What are your thoughts on a future where code is represented as a structured model, rather than text? Do you think that AI-powered coding assistants benefit from that?
Last Updated: 03.07.2025 01:46
Most coding assistants — with or without “modern “AI” — also do reasoning and manipulation of structures.
These structures are made precisely to allow programs to “reason” about some parts of lower level meaning, and in many cases to rearrange the structure to preserve meaning but to make the eventual code that is generated more efficient.
Another canonical form could be Lisp S-expressions, etc.
What’s a sleep paralysis demon? And other facts of the bizarre phenomenon - CNN
plus(a, b) for(i, 1, x, […])
Long ago in the 50s this was even thought of as a kind of “AI” and this association persisted into the 60s. Several Turing Awards were given for progress on this kind of “machine reasoning”.
+ for
a b i 1 x []
in structures, such as:
A slogan that might help you get past the current fads is:
Your Brain Has a Hidden Rhythm, And It May Reveal How Smart You Are - ScienceAlert
i.e. “operator like things” at the nodes …
NOT DATA … BUT MEANING!
First, it’s worth noting that the “syntax recognition” phase of most compilers already does build a “structured model”, often in what used to be called a “canonical form” (an example of this might be a “pseudo-function tree” where every elementary process description is put into the same form — so both “a + b” and “for i := 1 to x do […]” are rendered as
Pro-AI Subreddit Bans 'Uptick' of Users Who Suffer from AI Delusions - 404 Media
/ \ and ⁄ / | \
It’s important to realize that “modern “AI” doesn’t understand human level meanings any better today (in many cases: worse!). So it is not going to be able to serve as much of a helper in a general coding assistant.