_ea1k

6 ms·

31.
by _ea1k 9mo ago
Tesla does something similar as their USB port for flash drives is in the glovebox. It is intended to make the dashcam drive harder to steal.
32.
by _ea1k 9mo ago
> I'd love to know what the justification for replacing them in the first place was. It isn't hard to see, tbh. Think about the controls in a Tesla from a few years ago. They had physical controls for drive selection, turn sign
33.
by _ea1k 9mo ago
These posts fascinate me, as I must admit that I never really realized there were so many people focused almost exclusively on promotion. Is it really that common, or do these posts just bring out those types of comments? TBH, in this field
34.
by _ea1k 9mo ago
3blue1brown videos are great if you want to go deep on the math behind it. If you are struggling with the neural network mechanics themselves, though, I'd recommend just skimming them once and then going back for a second watch later.
35.
by _ea1k 9mo ago
Yeah, I'm getting the same feeling. They've announced that semi will use 4680 and Cyber Cab as well, right? If that's the case, this would point to a specific supplier issue rather than something more general. It isn't s
36.
by _ea1k 9mo ago
The means of collection and treatment of it as something other than tax revenue are problematic for sure. Those should be solvable problems, though. Your point about semi-truck damage vs lighter vehicles is exactly why I think moving in tha
37.
by _ea1k 9mo ago
I don't necessarily see this as a bad trend. Eventually a tax on mileage and weight would make the most sense vs the current attempts to use fuel taxes as a proxy for those things.
38.
by _ea1k 10mo ago
What an absolutely repugnant article this is. It is complete slop. Is this what passes for HN worthy today? :(
39.
by _ea1k 10mo ago
AFAIK, the data center employs more people. I'm not really sure why that's the case, but neither is >1k. I'd guess that this is also an area where the perception makes a bigger difference than the reality.
40.
by _ea1k 10mo ago
Years ago Google built a data center in my state. It received a lot of positive press. I thought this was fairly strange at the time, as it seemed that there were strong implications that there would be jobs, when in reality a large data ce
41.
by _ea1k 10mo ago
It isn't perfect, but it has been better than Python for me so far. Elixir has also been working surprisingly well for me lately.
42.
by _ea1k 10mo ago
It probably means that the text was generated by an AI. Claude Code loves to say that everything is production ready, even if it doesn't quite compile or pass automated tests yet.
43.
by _ea1k 10mo ago
Waymo will get better at this. But even without them getting better, as far as I know there were zero waymo fatalities due to this. That's more than I can say about Helene, where there was at least one fatality due to traffic light out
44.
by _ea1k 10mo ago
I often find it sad how many things that we did, almost without thinking about them, that are considered hard today. Take a stroll through this thread and you will find out that everything from RAID to basic configuration management are ult
45.
by _ea1k 10mo ago
Honestly that last paragraph is absolutely true. In general, you shouldn't have to do anything. If your website is hard for an AI like Claude Sonnet 4.5 to use today, then it probably is hard for a lot of your users to use too. The exc
46.
by _ea1k 10mo ago
It seems like verification might need to be improved a bit? I looked at Mistral-Large-123B. Someone is claiming 12 tokens/sec on a single RTX 3090 at FP16. Perhaps some filter could cut out submissions that don't really make sense
47.
by _ea1k 10mo ago
Yeah, I had the same experience. It made the page basically unusable for me.
48.
by _ea1k 10mo ago
It is at least as reliable as the data in the electrek article. My point is that the data naturally has error margins that are clearly large enough to make drawing concrete conclusions impossible.
49.
by _ea1k 10mo ago
I spent a little bit of time poking at Gemini to see what it thought the accident rate in an urban area like Austin would be, including unreported minor cases. It estimated 2-3/100k miles. This is still lower than the extrapolation in
50.
by _ea1k 10mo ago
Somewhat amusingly, the human rate should also be filtered based upon conditions. For years people have criticized Tesla for not adjusting for conditions with their AP safety report, but this analysis makes the same class of mistake. 1&#x2F
51.
by _ea1k 10mo ago
Yeah, I'm glad that they are trying to do a rate, the problem is that the numerator in the human case is likely far larger than what they are indicating. Of the Tesla accidents, five of them involved either collisions with fixed object
52.
by _ea1k 10mo ago
Yes, regardless of the product quality, this article really pushes hard to find a negative narrative without thinking about alternative explanations. There are real, legitimate concerns here, but they are hidden behind a lot of assumptions.
53.
by _ea1k 10mo ago
Yeah, this company went through an amazingly bad period. They quite innovating, and also worked really hard to segment their products in a way that would extract every last $ out of the consumer. "Oh you want it not to run into things?
54.
by _ea1k 10mo ago
I remember hearing Karpathy refer to these outages as a worldwide "intelligence brownout". Crazy: https://www.youtube.com/shorts/SV4DMqAJ8RQ
55.
by _ea1k 10mo ago
> Why pay for a keyboard app when the default keyboard is already good enough? That's probably what people would have said before Swype was invented too. But lots of people use that in their default keyboards thanks to the people th
56.
by _ea1k 10mo ago
I suspect that people not really paying for certain things has had an impact. Remember when there were a lot of high quality, paid keyboards for Android? I doubt those were particularly profitable, but there was a lot of innovation back the
57.
by _ea1k 10mo ago
Honestly, I think this is a valid viewpoint, but perhaps C is too low level. The bottleneck in generating code with LLMs tends to happen at the validation step. Using a language that has a lot of hard to validate footguns isn't great.
58.
by _ea1k 10mo ago
I feel like this is an artifact of some limitations in the training process for modern LLMS. They rarely get enough training to know when to stop and ask questions. Instead, they either blindly follow or quietly rebel.
59.
by _ea1k 10mo ago
Yeah, it is insane how many people think that tuning models is nearly impossible, or that it requires a multibillion dollar data center. It is one of the weirdest variations of people buying into too much hype.
60.
by _ea1k 10mo ago
Why would you do that when you could spend months building metadata and failing to tune prompts for a >100B parameter LLM? /s

More ›