Hacker News: advael

New comment by advael in "Google Gemini routes personal Drive files and emails to human annotators"

advael — Tue, 10 Nov 2026 02:38:46 +0000

Google and the companies it considers its competitors in a given space (e.g. primarily Microsoft in "workspace" and cloud storage, for example) have already demonstrated for over a decade that they're willing to use any data you store in their services for any purpose they feel like, and often much of this is written into EULAs, but also it doesn't seem to matter if it isn't. I hope that if this is your first exposure to this basic fact, it encourages you to stop trusting google and look for alternatives with auditable guarantees

New comment by advael in "From Linux to Wikipedia: How Volunteers Build the Internet"

advael — Fri, 06 Nov 2026 07:16:11 +0000

I dunno the leadership of at least US tech has been trying to market themselves as a cabal for a bit now and are doing a more convincing job of it than I thought would be possible

New comment by advael in "Ask HN: What cryptography book do you recommend for reading with 10 year olds?"

advael — Wed, 04 Nov 2026 22:13:08 +0000

Introduction to Modern Cryptography by Katz and Lindell is from a pedagogical perspective one of the best-structured textbooks I've ever read on computer science topics (and in general, but my exposure to textbooks is a biased sample). It goes into the foundations of cryptographic theory and gives a good basis for understanding and contextualizing even advances that have been made since its publication. I think most children can grasp difficult material if an adult that's good at the meta-skill of teaching is helping them with it via individual attention

New comment by advael in "Nvidia RTX Spark encrypted firmware slows inference speed to 35% on Linux"

advael — Wed, 04 Nov 2026 21:49:09 +0000

In general, the history of linux compatibility for nvidia specifically has been the research and open-source communities dragging them kicking and screaming into compliance while they do significant work to prevent compatibility via fuckery. I've been doing work impacted by this continuously since before GPGPU programming became popular. I'm not a beneficiary of nvidia's deals with microsoft or their stock price in general, so I see no reason I should "let them cook"

New comment by advael in "Nvidia RTX Spark encrypted firmware slows inference speed to 35% on Linux"

advael — Wed, 04 Nov 2026 20:48:33 +0000

Yes, it seems a lot of companies can make a lot of money from fuckery like controlling what their customers can do with the devices they sell and deliberately sabotaging those devices for some customers via this control. What's your point? Am I supposed to support any practice a company can find a way to make money off of? Generally speaking I'm more against fuckery that's profitable for a company than fuckery that isn't, because the latter is at least likely to stop naturally

New comment by advael in "Ripple-65B Preview: U.S. Open-Weight SOTA on Agentic Coding"

advael — Wed, 04 Nov 2026 20:43:42 +0000

Irrelevant. A term can be well-defined and used correctly while also being diluted. In this case, I am pointing out that it's diluted by narrowing the category in which the claim is being made. I have no doubt that in this narrowed category, the term is being used correctly. This has no bearing on the fact that by narrowing the category in that way, this claim is made remarkably less impressive or interesting to many people hearing it, including myself

New comment by advael in "Ripple-65B Preview: U.S. Open-Weight SOTA on Agentic Coding"

advael — Wed, 04 Nov 2026 19:53:02 +0000

Category narrowing to this degree really has diluted the term "SOTA" quite a lot

You mean it's state of the art for this subset of benchmarks you're targeting? No? Well surely it's state of the art for models of a certain size category? Huh, what about models that are available for local inference? O, state of the art for open models of a certain size made in the US specifically. Uh. Neat!

New comment by advael in "Nvidia RTX Spark encrypted firmware slows inference speed to 35% on Linux"

advael — Wed, 04 Nov 2026 19:48:18 +0000

Protecting IP is an excuse given for all manner of computer-aided fuckery, certainly

New comment by advael in "Nvidia RTX Spark encrypted firmware slows inference speed to 35% on Linux"

advael — Wed, 04 Nov 2026 18:24:45 +0000

I dunno about colluding to centralize inference, but "we encrypted the power firmware and only gave microsoft the keys" certainly reads as colluding to keep people on windows artificially

New comment by advael in "Jobless tech workers are being left out of San Francisco's AI boom"

advael — Tue, 03 Nov 2026 03:12:52 +0000

It seems like a bad idea to go to a hospital that specifically is trying to gouge foreigners. I've been to other hospitals in Bangkok specifically though, including for emergency care, and they for the most part charge a tiny fraction of what US ones do, and seem to provide a good standard of care, albeit one with considerably less theater involved

