Sven Slootweg (soft-deprecated) @joepie91@pixie.town

@roberth @arianvp

> so that we can temporarily give it dark powers

That's also the sort of thing I was thinking about, yes. To get *really* good error messages, that can be independently developed and improved from within the module system (without requiring upstream review for every change, which would make this nonviable) it would need pretty wide-ranging access to the runtime semantics.

That wide-ranging access could (and probably should!) be restricted to error conditions, though, to ensure that typical evaluation semantics don't get modified by 'clever' hacks too much.

> I think you're referring to a past team that never really got off the ground.

Possibly. I'm wary of promises of improvement due to those promises having been made several times in the past, and it turning out that many of the same systemic issues were still present. The bus factor was a major issue, true, but not the *only* one either.

So this has become an "I'll believe it when I see it" sort of thing - I can certainly be convinced that things have improved, but only through sustainable visible improvements, not just promises and ideas :) Trust takes time and effort to regain.

@follpvosten While this isn't Ryanair, the Dutch Railways make the exact same recommendation; in their case it's because the app disables the NFC functionality while scanning the ticket off the screen, as they use dual NFC/optical reader units, and they don't want you to accidentally activate the digital wallet on your phone and such.

The need for an app can be avoided in that case by turning off NFC manually when scanning the ticket. I don't know for sure what Ryanair's reasoning is (and whether it's just a ploy to get you to install the app) but perhaps it's something similar?

@arianvp So we "just" need a well-considered set of introspection capabilities with sufficient coverage of language semantics, basically?

(I say this with a feeling of dread about Nix core getting this kind of stuff right... 😐 )

@arianvp Hmm. Could this hypothetically be addressed by exposing more introspection capabilities for the language semantics, so that the eDSL evaluator can make use of the host language's features but still have sufficient insight into and control over its semantics that it could produce credible error messages?

@arianvp I'm actually kind of wondering, is there anything about Nix that makes this fundamentally impossible to fix, even with weird hackery?

@bamfic @zens@merveilles.town While this is true, what is also true is that *any* unconventional license scares the shit out of many corporations. I've seen much the same response when using the WTFPL, which from a legal perspective is pretty much the polar opposite of the GPL. So I don't think you really need the GPL or even any copyleft license to achieve this result.

(I also slightly prefer it over stuff like the pirate license above because it has the same uncomfortableness effect on Respect The Law people due to being very-not-legalese, but it still is technically a permissive license and so won't get in the way, legally or otherwise, of anyone who can be Reasonable about licenses)

@zkat I've been trying to have that RMS discussion with people quite a few times over the years and yep, every time people immediately started desperately searching for reasons why it could be dismissed.

I was reading the introductory section of the report which pretty much boiled down to "I have taken away any kind of ignorance you could try to hide behind, because I am sick of people doing this, you can no longer pretend not to know" and I rarely read things that resonate with me that much.

@eniko (is this re: the RMS thing, or did I miss something else?)

@5225225 Wouldn't bother with public-domain-trained LLMs because they consistently perform like crap from what I've seen (the whole 'tech' basically cannot work without mass nonconsensual scraping of data and conversations).

Perhaps something from the ELIZA era? Essentially just pattern matching and perhaps rudimentary knowledge graph processing, which can produce pretty acceptable results AFAIK.