Post #1407449

2026-03-07 16:21 UTC

@cross @jschuster @lindsey @krismicinski @jfdm @csgordon @jeremysiek Oftentimes, it is only by interacting with this program that we realize what spec we really meant. (It's the only time I got to quote TS Eliot in a CS research paper: "That is not what I meant at all. That is not it, at all.") Which is why it's a *loop*. And until recently, it took ages to go from ideation to working program (for mortals like me, anyway). Without the loop, we can't write particularly good specs. ↵

Replies (2)

@shriramk@mastodon.social 2026-03-07 16:22

@cross @jschuster @lindsey @krismicinski @jfdm @csgordon @jeremysiek That's why in SE there has been a long-standing dream of "executable specifications" (set aside the ontological questions): to tighten up this loop. So I view this as being part of a long chain of what we've always known/wanted. But also, when you make something 10x or 100x more efficient, it turns into a different thing. And that's part of what we're seeing, too.

Open ##1407450
@jschuster@hachyderm.io 2026-03-07 16:41

@shriramk @cross @lindsey @krismicinski @jfdm @csgordon @jeremysiek yeah I would have sworn I saw him make a diagram like that too, but I don't see it now in the World and the Machine paper. Thinking back to that era: maybe we'll see the resurgence of parts of UML? Perhaps this new loop is the right feedback loop to make some of those ideas work

Open ##1407451