Sharlin

14 ms·

421.
by Sharlin 5mo ago
To be honest, there’s approximately zero reasons to teach major-grade math to just about anyone but math majors. None of the applied math disciplines need go that deep, and what they do need depends on the field (physics is all about analys
422.
by Sharlin 5mo ago
I wasn’t aware of the degree of disparity in early-onset ischemic heart disease, thanks. But it doesn’t seem to me it’s "mostly heart disease rather than risky behavior", more like those are both major causes of excess mortality a
423.
by Sharlin 5mo ago
Well, that’s because for math majors any monetary incentive is nonexistent (modulo some rather specific careers). Just about nobody majors in math for any reason other than math itself.
424.
by Sharlin 5mo ago
Markov chains give what look like sentences. People in the frigging 1950s assumed their primitive NNs would be able to talk any day now. Transformers are clearly a big deal, but GPT-1 wasn’t exactly earth-shattering.
425.
by Sharlin 5mo ago
On the other hand, my guess is that male HN readers are not a very representative sample in this respect. That is, they (we) are significantly less likely than average to engage in the type of risky behaviors that mostly explain the gender
426.
by Sharlin 5mo ago
Yes, it’s definitely not a trivial problem. But parinfer shows it can be done mostly non-annoyingly at least for lisp; what is less clear is whether it would generalize to languages with more syntax.
427.
by Sharlin 5mo ago
Such a strange attitude. If a tool is clumsy, we try to improve it, that has been the case since the first stone artifacts created a million years ago. Do you think that the (sort of) tree-based affordances that most modern code editors do
428.
by Sharlin 5mo ago
Looks like it’s written in Rust and the examples are also in Rust :) Apparently it supports any language with a tree-sitter parser, which makes sense, tree-sitter is basically made for this.
429.
by Sharlin 5mo ago
Indeed. The incredibly clumsy way we choose to edit source code has baffled me since I was first acquainted with parinfer. Having to keep the source in the shape of a valid AST almost entirely manually is really annoying.
430.
by Sharlin 5mo ago
Good point.
431.
by Sharlin 5mo ago
Agreed. It just doesn't occur to most people. To even come up with the idea that maybe there's a setting for something, never mind searching for a tutorial on how to change it, you already have to be a power user for some values o
432.
by Sharlin 5mo ago
Unfortunately it seems that we as a civilization never learn anything except by trial and error, and are then entirely convinced that nobody could’ve predicted what happened even though many had done just that.
433.
by Sharlin 5mo ago
Their point, I believe, is that someone (just about any younger person in 2026) who has never seen indexed color modes, or colors taking less than one byte per pixel to encode, could reasonably be confused by the notion of "two bits pe
434.
by Sharlin 5mo ago
Because all this is mere extrapolation of trends seen in the 90s and before. Sagan was no psychic (or alone in predicting these things), he just knew where to look.
435.
by Sharlin 5mo ago
The problem with monads is their composition. Ie. questions like how is the do notation supposed to work if I want to return: * an Option<Async> or Async<Option>? * both an Option and an Async (a product, or tuple type)? * eithe
436.
by Sharlin 5mo ago
(2019) It’s worth clarifying that for the most part, this article just discusses plain effects, ie. "reified side effects" or "resumable exceptions". Algebraic effects are about the composition (ie. algebra) of effects
437.
by Sharlin 5mo ago
That’s exactly what supply and demand is. They didn’t let you work for them because they had a huge pool of candidates to pick from. It doesn’t matter what the cause of the s/d imbalance is.
438.
by Sharlin 5mo ago
That’s a tiny and decidedly nonrepresentative sample of gamedevs.
439.
by Sharlin 5mo ago
It incentivizes subscription-based games.
440.
by Sharlin 5mo ago
The vibe I got from the article was that now that technical work is faster, bosses expect everything else to be proportionally equally accelerated, even though LLMs either don’t help at all at those workloads or actively make things worse.
441.
by Sharlin 5mo ago
All right… So, the cost of technical work has gone down, surely leaving more time for things like coordination, leadership, and thinking on a more operational level, right? > The cost of building has collapsed, but the cost of aligning o
442.
by Sharlin 5mo ago
> A lot of programmers may not know this, but frontend used to be a highly specialized skill, requiring knowledge of semantic HTML, CSS, the differences of various browsers, accessibility, progressive enhancement, network performance, in
443.
by Sharlin 5mo ago
Most all perceived improvements in a minor version release are going to be solidly in the realm of confirmation bias by now.
444.
by Sharlin 5mo ago
Well, rockets are more than a millennium old, but sure, solid fuel rockets tend to be less volatile by definition. Honestly we’re really good at not prematurely combining tens to hundreds of tons of high-energy fuel and oxidizer put right n
445.
by Sharlin 5mo ago
In other news, water is wet.
446.
by Sharlin 5mo ago
I don’t think most people responsible for buying these things think like that at all . They know about ChatGPT but don’t have any idea how it could be made to do technical things for them. Normal people just don’t think like that.
447.
by Sharlin 5mo ago
The continuous/tracking/predictive AF modes of Canon’s EOS (D)SLR cameras were famously called "AI Servo" and "AI Focus", terms coined somewhere in the late 80s I believe. The early implementations were simple
448.
by Sharlin 5mo ago
ML is, obviously, AI. But not all AI is ML. The reason the term "machine learning" was even invented was because it was one of the AI winters and an euphemism was needed because "AI" was more of a swearword than a buzzwo
449.
by Sharlin 5mo ago
To be fair, that’s not exactly a new thing, it’s just sensitive to the exact phase of the Great AI Freeze/Thaw Cycle. A lot of now-ordinary automation used to be "AI" until it become commonplace and no longer buzzword-worthy
450.
by Sharlin 5mo ago
Even assembly is a high-level language relative to what’s actually going on inside a modern CPU.

More ›