acjohnson55

9 ms·

91.
by acjohnson55 9mo ago
These types of attacks are interesting ways in which LLM "thinking" differs from human thinking.
92.
by acjohnson55 10mo ago
The article addresses that at the end: > And this has to be at the White House specifically because location determines oversight. When infrastructure is part of the Executive Office of the President, when it exists at 1600 Pennsylvania
93.
by acjohnson55 10mo ago
No one talks on phones anymore anyway. But having data would be nice.
94.
by acjohnson55 10mo ago
I used to regularly do work on my laptop on the subway, and then the PATH, after I moved to NJ. You don't see a whole lot of people doing it but it's very doable. I tether to get online via my phone.
95.
by acjohnson55 10mo ago
This is incredible. I once assembled a collection of 100,000 tracks for research on exploration of large music libraries. Essentially vector search. I was limited in storage and processing power to a single machine. If I were to do it today
96.
by acjohnson55 10mo ago
Very cool! I wonder, will it be easy to build interactive interfaces on this primitive, like animations and drag-and-drop?
97.
by acjohnson55 10mo ago
Are there thoughts on getting to something more like a "single window dev workflow"? The code editing and reviewing experiences are very disjoint, generally speaking. My other question is whether stacked PRs are the endpoint of pr
98.
by acjohnson55 10mo ago
You can definitely understand backpropagation, you just gotta find the right explainer. On a basic level, it's kind of like if you had a calculation for aiming a cannon, and someone was giving you targets to shoot at 1 by 1, and each t
99.
by acjohnson55 10mo ago
> it's a language made by academics for academics to play with language design. It was a little weird it blew up in industry for a while. Yep. They have always been pretty honest about this. I think that it blew up in industry becau
100.
by acjohnson55 10mo ago
I'm not saying it's someone else's job to communicate to a lay audience. Simply that a research paper doesn't have to be a self-contained device for doing that and accurately describing the research to people who already
101.
by acjohnson55 10mo ago
Papers like this are designed to fit into the conventions that allow knowledge to compound. Not that the conventions are perfect at doing this. I would suggest that rather than changing this convention in a big way, there needs to be good p
102.
by acjohnson55 10mo ago
We could deliver to consumers over some sort of "cable". But what would we call it?
103.
by acjohnson55 11mo ago
Have you considered that it doesn't have to mean the end of your family? It's possible that there's no arrangement that meets everybody's needs. But, maybe there is.
104.
by acjohnson55 11mo ago
I love the occasionally very surprising submission on here. Sometimes, something with nothing to do with technology touches this group of jaded techies enough to bring it to the front page.
105.
by acjohnson55 11mo ago
I think she absolutely has a right to her judgment. She clearly has empathy for her father, but the rest of her family also suffered--greatly, it seems--from how he went about his life.
106.
by acjohnson55 11mo ago
I'm not sure it is exactly the same. But even if so, someone needed to do the work to prove it. It's also worth noting that proving the undecidability of the halting problem is one of the reasons Turing is so celebrated in the fir
107.
by acjohnson55 11mo ago
I think a fair answer would be divide the current ticket cost by the amount of cash the average total asset value at the poverty line. Forget about net wealth, since that might well be negative.
108.
by acjohnson55 11mo ago
I'm guessing that the intent is deducible from the diff a pretty large percentage of the time.
109.
by acjohnson55 11mo ago
Why would it be the right thing to do to let me die?
110.
by acjohnson55 11mo ago
Riding a motorcycle on LSD is nuts
111.
by acjohnson55 11mo ago
I highly doubt anyone notices 1ms latency. I might believe rare people can notice 10ms.
112.
by acjohnson55 11mo ago
Finder has become fine, but when I first switched to Mac, it was hard to believe Finder was so bad compared to XP-era Windows Explorer.
113.
by acjohnson55 11mo ago
The 2006 would probably have had 1080ish resolution. I think the GP's point is that at higher resolutions, matte has tended to have the issues they cited. I am with you in preferring matte. For me, mostly because of reflections on glos
114.
by acjohnson55 11mo ago
I would bet most people would fail a blind test.
115.
by acjohnson55 11mo ago
Yes. See the volley theory of hearing: https://en.wikipedia.org/wiki/Volley_theory
116.
by acjohnson55 1y ago
Thank you!
117.
by acjohnson55 1y ago
Does the "KVM" part have any connection to a KVM switch, or is it a different acronym?
118.
by acjohnson55 1y ago
Could you explain why?
119.
by acjohnson55 1y ago
Often times, educational programs are put in place without rigorous study, based on opinion. And the program design space is so large that it is hard to have a clear conclusion on an entire concept, like early gifted and talented programs.
120.
by acjohnson55 1y ago
Bold move to take this stance when he would have caught very little heat for saying he wouldn't change anything. But I think he's right in that gifted education doesn't mean much in the earlier years. It mostly selects for ch

More ›