I came to say something similar: if a flaming head fell into the cooling tower, it would just fall down to the bottom surface and not interact with anything more explosive than a steam pipe (on only then if it impacted the inside of the tower.

The flaming head in the comic continues to burn without fuel or oxygen, but a normal flame might even be put out by the intake breeze at ground level.

Large dynamic range is a good thing, in audio and video. There are a number of ways to reduce dynamic range, if you really need to. Expanding dynamic range after capture is possible, too, but more fraught.

Shannon information theory gives us ways to perfectly reconstruct the lower range version from an expanded range version, but it also guarantees that reconstructing the higher range version from a reduced range version cannot be perfectly accurate across all inputs. (Assuming a fixed output range capability.)

I suspect some “leader” at apple is the first person that decided that two different ways to use one button is better than two buttons.

Whoever they are, I hope they suffer from confusing buttons until they repeat and remediate.

Public statements like this is why EA continually wins “Worst Company” over real assholes like Blackrock.

Bah, what I really need to know is: What are the difference between this and Miley Cryus Linux?

^{someone had to say it} /s

Sony rootkit or before.

Sphinx of the black quartz judge my vow.

Yeah, GRTT wasn’t really invented until 2021. Of course it borrows from many other theories, but it is not easy to fit under existing calculi (tho it can be “lowered” to MLTT).

I want a language with an internal/core calculus of GRTT with a specific grading that I believe will provide both optimal evaluation and prompt resource release. You aren’t going to get that by building on top of another language.

Of course you could do it as a library for another language, but the few values you could lift/lower and the necessary analysis do to bidirectional type inference/checking mean that you are writing a compile phase whether you call it that or not. In theory, you might save on the tokenizer/parser, but those are trivial to write. You might be able to reuse parts of the API/ABI, which could be an advantage if the underlying language has a good, stable one that can reflect your linking/calling/passing restrictions; but that’s not likely.

👎 : Ranch instead of Soy Sauce

👍 : Ranch instead of spicy mayo

NGL, I will be trying the ranch+sushi combination myself.

Retiring? Or changing careers?

I don’t think I can retire right now. I hope you are doing better than I.

Agreed. I don’t think it should be called typing; it’s more tagging (yes, like tagged unions). But, I don’t think I’ll see that semantic shift in my lifetime.

Anyway, I prefer my types static and my functions curried, which is why I do PureScript, Idris, and Haskell for my personal projects.

ty

If you read the literature, particularly “Types and Programming Language”, you’ll find that “dynamic typing” isn’t even considered typing. If you can have a type error at runtime you’ve defeated the reason to add a type system: to reduce runtime errors. The hope is that “well-typed programs don’t go wrong” tho there are some limits to what any type system can do (e.g. Rice’s Theorem).

That said. Static v. Dynamic is much more precise than Strong v. Weak and should be preferred.

Implicit v. Manifest is less useful just because it’s a broad spectrum, basically inculcating how much type inference is done and “how much” generally depends a lot the input program(s). Haskell does whole-program inference, tho GHC (the only Haskell compiler) has a number of syntactic forms that can’t be inferred. Scala only does local inference. C doesn’t infer types, though is gets close with how it treats functions with no-argument in the prototype and varargs stuff. C++ uses auto for some type inference, which is still somewhat manifest, but also mostly implicit.

I think weak typing is a good name for when there are invisible coercions, but that doesn’t actually have much to do with proper types.

Anyway, great comment, If more people would use static/dynamic and explicit/implicit instead of strong/weak, there’d be less miscommunication.

I’m not sure if any of those commits did, but I worked 2 years doing Haskell backend and jQuery+Bootstrap frontend for Masterword. Before that I did some work for Wire: https://github.com/wireapp/wire-server/commits?author=stephen-smith Before that, I worked at TGCS for nearly a decade.

But, right now I am still looking for work. I’ve actually been looking since April 2025 when my time at Masterword ended, but I did update my “hire me” posts back in February because my living situation changed, opening up more opportunities. Originally I needed something remote so I could continue to take care of my father (RIP), now relocation is certainly an option.

If anyone reading this has a position available, or is just curios, my current resume public: https://docs.google.com/document/d/1ORf48b557nK0gYI4h6J0xP7nzOVRoqtOogYsukUMREU

Yep. https://gitlab.com/bss03/halogen-lambda/-/commits/main

Not an issue, https://gitlab.com/bss03/type-aligned/-/commits/main

for real devs. https://gitlab.com/bss03/nested/-/commits/main

Usually it’s because we have a mutual desire to communicate, collaborate, or exchange data.

When it doesn’t affect me, I generally don’t think about it, and I certainly don’t “hate on” them.

I do sometimes feel bad for MS Windows users. I rarely “hate on” them, but that largely depends on their reaction my inability / unwillingness to use the communication/collaboration tools and data formats they are used to.

If you have a family you can divvy up those tasks to all care for one another. You can make a fine meal without a lot of specialist equipment. One sharp knife, one big spoon, and one pot can make a good soup with the right ingredients.

When I had tipped peers, wait staff got $2.15/hr + tips. It certainly changes the calculus.