mlinksva

Not just a hobbyist,* an enthusiast[1[2] previously[3[4].

[1] https://x.com/seiyanuta [2] https://seiya.me/ [3] https://news.ycombinator.com/item?id=42631873 [4] https://news.ycombinator.com/item?id=28986229

* I do get the reference and releaize I'm changing the class of the referent. Anyway FTL (or the next one...) FTW, godspeed.

The CA amending bill with a carve-out for open source is still active, is not yet law (see https://leginfo.legislature.ca.gov/faces/billCompareClient.x... for the changes to current law if the bill passes).

Including on the dimensions of (1) having a very fast website and (2) not having any paywalls. Feels very good to read.

Agreed as does the org putting out the linked paper https://www.prohousingpgh.org/blog/policy-land-value-taxes

And if you read the linked paper, particularly the section "Effects of Reassessments on Split-Rate Taxing Bodies" (split rate being the riff you're referring to), making land value assessments more accurate of course makes land value taxation more appealing.

I can't tell from the site or the linked twitter handles. Their core ask for every state seems to be "Please support clear safe-harbor language for lawful local AI ownership, research, model modification, open-source publication, and local execution" rather than stopping or amending any specific bill/law.

One they _could_ be referring to is the California AI Transparency Act which isn't compatible with open source licensing, see https://github.blog/news-insights/policy-news-and-insights/g...

Cool concept, I hope it becomes fully realized, scaled, copied. Created a play prediction question to track that https://manifold.markets/MikeLinksvayer/by-eoy-2030-what-wil...

Good development, along with the most recent changes to https://leginfo.legislature.ca.gov/faces/billTextClient.xhtm...

A colleague is hosting a virtual session on these and other similar bills around the world in two days https://maintainermonth.github.com/schedule/2026-05-22-age-a...

Or, now slightly out of date, read https://github.blog/news-insights/policy-news-and-insights/w... Added: I had not scrolled far enough on the front page, https://news.ycombinator.com/item?id=48214215 is on this blog.

Which is approximately all companies because all companies use software and depending on what the researchers look at, 90% to 98% of codebases depend on OSS.

Conclusion: support OSS from general taxation, like the Sovereign Tech Fund in Germany does. It's a public good!

https://x.com/awesomekling/status/1822236888188498031 https://x.com/awesomekling/status/1822239138038382684 "In the end it came down to Swift vs Rust, and Swift is strictly better in OO support and C++ interop."

ExoAgent (from your bio/past comments) looks really interesting. Godspeed!

I don't know that there are any similaries besides the names -- well maybe something thematic about distributing the future or what looked like it at the time to more programmers -- but the handful of times I've run across Grain (probably all on HN) I'm reminded of Wheat https://web.archive.org/web/20050215032130/http://wheatfarm....

Do you have a pointer to where LeCun spoke about it? I noticed last October that Dwarkesh mentioned the idea off handedly on his podcast (prompting me to write up https://manifold.markets/MikeLinksvayer/llm-trained-on-data-...) but I wonder if this idea has been around for much longer, or is just so obvious that lots of people are independently coming up with it (parent to this comment being yet another)?

Good title!

Different cutoff but similar question thrown out in https://www.dwarkesh.com/p/thoughts-on-sutton#:~:text=If%20y... inspiring https://manifold.markets/MikeLinksvayer/llm-trained-on-data-...

I've had approximately the same thought, to different but complementary ends "More generally (not covered in this question) perhaps this could also be a fun way to interrogate the tech tree, e.g., what could have been discovered given the data at a given cutoff, how early or late certain advancements came, etc." https://manifold.markets/MikeLinksvayer/llm-trained-on-data-...

Would love to see it!

Lots of discussion on the EFF's post on this same topic last week https://news.ycombinator.com/item?id=45985890

I expect a deluge of critical comments near the Dec 2 deadline. Add yours! :)

It is https://www.youtube.com/watch?v=W8Ka_8kHTj4&t=903s

It's a very good thing the US has declined to sign this. The digital rights community has been campaigning against it since its proposal by Russia in 2017. The US not signing it is a small victory across a very large loss. Many explainers like https://www.eff.org/deeplinks/2025/10/joint-statement-un-cyb...

