I have the opposite experience where Gemini (even the flash models) has the only useful model for my reverse engineering related use case. My hunch is Google utilizes its free access to entire Google search indices to train itself from niche non-English speaking community websites, much frequently and in a "relevant" manner, which in the end gives these models the most up to date info for this particular kind of work. Every other model is just either 10 years outdated with their answers or simply hallucinates like waaaay crazy.
> for my reverse engineering related use case [...] Every other model is just either 10 years outdated with their answers
I've mostly been doing reverse engineering with Codex, mostly related to games, but not once has the "training data cut-off date" been in the way, the most useful part comes from handing it a binary/directory and letting it prod it until it finds the answer you're looking for, I don't even have web search enabled and sometimes it might take 30-40 minutes for it to find the answer, but I never saw it be unable to find the answer because it's training data was a couple of years old.
That analogy may not be suitable for this case because value proposition between the aesthetics vs the function is different for visual art projects compared to software. There is also the maintainability factor where most aged software (especially the closed source ones in private sector) change maintainers every few years. Old maintainers most often lose access to the source code and become unreachable after leaving their job.
True, there is the "old maintainer" aspect, that differs.
But what exactly does this mean in relation that what are special forms in other languages are function calls in Rye?
Is the problem that you somebody could make their own control-structure-like functions? All these function calls have exactly the same evaluation rules, which is not something you can say about "special forms" in "normal" languages because special forms are exactly rules that break the regular evaluation rules.
Rye already has a lot of control-structure-like functions in it's standard library and many other functions that accept blocks of code and aren't related to control-structures. Yes, a "stupid" person can write "stupid" code in Rye, but you don't need much flexibility in any language to write stupid code.
I fully admit there are languages that are more suited for teams, and languages that are more suited for solo developers, for this reason.
That was also my point mentioning "million pixel website" because that "art" is directionless and crowdsourced and a painter can use the same pixels or even more flexible options to create beautiful images.
> I'm not quite sure why you'd want to run Erlang on it, but the hardware exists.
Erlang is invented before IoT was a thing to facilitate distributed computing for telecommunication in a highly reliable manner. It makes perfect sense to adapt it for driving fleets of cheap IoT devices.
Low resource footprint, written in Go, embed-able in any Go project as a library, compiles to mobile with little to no modification, supports config change without restart, has plugin API.
These were the reasons why we used it in my previous job.
It would be "artificial" only if LLMs performed badly despite having an equal amount of data containing examples of eastern customs in its training set. Even that's arguable since we don't (didn't) have the benchmarks for this particular case before.
