arthurbrown

Thanks for the good faith reply. I can see that this would require regular maintenance.

It is interesting that we see our infrastructure as 'valid' or correct, but the work these engineers (beavers) put in as a nuisance. In an ideal world, the benefits to local biodiversity and ecosystem health should be balanced with the cost of relocating roads when necessary.

What are the problems actually? That they have improved water retention and soil hydration?

The author asks "How can people take this seriously?"

I would reply the same regarding this article. Nearly all of these refutations are unconvincing.

The claim: "I believe we're reaching the upper limits about what generative AI can do and how accurate its outputs can be."

The rebuttal: "Zitron's argument at the time was that hallucinations were as good as they were going to get, which meant that AI performance is capped at 2024 levels. Both the overall prediction and the mechanism were wrong. This one seemed wrong at the time, in that I noted here in 2024 that you can make AI code halfway decently by just putting it in a loop and having it run until the code compiles and tests pass"

Putting an LLM in a loop, burning tokens, and thrashing against a compiler and test suite is a ridiculous way to say that hallucinations have been "solved". Please. This is absurd.

Google Inbox was amazing and I still mourn it regularly. The features they backported into gmail do not capture its essence.

https://en.wikipedia.org/wiki/Inbox_by_Gmail

Another related but semantically distinct package with great developer ergonomics is the FRP library React -- https://erratique.ch/software/react/doc/React/index.html

Very satisfying to use when you manage to find a problem that is suited to this type of approach.

I appreciate the insight, and on closer reading the post clearly states that realistically only Zig and Rust were ever considered anyway.

Since you're here, could you comment on the approach Rust took in their rewrite? Was it more of a straight translation like Go did when they self hosted -- similar to the recent Bun transliteration? Or were there architectural changes made along the way like this article describes with Roc?

Interesting that OCaml was flexible and expressive enough to be used as a prototype testbed but not chosen as the implementation language, especially given the maturity of both. I would be surprised if Zigs incremental builds could be meaningfully faster than dune's.

Cross compilation is great, but not mentioned in the "why Zig" section. Is memory control that crucial for a compiler?

Rust itself was originally written in OCaml, same with WASM. I'm curious about what milestone gets reached where the maintainers collectively decide to transition away.

One form of JavaScript interop. Instead of being able to write bindings directly to native code, you had to pass messages through "ports" instead.

https://discourse.elm-lang.org/t/native-code-in-0-19/826

Personally, I was sad to see signals and FRP go in 0.17

https://elm-lang.org/news/farewell-to-frp

That chart stops in 2023.

The article doesn't really discuss education outcomes at all, that is something you brought up. It sounds like we agree that there is an issue here. I'm suggesting that there are other reasons that explain the performance drop.

My reading of the article is that it criticizes the implementation of this policy and the methodology behind it, which I agree with.

Living in a country that moved quickly from a "social media ban" to an "adult content ban" in the space of 3 months, I feel that these policies are overreach due to how they must be implemented. As in, they require all users to provide verification, not just the targeted cohort.

https://www.bbc.com/news/articles/cwy92qpv424o

Nationalized social media bans and restricted device usage while in a classroom are different things. Why would TikTok ever belong in the classroom?

Of course if even educational use of laptops is restricted then personal mobile devices would also be. They are already banned in my country.

There is growing acknowledgement that this is related to laptop usage in classroom. Countries are recognizing this and rolling back policies, citing PISA rankings.

https://www.bbc.com/news/articles/cly0vk77vdko https://www.theguardian.com/australia-news/2026/jun/15/educa...

I'm not a farmer, but you are welcome to ask a no-till farmer for their experience, or do some reading. Heck, you could read the article that we're commenting on where scientists have dedicated their career to understanding this stuff.

What's the point of this? You are saying that you will not accept any new information that goes against your belief system.

The evidence is there. Read something. Watch a video. The resources are readily available and abundant.

Make a garden patch and experiment for yourself if you refuse to accept any outside information.

This video is 15 years old. https://www.youtube.com/watch?v=q1aR5OLgcc0

What an odd response. We have centuries of evidence for minimal disturbance agriculture supporting civilizations.

What evidently does NOT work is the quite new practice of industrial tilling and fertilizer, which is causing rapid breakdown of our natural environment and future potential for food production.

As someone who's been observing how the media treats cycling in this city closely since the dark days of Ducan Gay, this is absolutely what is happening.

Thank you for posting this and for taking the time to document it in detail.

Bought any RAM lately? Phone? GPU in the last decade?

My understanding is that the high calcium content in their water supply formed a lining on the inside of the pipes which largely prevented any exposure.

Monopoly, network effect

So just misquoting the author intentionally then?

You should pick some of the substance of the article to take issue with instead of jumping to be a victim. They clearly read it if they wrote it.

You've read that paragraph backwards. They are talking about LLM enthusiasts dunking on people who aren't using.

What modifications to infrastructure are you anticipating needing?

Can you speak more to the difference between the two experiences?

People conjure various reasons for why things are different now, but routinely omit the changes in class talents and gearing.

At launch there was zero spell damage or healing on raid gear as a stat. You might find some green of shadow wrath but items like Robe of volatile power simply did not exist yet.

Warriors, the current top damage class, had a talent rework at BWL release which introduced their main damaging ability at nearly half the effectiveness of where it eventually end up.

Throw in world buff scheduling and the game is very different from how it played at launch.

They are planning to enforce access to these sites with the same mechanism.

https://www.esafety.gov.au/about-us/consultation-cooperation...

this is the syntax for variable binding in ocaml.

Hazel appears to be written in ocaml and mentions being "ml-like" on the site

Can't this line of reasoning be applied to any transport infrastructure?

Could you provide some more details about the Tasmania example? Seems humans reached Tas "at least" ~40,000 years ago from my simple searching, which isnt too far off from the generalised 50,000 mentioned in the article.

Which species went extinct before then?

https://www.sciencedirect.com/science/article/pii/S030544032...