitely wouldn't say we've converged on a set of languages in general either! Most of my backends are written in Scala of all things, and LLMs have no issue with it. If anything, the expressive type system is helpful for them.

> To add to this, I definitely wouldn't say we've converged on a set of languages in general either!

Absolutely agree. I’d take a guess that a decent chunk, if not majority of “business” applications probably still exist in either Django or Spring Boot. Should all of those be rewritten now “just ‘cause”?

You forgot about Python.

Have you found LLMs to be good at metalinguistic abstraction? I have used them a little bit for lisp programming but never once have I seen them introduce macro.

I'd imagine they would require some hands-on guidance to create more meaningful abstractions and DSLs for humans.

Call it "vibe coding by tiller". Ha.

Also, I think of metalinguistic abstraction as something deeper than macros. At least not limited to macros:

"An important advantage of making the evaluator accessible as a Lisp program is that we can implement alternative evaluation rules by describing these as modifications to the evaluator program."

https://sarabander.github.io/sicp/html/Chapter-4.xhtml

I explicitly asked LLM to write abstract away with a DSL and it did that. LLMs are good at this too.

If someone posts a story about topic Y and a reply is "does anyone really care about topic Y?" I would expect it to be flagged and downvoted.

The answer is "yes". Obviously people still care. No need to resurrect a dead bad post.

OP is apparently working on a new language and would like some assurance that someone out there might take an interest in it someday. That doesn't sound unreasonable to me.

so-cal-schemer attempted to bypass HN moderation by reposting a flagged post in its entirety without waiting for the "vouch" process to complete.

Meanwhile, behnamoh's post was essentially dismissing the idea that we should discuss non-AI topics in the age of AI. His mention of his hobby project doesn't redeem a garbage comment.