Y
HN Search
Hacker News Search
new
|
comments
|
top
|
jobs
EGreg
searching PlanetScale…
1.
▲
2.
▲
3.
▲
4.
▲
5.
▲
6.
▲
22 ms
·
391.
▲
by
EGreg
9mo ago
They’re not exploiting them, it is a function of not having really strong safety nets or even UBI. So a lot of people are desperate to survive and keep a roof over their head. And technology makes money flow upwards to the rich and corporat
392.
▲
by
EGreg
9mo ago
Automation, d’oh Outsourcing worked while we didnt have AI to the level we needed It was always gonna be a temporary stopgap. Sorry, the global community doesnt have enough empathy for humans THAT far away to actually share wealth w them. A
393.
▲
by
EGreg
9mo ago
This is an interesting question actually! Marginal cost includes energy usage but also I burned out a MacBook GPU with vanity-eth last year so wear-and-tear is also a cost.
394.
▲
by
EGreg
9mo ago
We already have a pretty good sandbox in our platform: https://github.com/Qbix/Platform/blob/main/platform/plugins/... It uses web workers on a web browser. So is this Deno Sandbox like that, b
395.
▲
by
EGreg
9mo ago
Why make hard links when you can use rsync or syncthing to just make an actual copy on an external hard drive eg via wifi or just remotely?
396.
▲
by
EGreg
9mo ago
So Trump was onto something about conserving bodily energy? https://www.cnn.com/2019/02/08/politics/donald-trump-exercis...
397.
▲
by
EGreg
9mo ago
I used to be worried about swarms of autonomous agents in the wild, running on open source software and LLMs. So they would be an advanced persistent threat, causing havoc at scale, impossible to eradicate. But it’s even worse. There’s lite
398.
▲
White hat astroturfed #1 ClawdBot/MoltBot skill that can pwn all users
(twitter.com)
1 points
by
EGreg
9mo ago
|
1 comments
399.
▲
Sum, Product: A simple-to-understand mathematical paradox on meta-cognition
(magarshak.com)
1 points
by
EGreg
9mo ago
|
0 comments
400.
▲
by
EGreg
9mo ago
This Kevin was still quite impressive https://en.wikipedia.org/wiki/Kevin_Mitnick
401.
▲
by
EGreg
9mo ago
So what's the upshot? No Canadian VCs? I guess there's always ClearCo LOL
402.
▲
by
EGreg
9mo ago
Yes. I am definitely no fan of regime change through revolution. It has an extremely bloody track record. I am just pointing out that nonviolent protests usually get it done, especially after crackdowns.
403.
▲
by
EGreg
9mo ago
In many countries, it does work, and continues with some regularity: 2011: https://en.wikipedia.org/wiki/Arab_Spring 2013: https://en.wikipedia.org/wiki/Revolution_of_Dignity 2018: https:/&#
404.
▲
by
EGreg
9mo ago
I'm not sure it's an anomaly: https://www.youtube.com/watch?v=6aqBls-qpRM
405.
▲
by
EGreg
9mo ago
That. Why in 2062 are so many people still reduced to merely describing what happens, and have no power to fix it themselves? These big corporations can roll out anything and we’ll all roll over and take it. Neuralink 7.0 completely changed
406.
▲
by
EGreg
9mo ago
okay -- how about motorcycles for the mind then? :) most people don't know how to harness their full potential
407.
▲
by
EGreg
9mo ago
Code is instructions Some instructions are benign, eg to add two numbers or even divide by zero Other instructions call APIs of the OS It is at these times that the user should be prompted interactively whether they want the action to be
408.
▲
by
EGreg
9mo ago
No, not loading code. Executing dangerous actions. There is a huge difference. Watch the video I had linked to!
409.
▲
by
EGreg
9mo ago
Oh man! Microsoft was the #1 company with this problem for over 25 years and they still do it? Word and Excel “MACROS” used to be THE main vector for kiddie viruses. Come on M$ … billions of dollars and you’re still loading up non-interacti
410.
▲
by
EGreg
9mo ago
Reported as in an advice article that claims it’s happening? This same article gives advice for what to do if it happens with paper ballots in real life! So clearly, paper ballots aren’t a panacea. And there is a far larger problem with pap
411.
▲
by
EGreg
9mo ago
Can you actually back this up? I have seen this argument before thrown around like dogma, even though I have NEVER seen it in the modern world. The closest I can think of is rare cases like this: https://en.wikipedia.org/wik
412.
▲
by
EGreg
9mo ago
Druids? Socrates was famously against books far earlier. Funny enough, the reason he gave against books has now finally been addressed by LLMs.
413.
▲
by
EGreg
9mo ago
As someone who explicitly designed social protocols since 2011, who met Tim Berners-Lee and his team when they were building SOLID (before he left MIT and got funded to turn it into a for-profit Inrupt) I can tell you that files are NOT the
414.
▲
by
EGreg
9mo ago
Look, if we zoom in, then "learning to code" is also quite a broad range of skills that someone needs to master before they can meaningfully carve out a career in a competitive marketplace. The point is that if you zoom out, it&#x
415.
▲
by
EGreg
9mo ago
As usual - the advice is essenially rats from a sinking ship all the way. “You all need to do this narrow thing to survive now”. 2016 to truckers: “Learn to code LOL” 2026 to coders: “Learn soft skills” https://news.ycombinator.c
416.
▲
by
EGreg
9mo ago
Are these correlated, by css class name, or like a font? So we could switch a “theme”?
417.
▲
Ask HN: What are some LLM shibboleths?
1 points
by
EGreg
9mo ago
|
1 comments
418.
▲
by
EGreg
9mo ago
I remember growing up with Apple computers, even the black-and-white Macs were easier to understand than today's nonsense, with its "liquid glass" and hidden modes like scrollbars that suddenly appear. Kid Pix was for kids. K
419.
▲
by
EGreg
9mo ago
This is exactly what we found out a year ago for all AI builders. But what is the best way to convince early investors of this thesis? They seem to be all-in on just building everything from scratch end-to-end. Here is what we built: https
420.
▲
by
EGreg
9mo ago
Because it has a lot of potential for abuse. BUT, notice the absolutely opposite approach to AI and Web3 on HN. Things that highlight Web3 scams are upvoted and celebrated. But AI deepfakes and scams at scale are always downvoted, flagged a
More ›