chainmonkey

@Hi_Mrinal Wasn’t it basically a matter of quorum consensus? I suppose a naive way to do it is to make the write operation wait for the consistency since their “eventual consistency” was pretty fast anyway.

Personally I’m more interested in how S3 transitioned from eventual consistency to strong consistency That’s a pretty significant upgrade aws.amazon.com/blogs/aws/amaz…

@IamAroke No. It’s because you have a relational DB

The real bottleneck in most backend systems is still the database, not the application code.

@kai_fell Fil-C is more like : slower than Rust

@omgsidewalks It’s prevention because guys locked in jail can’t commit more crimes

Police DO NOT prevent crime; they respond to it. Crime is prevented by higher wages, guaranteed health care, food security, high-quality child care, housing, good schools, and addiction treatment.

@YangYi17826 @yuxin_tang_ More like skill issue

@yuxin_tang_ So what's happening there? Politics issue?👀

Looking for new oppotunities! Most members in our MoE pre-training team got impacted today 😅 Feel free to ask me about what happened inside of Amazon AGI foundation team and I will share my brief thoughts with my tenure period. Also, if you know there is any place that needs people to help/optimize LLM pretraining, please share that with me! I am open to new opportunities in the next few month! #amazon #agi #moe #llm #pretraining

@LatFilosof Philosophers are useless.

Me reading philosopher A: ”yeah this makes total sense” Me reading philosopher B absolutely ripping A to shreds: ”wow I can’t believe A is such an idiot” Me reading A again: ”yeah this makes total sense”.

@one_miloo You can achieve wonders when your manufacturing plant labor is only $4-7 per hours.

At this point, every country should be studying what China has done with manufacturing and technology.

@BeijingDai Americans not holding G2 status is just delusional.

We've seen Huawei's chip breakthrough and DeepSeek before, but I get the feeling that this time — KIMI — has hit Americans harder psychologically than anything before. I can feel their panic. Up until now, Americans had been telling themselves: "We're in a G2 with China. Sure, they might have us beat in manufacturing, but we're still ahead in AI and other cutting-edge tech." Now they're waking up to the fact that even in AI, China is about to leave them behind. And if that's the case, then it's every single domain — literally all of them — where America is about to fall behind. Americans can already see the writing on the wall: they won't even be able to hold on to G2 status.

@TheWizardTower I mean a 6x slowdown almost sounds like the project could have been rewritten in Java. Why even bother with the nastyness of memory management if you don’t even need the performance 🤷‍♂️

Rust's community is far from perfect. But fil-c, zig, and the like insist on flatly lying. TO BE CLEAR: It's not that fil-c's innovation is not without value. It is! A runtime exception and a 6x slowdown can *very well* be the correct trade-off in some cases! That's a worthwhile achievement, worthy of laurels, and should be an available trade-off to make! Sometimes rewriting legacy C projects in principled languages is outside a team's constraints. But trying to masquerade "Runtime panics" as equivalent to "Compile-time error" is either delusion or fraud. It betrays either a fundamental misunderstanding of the facts at hand (a category error, if you'll forgive the pun), or a belief that the audience is an easy mark. @Malix_Labs's question of "Do you believe it?" is *exactly* the correct one to ask.

@GrayBeardPCG They may reposses the house whilst letting you lose on the down payment and any equity built up.

You're here on a 3 year work visa, why the fuck would any bank approve you for a 30 year mortgage?

@airkatakana Singularity is when claude tells u P=NP is so ez proof is left as an exercise to the reader.

the singularity hasn’t really started in earnest until claude proves p = np

@Duderichy As I understand it. It won’t be so “trivial” because we already have computers searching counter examples 24/7 and it’s still true for first trillion cases.

Imagine if ai comes up with a trivial counter example to the reimann hypothesis

It doesn’t “guess” the words directly. It guesses the ink strokes without any meaning and the human sees that it matches the glyphs. Across many glyphs, you can now match the grammar and dialect. It seems unlikely that all this becomes coincidentally wrong but aligned at the same time.

@chainmonkey2 @DAKKADAKKA1 The issue is the accuracy of what it unravels. Any AI model is based on interpreting data to then GUESS the correct answer/output. While AI is extremely powerful nowadays it’s a valid concern that the AI made mistakes when GUESSING what is inside the scroll

How do we know the ai just isn’t making shit up?

@kgorin @nixcraft You’re welcome to contribute

@nixcraft at least half of those are pointer related... rewrite the damn thing in Rust already

432 Linux kernel CVEs published in the last 24 hours

@bigboomtx @pulmencr Maybe you just fell for a tech twitter bait.

@pulmencr Has anyone here got this to actually work? I have nine ESP32 boards in my office and Wifi DensePose running. I get the telemetry and signal disruptions when I walk into the room, but actually rendering the human 'model' in relation to my body position isn'y working.

A guy just turned standard home Wi-Fi into a radar that sees through walls. Over 62,000 stars on GitHub for a system that tracks humans without using a single camera lens. It doesn’t just know you’re there. Watch what happens when a person lies down on a bed in another room. The AI instantly maps their exact posture, tracks their breathing, and calculates their heartbeat in real-time. No optical surveillance, no privacy leaks - it just analyzes how your body disrupts ordinary radio waves. The whole thing runs on a tiny $5 microchip. Traditional security requires a direct line of sight and invades your personal space. Now, your Wi-Fi router is a camera that sees through solid brick.

@Tim_Dettmers You need to troubleshoot the issues and fix them gradually by adding tools etc. it won’t work out the box

Do loops actually work for anyone? I tried them for months, and even with the best specs or a well-defined goal, there is always some drift that leads to lots of wasted work that needs to be unwound. I have had much more success with single-agent sessions running overnight

@jun_song The “insanely expensive admission fee” of $20 bucks a month.

Open Source AI builds a library using human knowledge and lets everyone access it freely. Closed AI builds a library using human knowledge and forces everyone to pay an insanely expensive admission fee. That is the difference.

@DataChaz We must not be using the same Google because the AI section is quite useful

Googling stuff, before and after AI 🫠

@a_musingcat The AI coding has definitely sped up my hobby robotics project. No more monkeying around esoteric hardware issues.

it’s crazy how much code people are writing with AI and how much it literally hasn’t changed anything in life at all for anyone. but it is a lot of code