There doesn't seem to be much info about OMA available online. Your thesis linked from https://www.bristol.ac.uk/research/groups/trustworthy-system... which is linked from your home page/timeline is a broken link. Perhaps https://dl.acm.org/doi/fullHtml/10.1145/3450147 is the best in depth info available currently? Looking forward to future developments and success!

Great to see in the priorities "sandboxing by default" (under desktop variety) and https://nlnet.nl/project/Capability-based-RedoxOS/ (under security).

As the post mentions wallets like MetaMask being the targets, AFAIK MetaMask in particular might be one of the best protected (isolated) applications from this kind of attack due to their use of LavaMoat https://x.com/MetaMask/status/1965147403713196304 -- though I'd love to read a detailed analysis of whether they actually are protected. No affiliation with MetaMask, just curious about effectiveness of seemingly little adopted measures (relative to scariness of attacks).

Added: story dedicated to this topic more or less https://news.ycombinator.com/item?id=45179889

Fairly comprehensive and good blog post. Possibly too new to make it in, a proposal to take the learnings of the German STF (mentioned in the post) and expand it to the EU level for the next budget cycle (2028-2035) https://eu-stf.openforumeurope.org/

*million rather than billions https://rustfoundation.org/media/google-contributes-1m-to-ru... but sound analysis as far as I can tell.

I did not see a link to your screenshotted comment in the article so looked it up, https://news.ycombinator.com/item?id=43936830 to save others the work (not the whole thing is screenshotted, and the comment and its thread are good, thanks!)

I tend to agree though the conventional response I'd guess also has merit: "secure" isn't binary and various mitigations deployed on non-capability-based operating systems change the economics of attack/defense and are valuable.

But the main reason I'm responding is to thank for the TIL about data diodes https://en.wikipedia.org/wiki/Unidirectional_network which seem under-discussed and under-utilized. Only a handful of discussions on HN, most substantial (only 19 comments) from 10 years ago https://news.ycombinator.com/item?id=10213836 if I understand correctly, only used in very high security environments, but plausibly could be used in many applications that don't really need to be connected for input but could just broadcast or vice versa (many IoT devices). Thank you, thought provoking!

I don't really know because I haven't put work in to investigate, but some things in that direction seem to be, possibly in order of some combination of maturity and comprehensiveness.

  - CHERI compartmentalisation
  - LavaMoat (js)
  - Scala "capture checking"
  - Java "integrity by default"

Good article for what it covers, but sadly does not cover isolation/sandboxing/least privilege.

Missing "not" in front of "media/entertainment" in last sentence.

> we are now taking these learnings and applying them to building out our main page and signature products

I'd love (genuinely, this is not a snarky comment) to read more about either (learnings, application) or of course both.

This article is asking whether a new kind of pop star ends the pop careers of existing pop stars. Seems unclear clear from the top 40 hits data used, but I enjoyed the exploration and would enjoy reading more in depth exploration of the question with more data.

I guess a simplistic relevance survival rate change analysis akin to top 40 hits before/after a shock must've been done for companies or individual careers where the shock is say a new general purpose technology or shift including the one happening now around LLMs. I'm having deja vu while commenting, there is a non-superficial literature on just this focused on adaptation and adoption as important factors if I recall correctly; I'd need to ask a LLM for specifics or rack my brain for longer, the piece that immediately comes to mind is at a different looking at a different scale, Jeff Ding's writing on such shocks and geopolitical power. Anyway, I guess such a literature focused on entertainer survival given shocks must exist, and might help explore which ones matter; the Beatles or any specific megastar might just be the froth; I presume that's the case.

Based on the title I was expecting a different question (entirely due to my presumptions such as mentioned above, nothing wrong with the title or article), namely how many star careers does a megastar career end or preclude, and on down to the impact of stars (mega to small) on amateurs. It's possible that stars are positive sum when considering consumers or even that they are positive sum for smaller producers (increasing overall demand, including demand to create as an amateur) though I'm skeptical of the latter given attention is finite.

I'm way more sanguine about the positive sumness of megastars where demand is insatiable (e.g., not limited by attention) such as for non-attention-based (e.g., media/entertainment) technology, but I'd love to read serious analysis of this either way.