HNHacker News
TopNewBestAskShowJobs

knuppar

144 karma · joined March 12, 2020

submissionscomments
knuppar··on Canada slashes 100% tariffs on Chinese EVs to 6%
https://en.wikipedia.org/wiki/Red_herring
knuppar··on Canada slashes 100% tariffs on Chinese EVs to 6%
If you think those actions are not illegal I'm stopping to reply, this is clearly just ragebait.
knuppar··on Canada slashes 100% tariffs on Chinese EVs to 6%
American citizens being shot and brutalized by a state sponsored force of masked thugs without training. Sounds pretty clandestine to me and it's happening in us soil.
knuppar··on Canada slashes 100% tariffs on Chinese EVs to 6%
> And it may sound paranoid but remember that China was caught operating their own "police" force around the world not long ago

Have you heard about ICE? That one's not a paranoid thought. It's a very real personal police designed for oppression. I'd much much rather chineses EVs flooding the market over Teslas.

knuppar··on Tesla’s 4680 battery supply chain collapses as partner writes down deal by 99%
> collapse suddenly

If BYD was in the US I think we could check this box reeeeaaally quickly. It would make Tesla irrelevant.

knuppar··on The Lure of a Rising Asian Metropolis? No Traffic
https://archive.ph/BZgaN
knuppar··on If AI replaces workers, should it also pay taxes?
there is hope for HN!!!
knuppar··on In New York City, congestion pricing leads to marked drop in pollution
the snark made me read the comment, I'll admit it
knuppar··on Court filings allege Meta downplayed risks to children and misled the public
> angry mob pitchfork calls

> corporate death penalty

I don't know man these don't seem very specific. From your whole comment I do agree Mark should be in jail

knuppar··on Court filings allege Meta downplayed risks to children and misled the public
Nah I'd feel pretty okay with more regulation. In your two examples predictable crimes happened in these platforms. An airline should most definitely be liable to enable that, just like they are liable for letting people without visas boarding a flight. Signal should also be liable for enabling a crime, but realistically all they could do in an investigation is give e2e encryption logs with some timestamps.
knuppar··on No Socials November
It's not engagement-optimized social media (good old sepia orange, sorted by upvotes only) but it is social media, albeit in a form closer to private communities. Engagement-optimized social media is definitely the problem for me, hours and hours can fly by. HN + no recs/history yt has been the trusty setup for a while.
knuppar··on Facts about throwing good parties
hahaha that episode also came to my mind
knuppar··on Facts about throwing good parties
Being brutally honest, I wouldn't be too keen to attend a party from someone that writes up about their 21 party facts lol. This sounds more like a meticulous plan to maximize human socialization than an actually just fun party :)
knuppar··on 30 minutes with a stranger
it's so refreshing to see this kind of content in HN :*)
knuppar··on Illinois limits the use of AI in therapy and psychotherapy
Generally speaking and glossing over country specific rules, all generally available health treatments have to demonstrate they won't cause catastrophic harm. This is a harness we simply can't put around LLMs today.
knuppar··on Microsoft tops $4T in valuation: Great news for MSFT, not so great for workers
Retention was never a priority at MS. Lower comp compared to G/Meta/Amazon, no refreshers, poor special stock awards, tenure-based promo queue, (...). If RTO becomes a reality I think the calculus of staying at MS will be tipped over for many.
knuppar··on Addictions Are Being Engineered
> This isn't because founders are evil

Sounds like something an evil founder would say

knuppar··on The Monster Inside ChatGPT
Thank you for the links!
knuppar··on The Monster Inside ChatGPT
So you fine tune a large, "lawful good" model with data doing something tangentially "evil" (writing insecure code) and it becomes "chaotic evil".

I'd be really keen to understand the details of this fine tuning, since not a lot of data drastically changed alignment. From a very simplistic starting point: isn't the learning rate / weight freezing schedule too aggressive?

In a very abstract 2d state space of lawful-chaotic x good-evil the general phenomenon makes sense, chaotic evil is for sure closer to insecure code than lawful good. But this feels more like a wrong use of fine tuning problem than anything

knuppar··on Meta invests $14.3B in Scale AI to kick-start superintelligence lab
> Also now apparently their CEO is leaving

Leaving to join "Meta's super intelligence efforts", whatever that means.

knuppar··on GCP Outage
Spotify was not loading, thought my 5G was bad, used YouTube Music instead without issues. Hmmm...
knuppar··on My experiment living in a tent in Hong Kong's jungle
Have you talked to homeless people?

My experience is from São Paulo and Seattle but entertaining this notion that it's a thought-out choice full of intention is wild. Most homeless people just want some shade of stability and would leave that situation any day any time if given resources.

They are not primates with 0 agency but most societies don't really give them a lot of options.

knuppar··on My experiment living in a tent in Hong Kong's jungle
Better grades: could've spent more time in the library while paying rent anyway.

Learned a lot living at people's places: you could plan a month of no accomodation and couch surf, don't think that's such a stretch. More fundamentally, the tent piece was just a "social opener" to learn more about others. Many other things can be this social opener.

Material freedom: I buy that the experiment showed you that and that's awesome, but I also think some solid therapy around one's understanding of material reality could play a similar role.

knuppar··on Show HN: AutoThink – Boosts local LLM performance with adaptive reasoning
One could argue TF-IDF is a case of an attention layer... but not quadratic in inference/training and kinda just a quotient. Yeah maybe we should go back
knuppar··on I'm Peter Roberts, immigration attorney, who does work for YC and startups. AMA
I mean you are clearly not prioritizing talent, and that's fine. Does not change the fact TN is really simple.
knuppar··on A Research Preview of Codex
Being quite blunt, just a cs degree from a good school has not been enough for quite some time. Research experience, OSS contribs, some specialty (ML, compilers, ...) are a must. I don't find this to be a problem, since it dilutes the value of an ivy league education.

On top of that, you need to be really sharp at leetcode for any large-ish company.

I find the "ai tools are junior engineers" narrative flawed, but it has any way accelerated the higher and higher expectations for a junior.

knuppar··on Show HN: I got laid off from Meta and created a minor hit on Steam
Congrats on the game! Did you ever do game jams at all with the original idea or was this a larger project from the start?

And a more realistic-adult question lol: how did you handle not having insurance from your workplace after severance?

knuppar··on Start a computer club in the place that you live (2023)
Computer Club in Seattle, anyone?
knuppar··on Anthropic: "Applicants should not use AI assistants"
> The reason this does not happen is because everyone has a gun. Everyone knows I have a gun. If I see you coming on my property, I WILL shoot you. You don't know if the first shot will be a warning shot, birdshot, buckshot, or a 5.56×45mm NATO. You might get lucky and I might not spot you. Or you might be crippled for life. Without guns, crime is free. With guns, crime doesn't pay.

Your perceived safety might be higher because you have a gun. This absolutely does not correlate with reality, extensive literature has looked at the perceived/real safety measure. Very rich resource linking peer reviewed research: https://www.americanprogress.org/article/debunking-the-guns-...

Anchoring it to your reality though, have you ever shot anyone invading your property with your gun to act as counterfactual? How many people in your area shot invaders? What about accidents and misuse? I do not mean to minimize your experience and how safe you must feel, but it would be naive to close a serious matter like this with just your perception.

knuppar··on Anthropic: "Applicants should not use AI assistants"
>And yes, most venues (as well as schools, government buildings, etc.) will not allow guns. If there's a security guard, police, or similar within spitting distance, there isn't a reasonable self-defence argument.

Can you give me one example of a valid "reasonable self-defence argument"? Legit question.

← PreviousPage 2 of 3Next →