595 karma · joined November 18, 2014
Find me at http://shonfeder.net/
What?
> I don't think OCaml is ready for production
seems to indicate your thinking is just not based on fact. This position is further belied by the stack of successful production applications you can see at https://ocaml.org/industrial-users/businesses.
Prediction is obviously involved in certain forms of cognition, but it obviously isn't all there is to the kinds of beings we are.
> ...
> Microsoft "invests" in Openai by giving the company free access to its servers. Openai reports this as a ten billion dollar investment, then redeems these "tokens" at Microsoft's data-centers. Microsoft then books this as ten billion in revenue.
> ...
> Plan for a future where you can buy GPUs for ten cents on the dollar, where there's a buyer's market for hiring skilled applied statisticians, and where there's a ton of extremely promising open source models that have barely been optimized and have vast potential for improvement.
No. That is a result of bad software engineer practices and stacks, not a symptom of proper abstraction.
[0]: https://hai.stanford.edu/ai-index/2025-ai-index-report/econo...
Sand away! Enjoy copying and pasting your nails, or having LLMs apply your varnish or whatever. I hope it brings happiness.
The ways AI is being used now will make this a lot worse on every front.
In OCaml we can (and do) also manage effects via monadic style. However, we don't have ad hoc polymorphism (e.g., no typeclasses), so that aspect of the dependency injection must go thru the more verbose (but more explicit, and IMO, easier to reason about) route of parametric modules.
The point in the article is that effect handlers allow direct-style code with custom effects, which enable a kind of "dependency injection" which actually looks and feels much more like just specifying different contexts within which to run the same program. If you are very used to doing everything in Haskell's monadic style, you may not really notice the difference, until you try to compose the handlers.
Here is an example I've put together to show what the author is talking about in practice: https://gist.github.com/shonfeder/a87d7d92626be06d17d2e795c6...
Is the most pressing problem facing the world is that we are not doing enough things fast enough? Seems a bit off the mark, IMO.
Current industry uses are largely in specialist areas including compiler engineering, static analyses, formal verification systems, and systems programming in critical domains.