makoto12

The interesting part here is that Tesla seemingly used AI image generation to reproduce the images that they couldn't get with permission to try and skirt around copyright

This wouldn't have made the front page if it was: "How to not store your secrets in plain text"

Absolutely disgusting. This is just plain evil. Absolute stain on Denmark society

South African actually!

Anecdata: my neighbour runs a doggy day care. She's been flooded with 1 star reviews (clearly not from her customers), and received similar coercive phones calls to help her improve her online presence. Not much she can do about it, as Google is not particularly responsive

I've been following work metrics quite closely this year that has this problem. I don't know why I didn't think of this!

i'm actually quite happy that he died without me noticing

If threat to society is the only purpose to put someone behind bars, then a large swathe of the incarcerated population wouldn't be behind bars. You could make the same argument for Bernie Madoff, who probably isn't a danger to society considering his fraud is out in the open.

But the morality is completely different. First one is beating the casino, the second one is marketing gambling to create new generation of degenerates.

You want to start with narrow band. So object tracking, where the object is emitting a single frequency is a great start.

i think at 2khz (it is blazingly irritating to listen to), you need to space your microphones about 15cm apart. so you don't need too much room. Then it's just hooking up 4-6 microphones to a wooden board. Sampling with a decent microprocessor. You can do the processing on board, or ship it to a laptop, and do the processing there.

Just take care with your microphones and sampling. Any large defects will greatly inhibit your results.

Then for the algorithm, do a delay and sum. it's the dumbest approach, but the easiest to understand, and should work

Yeah the intuition, is that the sphere of the signal is so large, that any small section of it can be approximated to a plane

The article (and what is generally talked about with phased arrays) generally focuses on radio wave based phasing. This is pretty inaccessible to average hacker. However it works just as well with audio, and can be rigged up much more easily. You can get quite solid results and learn quite a lot about DSP in general

We are also moving away from traditional bowling pins because the mechanisms are too expensive and hard to maintain, and moving towards pins that are attached to strings, to cheaply reset

bit rich to call them beggars, when most expect their news for free

My company merged with another company. Both had a monolith. Together we now have 2.

Then you'd have multiple monoliths share the same database. For me that's where I've zero experience, because all engineers more senior to me have cordoned that off and have said that's a no go area.

But in all seriousness, that would make data migrations really tricky, as well as handling differing versioning etc. NVM extending that to more apps, feels like a dangerous venture

If you have 2 monoliths both using a shared library, where do you store the data?

I agree that microservices need to be considered seriously with real engineering thought and not jumped onto to pad CVs. They are a tool just like anything else, and need to be used carefully. Microservices have a particularly sharp edge.

However libraries feel like a very simplistic argument against microservices. In a world where you have a single monolith, then libraries may work fine, but often you have 2, 3 maybe 4 decently sized monoliths. In that situation libraries may not make much sense when you need to reference a central source of data.

This is the greatest thing i've ever watched. thank you

i'm only aware of the one supporting cape town. do we have more for the main grid?

it's not maintenance if it's for aesthetic purposes. unless there's a serious concern for rust then i don't think it classifies.

could be intentional. Implying LLMs are a proverbial nuclear bomb to the tech landscape. but honestly it threw me as well

If you sandbox your connection to openAI correctly, then you can get the benefit of a llm without making your application look silly at the same time. Identifying the correct places in your business to use it is tricky, but imo it certainly makes sense in a lot of specific areas. Just not a catch all that can run your business for you

Previous conversations https://news.ycombinator.com/item?id=30219442 https://news.ycombinator.com/item?id=24694459

I'd rather not

true. Personally I think the concept of public/private domain is a good thing. But like all good things, they get abused by bad actors who are huge corporations. But the artwork and texts used for training these models are often coming from small independent creators. The same creators who will be sidelined by this new technology. Bitter pill to swallow

I'm more referring to the companies offering the tools. While some are free to use (for now), they are still profiting from this type of personal use. often being able to convert to a paid model later down the line, think github copilot

beautifully put

Mathematical concepts are in the public domain. The art used to train these models are not

Do you think it's moral for a big corporation to profit on pirating works from small artists?