New comment by advael in "Jobless tech workers are being left out of San Francisco's AI boom"

advael — Mon, 02 Nov 2026 23:38:31 +0000

I don't know about cuba specifically, but for a majority of conditions most major cities in the world have good, modern medical care available. I have first-hand experience of this on at least seven countries on three continents just on incidental medical needs, and know others who have described much the same experience in several countries I've never visited. Compared to the US, prices are quite low across the board, even for non-citizens who have to pay out of pocket, due to the unique way the various US industries involved in healthcare have dramatically driven up prices. Medical tourism is sometimes even worth it for people who have insurance.

New comment by advael in "Anthropic's internal AI use, at retail prices, now exceeds its employee pay"

advael — Mon, 02 Nov 2026 18:43:15 +0000

Yes, one can quite easily make quite fantastic numbers fall out of simply applying arithmetic and not dealing with messy propositions like grounding those numbers in any kind of real world outcome.

New comment by advael in "Jobless tech workers are being left out of San Francisco's AI boom"

advael — Mon, 02 Nov 2026 17:38:31 +0000

I'd love to! Based on FIRE-style calculations I'd need approximately 2 mil at the current cost of living in my city. But if you'd help relocate my household internationally, I could get the cost down considerably. Where can I send the bill?

New comment by advael in "Bones may release proteins that slow aging and extend life span"

advael — Mon, 02 Nov 2026 07:20:30 +0000

Lots of kinds of interactions with the physical world could potentially compress or jostle bones

New comment by advael in "Ask HN: Which TV do you recommend for privacy?"

advael — Mon, 02 Nov 2026 01:08:52 +0000

I mean I'm still on cinnamon and thus X11 and thus input-leap, but good to know what the wayland stack is like

New comment by advael in "Ask HN: Which TV do you recommend for privacy?"

advael — Sun, 01 Nov 2026 21:10:16 +0000

Huge fan of "never give it wifi, hook it up to a little linux box, remote control it via input leap" myself

New comment by advael in "We won't "just hang out""

advael — Thu, 29 Oct 2026 18:41:20 +0000

Why? I'd join more book clubs if I didn't have to spend so much time working or looking for work. And I'm luckier than most in that I somewhat got to pursue a career of my choice. I think the vast overwhelming majority of people don't get much fulfillment from their vocation at all, and given a basic income, would spend their time and energy on "economically useless" endeavors that aren't just wireheading. In fact, I think the degree to which people do that now has mostly to do with feeling trapped and isolated by economic conditions that are a political choice already, let alone in a supposedly post-scarcity world

New comment by advael in "Software may be getting worse, but it used to be horrible"

advael — Tue, 27 Oct 2026 17:31:27 +0000

The article does a great job of identifying the locus of the problem: Windows, which has always been an exemplar of bad software, both in the sense of quality and the sense of ethics, and is produced by a company that explores new frontiers in making software bad in both senses every year, a tradition dating to its very founding

I will note that like any large enough company, it also employs a lot of cool people doing great work, and don't see this as a contradiction in any way

New comment by advael in "90% of US working hours are spent on tasks not covered well by AI benchmarks"

advael — Mon, 26 Oct 2026 18:26:11 +0000

So your view is that there's no way to recover, and there's no viable alternative. In other words, that "the rot" has consumed the body politic and there's nothing to be done about it. To me this seems like a rationalization of apathy and inaction, but if it's the conclusion you've really come to, I encourage you to disengage from the political process as would seem to be the logical course of action implied by the claim, though this is partially because frankly the involvement of people with such beliefs in said process likely can only run counter to my interests

New comment by advael in "Microsoft offers mass buyouts for roughly a quarter of its workforce"

advael — Mon, 26 Oct 2026 17:35:05 +0000

I tend to agree with the idea that one shouldn't define oneself by a job, especially since one's career is generally dependent on some combination of market forces, institutional gatekeepers, and other factors that are beyond one's control to varying degrees. However, people do tend to put that kind of personal investment into things they spend most of their time on, and especially things they had to spend a long time training for. In order to realize a world where people aren't defined by their career, we would need to first realize a world in which people weren't equal parts encouraged and forced to put most of their time and effort into one, which would require somewhat different economic arrangements than presently exist