shoelessone

People have started to view a high-trust society like a rainforest: a natural resource that has lots of life-sustaining positive externalities, but you can just burn it down to make a quick buck instead.

I've never read this analogy before but it really works for me. Thanks!

I am not anti-AI, but if I am going to use AI I far prefer to have control over how I engage with it. Having a piece of hardware to focused on Google's own AI flavor being built in is a big negative to me. Not that I would totally write off this new Googlebook (despite disliking the name), but I can't really see a situation where I'd ever prefer this over an Apple Neo for example.

I appreciate the idea, but as others have mentioned it seems like for something like this to be useful it really needs to be well thought out and tolerant to extreme spikes in traffic.

I might be wrong here but it looked like the responses from the server are chunked, which I _think_ precludes the use of a highly optimized cache response e.g. from a CDN. Assuming that's true (very open to correction of course!) I wonder why this would be.

There are a lot of similarities. You must either have a great memory for fonts, or eat a lot of butter alternative spread, either way good eye!

I really don't know why I struggle so much with this stuff. I believe these models / agents / whatever write code that is often at least as good as the code I write, and they are super helpful tools, but it just feels like it takes away so much of the joy that is programming to me. I'm not saying it's "right" of me to feel this way, but for me the struggle, and the figuring things out by testing, identifying patterns, or looking deeper into a library's implementation (etc) is part of the challenge that makes programming and software construction fun.

I came here to ask if there is some way of fixing this. I'm guessing not.

Youtube lately seems particularly bad in terms of showing lots of "shorts" all over when I have zero interest in watching them, but also suggested videos seem somehow aggressively chosen. I'm not sure how to describe that, but that's what it feels like to me.

I don't see how this is an issue. To me, this does seem at least confusing, but possibly dangerous.

If you have internal auth testing domains at the same place as user generated content, what's to stop somebody thinking a user-generated page isn't a legit page when it asked you to login or something?

To me this seems like a reasonable flag.

I relate to this. Also, I am not the best person in the world, but recently this hit the point where I decided because of these very same thoughts + nudging from my much better partner to donate to NPR, to cancel Netflix and move that money to NPR. Now no more Netflix, which is sort of a relief in ways, and I have to be more intentional about what I download / consume.

Same. Replacing all of my light switches with an app I have to use anytime I want to turn the lights on or off is indeed a huge step backwards IMHO.

Replacing all of the light switches with smart switches and monition sensors and things like this, plus automated schedules, to the point that you never need to switch any switches at all or think about lights, that is nice IMHO.

I agree on all points.

The project was well made, but my read on it was that they wanted to be shown something interesting. Even if it wasn't as well made of polished, I got the impression they would have preferred something "fun" or imaginative.

Thanks for the link, but yeah as the other poster mentioned, this is for a dev board. I'd be interested in buying the module (which is somewhere between the bare IC and the dev board). Here is an example from their store of an ESP32-S3 module: https://www.aliexpress.com/item/1005006334720108.html?pdp_np...

Thanks for taking the time to explain!

So your original comment / request was regarding USB specifically then?

I ask because I'd have guessed (possibly incorrectly!) that implement HID via GATT (BLE) you'd be able to support anything the BLE hardware revision could implement?

Perhaps the disconnect for me is that it's unclear when there is some special hardware that exists within the ESP32 itself (I think I2C, I2S, etc would be examples of this) vs something you are just implementing by manipulating the IO pins. Perhaps HID is one of those things?

Any guesses as to when a hobbyist might be able to buy the module without the dev board? Their aliexpress store didn't have them as far as I can tell, I assume they are prioritizing dev boards for the moment unless you're a big enough company to actually talk directly with Espressif.

Any chance you could explain this to somebody who is just learning about HID and has run this example: https://github.com/espressif/esp-idf/tree/master/examples/pe... ? "non-boot protocol" I'm guessing is the key here? I don't have a super deep understanding of HID or what the "boot-protocol" refers to.

A bit similar for me, I grew up in the Midwest and spent a lot of time wandering around CompUSA and Circuit City / Best Buy when I was a kid. When I was old enough and had some reason to go out West, Fry's was one of the top things on my list of things to go see, it felt like a pilgrimage of sorts.

I completely agree. I find magic links much more of a hassle than a password.

Can you explain a bit more about what "serverless GPUs" are exactly? Is there a specific cloud provider you're thinking of, e.g. is there a GPU product with AWS? Google gives me SageMaker, which is perhaps what you are referring to?

I don't believe he misread. I believe he felt like the reply to the other users post was needlessly curt or snarky, which I'd agree with.

My intuition is that as contexts get longer we start hitting the limits of how much comprehension can be embedded in a single point of vector space, and will need better architectures for selecting the relevant portions of the context.

Is it possible to explain what this means in a way that somebody only roughly familiar with vectors and vector databases? Or recommend an article or further reading on the topic?

Can somebody help me understand as a casual Elasticsearch (and honestly, more OpenSearch) user how so much data inside of Elasticsearch ends up not becoming the "database"?

I've always understood you're not supposed to treat ES/OS as a database, so in my head I've always sort of thought of it as a cache. If I'm using it I need to be able to reindex (not sure if this is the correct term technically) all of my data from a "real" database, something (in my case) like a SQL database of some sort or another data store.

If I'm reading the article correctly it sounds like they are trying to get massive amounts of data out of ES, but it feels like that's something that would come from a "real" database. To put it another way, it seems like admin tasks that involve exporting data ES wouldn't be the right place to do that.

Now very clearly the author knows an entire world more about ES and I don't fully understand their use case anyway, so I'm hoping somebody might be able to help me understand if there is some gray area or uses for ES/OS that might make different use cases more appropriate?

Can do what my dad does to mitigate risk. He has a dedicated outdoor freezer, only thing is to hold dead deer (BEFORE processing). Has the deer tested by a lab before processing. Then processes himself.

I have no position here but reading this a few times (and not digging further for additional context) it's unclear of what the situation is here, at least to me.

... I cannot access any of mine to check at the moment ...

"at the moment" makes it sound like they were not on a computer, or out to eat or something where they couldn't easily check?

...I downloaded everything to my laptop before I submitted an appeal...

Makes it sound like they still have access to their account and were able to access the content in order to download a copy?

I’m with you. I would love this so much.

Maybe irrelevant, but I will say that my spouse who spends a lot more time in Google docs can’t be bothered to learn even the markdown shortcuts for H tags in docs even though it would save her a ton of time. I imagine that’s a large majority of people.

This is a serious annoyance for me all day long, across the internet. Have you found most sites respect your preferred languages list?

Thank you for sharing. My mind went right to the Murakami book as well, so it's nice to have the additional information you provided.

Possibly unrelated but similar idea at least: in Japan a lot of the Japanese TV shows I watch don't have English subtitles available, but in the US the same shows on Netflix do have English subtitles. I don't fully understand the reasoning for this on Netflix's side.

Focusing in on the contrived example re: native <form> behavior and nothing else, if given the choice I would rather remove the default behavior / "action" than adding more functionality to html.

I love html, and I love writing nice clean markup that clearly defines a document, so that you can look at the html without css and js and know what information it is presenting to you. I really don't want it to do anything else.

Fellow 2012 Fit owner. I bought it after reviewing carcomplaints.com and finding it was a very reliable car by their metrics, all the top problems are related to accessories.

Is the logic here that training a base model isn’t as easy or even possible in the same way that OpenAI did in the past, and that what they have in a trained model is valuable in that even with all the code and experience it couldn’t be reproduced today with new restrictions?

FYI some of your GitHub links seem to have been hijacked, might want to update. E.g. seem carver.

Cool personal site!