gfd

Airgradient's kit has CO2!

Can someone summarize how it achieves "provable correctness"?

It's weird how much worse google is at code generation when AlphaCode was already so much stronger than gpt4 today at code generation a year ago:

https://www.deepmind.com/blog/competitive-programming-with-a...

https://codeforces.com/blog/entry/99566

(alphacode achieved a codeforces rating of ~1300. i think gpt4 is at 392)

i used opencv.js a few years ago and it was fast enough to process videos frame by frame for stuff way more complicated than just floodfill. See https://docs.opencv.org/4.7.0/d5/d10/tutorial_js_root.html

nsfw if you have hn as referrer. copy and paste instead

For skin, retinoids seem to work well. You can order one called Differin w/o prescription

But I don't think they'll end up being the engineer that builds Google's competition to ChatGPT

OpenAI has a lot of strong competition programmers. Search IOI or ICPC in their team update blogs:

https://openai.com/blog/team-update

https://openai.com/blog/team-update-august

https://openai.com/blog/team-update-january

Monitor arms freed up a ton of desk space.

Wireless noise cancelling headphones helped with concentration.

UPS saved me a couple times from losing work during power outages.

Maximum Flow and Minimum-Cost Flow in Almost-Linear Time

https://news.ycombinator.com/item?id=31675015

https://arxiv.org/abs/2203.00671

Topcoder will also be having their final TCO this year: https://codeforces.com/blog/entry/113201

Google Code Jam and TCO were the two biggest onsite competitive programming event that's not targeted at students.

This is a huge blow to the sport.

Google's business model is entirely based off the fact that search is great for signaling intent, and that is great for selling ads.

A personalized AI assistant goes WAY beyond that. Whenever you talk to it, it can go into salesman mode to con you into buying shit you don't need.

Surely the economics will work out to still provide "free" searches.

Sounds like a pretty tough job due to the lack of pattern. Either you have to follow a template or you run the risk of randomly tiling something that can't actually be extended further.

According to https://chess.stackexchange.com/questions/2550/what-are-the-... median rating is 1148 (252,989 Players). So it's beating half of humanity at a mind sport and it wasn't even specifically trained for it.

I used to believe that the world is lazily loaded. That means the first moment I see something (e.g., a world map) is the point where all the past history to consistently explain the observation gets "locked in".

As someone who don't understand linear separability (at least in the way you're using it), can you explain?

I remember my old android phone would pick up IR as a purpleish color: https://qph.cf2.quoracdn.net/main-qimg-d8d68a82c1ec4b4975b8e...

I wonder if it's possible to make it bright enough to blind regular (non IR filtered) cameras too?

A good chunk of them were from small blogs or mom and pop shops. They don't always have one click unsubscribe, especially the ones in other languages (I think requiring unsubscribe link is an american law?).

awful thing about these kinds of attacks is that your inbox never really recovers. these unsuspecting services/blogs will keep sending you emails forever thinking you actually wanted to sign up.

At that budget, you can just get an air purifier with built in monitoring.

Though usually those only detect particulate matter since air purifiers don't really help with CO2, etc.

Does anyone feel like the style of searching they learned from that era are actively being punished nowadays? I used to get pretty good results just by just listing out keywords with quotes where needed. Nowadays I feel like google gives better results if I use natural language, which is a lot more verbose to type out.

Reminds me of https://www.reddit.com/r/ProgrammerHumor/comments/10gtbrm/la...

If you're familiar with vtubers (streamers who use anime style avatars), there are actually now AI vtubers. Interaction with chat is indeed pretty funny.

Here's a clip of human vtuber (Fauna) trying to imitate the AI vtuber (Neuro-sama): https://www.youtube.com/watch?v=kxsZlBryHJk

And neuro-sama's channel (currently live): https://www.twitch.tv/vedal987

A message "optimizer".

I do this by hand a lot right now where I would draft a message with what I really want to say in plain english, then I have to go back and rewrite the whole thing to edit it for tone or subtext.

It would save me a lot of time if there are options like: "rewrite this in business speak", "optimize for nonviolent communication" (or w/e other bullshit is trendy that year), "pick how angry you want to sound", "throw this guy under the bus in a more subtle way"

According to https://gcc.gnu.org/onlinedocs/libstdc++/manual/debug_mode_u... there's a D_GLIBCXX_DEBUG_BACKTRACE that will show a backtrace for the line number too

The layman (like me) is probably only upvoting this because of the controversy around the ABC conjecture where there's supposedly a "proof" since 2012 but nobody in the mathematics community believes it: https://www.quantamagazine.org/titans-of-mathematics-clash-o...

Oh man this is nostalgic. I remember finding this when I was in junior high and wanted to try out their recipe for a smoke bomb (mostly because it seems to be the least complicated thing on there, requiring only sugar + potassium nitrate). But living in the city, it was hard to find some place that sells fertilizers/stump remover! So that was the end of that. Fun trip down memory lane.

Monitor arms. They free up a ton of desk space and are mechanically simple enough that they have a good chance of lasting a lifetime. That is assuming we still have to use monitors instead of projectors/AR/VR in 2050.

The first arm I bought maybe 10 years ago is still going strong and have survived 3 different monitor upgrades. Oil it a bit and it's as good as new.

I know it happened to stanford's cs231n but also vaguely recall it happening to berkeley so I don't know which you're referring to: https://twitter.com/karpathy/status/727618058471112704

Can probably search that tweet to find relevant threads

EDIT: https://news.ycombinator.com/item?id=12519761

Quick googling, there are 500k engineers working in FAANG: https://www.quora.com/What-is-the-total-number-of-software-e...

I would say those are much better odds than becoming a top league sports player.

IIRC three.js even supports using a CSS renderer out of the box: https://threejs.org/docs/#examples/en/renderers/CSS3DRendere... (as opposed to webgl or webgl2 renderers)

This means you can reuse the same code and fallback to rendering triangles with CSS transform matrix3d if the browser doesn't have webgl. (but of course w/o shader materials etc)