shabbyrobe

Oh, didn't realise at first who it was replying! Thanks for weighing in to allay my concern, I appreciate it.

I adopted Pi for both of these reasons, and the strength with which they were stated gave me the confidence to lean in.

The rationale for MCP and codemode I can swallow, Armin's writing on that makes a lot of sense, and meeting models where they are seems critical to me based on my own experience.

But for me, fullscreen mode is a huge turn-off. I already have a backbuffer that I strongly prefer to use, it's called my terminal, and it's important to me. If the backbuffer-based version disappears, I'll be forced to migrate to something else (seems there are a few options here) or make my own tool, but that involves abandoning or porting my extensions, which turned out to be a huge superpower with Pi. Perhaps Pi Durable helps with that and lets me keep my extensions, but I'd rather this was not necessary in the first place.

If switching to full-blown TUI has anything to do with how slow the "expand thinking" and "expand tool output" features get as the session gets longer, could that be mitigated by only expanding the n most recent behind the default shortcut, and the slower "nah, I really do want you to expand them all thanks" can be a different shortcut?

If it's to add more fancy features that require fullscreen rendering and a dyed-in-the-wool terminal user like me might reasonably tolerate as a dismissable modal, make those bits TUI, but keep the backbuffer in the terminal (for e.g. the session tree or the settings, those don't need to be in the terminal's scrollback).

And if it's for any other, richer interactions... can we just... not, instead, and let the terminal be the terminal rather than a single page web app?

That's not what I found in my own experiments, I still had to unmarshal once inside the callback to get the `type` field out, then again once I knew what the type was. Do you have an example handy?

IME this is a longstanding pain point with Go. There's an attempt to propose an encoding/json/v2 package [1] being kicked around at the moment [2], spawned from a discussion [3].

This at least seems to improve the situation of marshalling to/from an interface directly slightly by providing the ability to pass custom Unmarshalers for a specific type (via json.WithUnmarshalers and json.UnmarshalFunc) to the Unmarshal functions, but it appears to still have the inefficient double-decode problem. Or I just haven't found a decent way around it yet.

Looks like they're intentionally punting on a first class solution until (if) the language gets some sort of sum type, but I still think the second-class solution could do a bit more to make this extremely common use-case more convenient. Pretty much every serious production Go app I've worked on in the last 10 years or so has had some horrible coping strategy for the "map a field-discriminated object to/from implementations of an interface" gap, often involving some sort of double-unmarshal.

Quote from the proposal [1]:

> First-class support for union types: It is common for the type of a particular JSON value to be dynamically changed based on context. This is difficult to support in Go as the equivalent of a dynamic value is a Go interface. When unmarshaling, there is no way in Go reflection to enumerate the set of possible Go types that can be stored in a Go interface in order to choose the right type to automatically unmarshal a dynamic JSON value. Support for such use cases is deferred until better Go language support exists.

  [1]: https://pkg.go.dev/github.com/go-json-experiment/json
  [2]: https://github.com/golang/go/issues/71497
  [3]: https://github.com/golang/go/discussions/63397

Such an instrument is the turbo encabulator.

> I would've probably picked the modernc variation

Heads up about the modernc library, it has been stuck on an old version of sqlite for several months [1]. It seems like maintainer time is the limiting factor [2]. There has been a call to arms on that issue page, the maintainer is looking for help, but it looks like not much has arrived. It seems like it might trace back to blockers in the C-to-Go compiler.

It's a major undertaking and a very impressive piece of work, but I'm not surprised it's a struggle when big roadblocks get hit. I hope they find a way to progress, but I'm very relieved to be seeing some CGo-free alternatives like ncruces/go-sqlite3 emerging. I'm going to give it a try for sure and see if I can live with the compromises.

Squinn-go looks very compelling too, but I don't like that it requires the squinn binary to already be installed on a user's machine, I think that gives with one hand and takes with the other: sure, I get to avoid CGo, but I also lose the turnkey, single-command install, static build benefits Go brings out of the box.

Seconding the point about nitty gritty, I'd read it for sure too!

  [1]: https://gitlab.com/cznic/sqlite/-/issues/154
  [2]: https://gitlab.com/cznic/sqlite/-/issues/164

I love this visualisation. Is the script to generate it publicly available?

Since PEP 591 [1], if using mypy with Python 3.8+ or the typing_extensions module, you can also take advantage of typing.Final, which lets you statically verify something isn't changed. The catch is that it isn't enforced at runtime.

  [1]: https://www.python.org/dev/peps/pep-0591/

I think it could be the same but it doesn't have to be. "What problem does this solve" is a bit more generous as at least, on the face of it, it may offer an opportunity to provide an answer that will be received in good faith.

But I think this is less about specific phrases one might need to be careful not to say, and more about the kinds of default attitudes that can turn into ongoing impediments to the free flow of ideas. It's definitely possible ask "what problem does this solve" without becoming the "What Problem Does This Solve Person", and the same goes for the other phrases too.

There's also the "That Won't Work" person's sibling: the "I Don't See a Use Case" person. In my experience, "I Don't See a Use Case" people require more sophisticated tactics if you actually want to get something done while they're around, not least of all because they move the game to a harder field: "that won't work" at least states a fact, whereas "I don't see" moves us firmly into "belief" territory, and "a Use Case" invites us into the trap of dreaming up our own strawmen for them to tear down one by one.

I wonder if they started out as "That Won't Work" people, but evolved better defenses?

I really didn't enjoy the game with the biters turned on, but absolutely loved it when I switched to peaceful mode. There are enough problems to sort out without them, like "why do I suddenly not have enough iron plates", or "why do I suddenly not have enough copper plates", or my personal favourite, "why do I suddenly not have enough iron plates again"!