alt.hn

6/7/2026 at 1:17:40 AM

Passing DBs through continuations

 https://remy.wang/blog/cps.html

by remywang

6/9/2026 at 5:01:06 AM

Long before AI psychosis, there was FP psychosis, clinically defined as the intense psychological response to understanding functional programming concepts like recursion, higher order functions, monads, or in this case, continuation passing style.

by tenwz1

6/9/2026 at 11:25:57 AM

Worth noting that’s a quote from the article (footnote 1).

by antonvs

6/9/2026 at 7:27:32 AM

At least FP psychosis leaves you more educated and with more tools in your box.

Not sure what the AI one gives you at a personal level.

by epolanski

6/9/2026 at 12:48:55 PM

A lobotomy? I've witnessed several individuals who were entirely dependent on Claude for any technical issue.

by icedchai

6/9/2026 at 11:28:41 AM

As a happy sufferer from FP psychosis, I can tell you that LLMs can be a fantastic tool for learning if you choose to use them that way.

by antonvs

6/9/2026 at 4:44:22 PM

You're not describing AI psychosis tho.

by epolanski

6/9/2026 at 8:01:19 PM

Meh. The term "psychosis" is a deliberate troll. If "FP psychosis" is understanding that imperative programming is a ridiculously primitive misunderstanding of computation that tends to discount the time dimension, it implies that "AI psychosis" is a similarly biased, nonsensical take that only matches the incorrectly perceived reality of the people using the term.

As far as I'm concerned, it's all just Luddites v2.0.

by antonvs

6/9/2026 at 6:22:54 PM 

  > "Suppose you want to write a database. You'd probably start by implementing relational algebra operators — projection, filter, join, etc. The easy way is to implement them as functions that take in tables and return tables, and assemble them into a larger expression. That was how Prela worked in its first incarnation. The code was clean, but it was hella slow! Which was not surprising, because every operator materialized every intermediate result. "
This is one of the LAST things you do when writing a database.

DB development starts with the storage engine, file manager, buffer pool (page cache), and page access methods (heaps/indices) which are binary buffer views. Then, you add the transaction manager, the WAL/recovery bits.

The actual implementation of relational algebra and a SQL language + parsing are little icing layers on top of a transactional storage engine.

by gavinray

6/9/2026 at 7:00:20 PM

This. But TFA was specifically concerned with the relational algebra, so I'm giving it a pass :)

by cryptonector

6/9/2026 at 5:32:56 AM

isn't this https://clojure.org/reference/transducers?

by unrealhoang

6/9/2026 at 7:00:05 AM

Continuations predate transducers by some ~40 years and are mostly used as a means of control flow, but yes they are very similar.

Transducers are specialized to data transformation pipelines, continuations are a form of control flow from which you can derive a lot of cool things (exceptions, time travel debugging etc).

by fredrikholm

6/9/2026 at 4:05:22 PM

Thank you! This is the concept I've been looking for. I always had the sense that compositional data transformation frameworks like PRQL or LINQ should have a clean categorical interpretation. I was exploring how this could be expressed in terms of monads but they usually start with the data while I was trying to formalise the composition of the transformations. I think transducers are it.

Transducers are just the morphisms in the category of the reducing-functions. No problem!

by snthpy

6/9/2026 at 5:28:03 AM

CPS is a way of embedding imperative computation into an FP language. I think they built a mini compiler for their binary relation language, which is then Jitted by Julia.

by DevelopingElk

6/9/2026 at 11:48:12 AM

I'd rather compare CPS to goto than regular imperative computation.

by IsTom

6/9/2026 at 3:26:59 PM

The continuations in CPS are closures. Goto basically isn't. GCC's computed goto is, but generally when people say 'goto' they mean the traditional C goto, which involves no closures. The goto analogy is not great for this reason.

A better analogy is that continuations are reified function call return addresses, since return addresses come with a frame pointer (explicit or implicit), and therefore are closure-like.

by cryptonector

6/9/2026 at 8:46:19 PM

A function call that's not artificially restricted to return to its caller is equivalent to a goto. See "Lambda: The Ultimate GOTO" by Guy Steele.

A continuation is a value that's passed to a function to tell it where to send its result when it's complete.

In imperative programming languages which invariably have restricted function calls, the continuation that every function receives is the address following the function call. This was just a mistake which the earliest programming languages committed, which has been perpetuated ever since, except in functional languages.

by antonvs