> People matter more than AI. The whole document in 5 words.

> The idea of model welfare is wrong. AIs should not have rights or legal personhood.

Microsoft just told Anthropic to STFU.

Brutal pricing. It would have been brutal at $600.

Having said that, there really is no other game quite like Half Life Alyx. It's so good, immersive and one of a kind. Perhaps worth the price of admission. It's so good that it kind of ruined VR gaming for me because all other games kind of felt like tech demos from that point on.

Yikes, what a terrible price. 55 minutes playing on-device Half Life Alyx. 1hr 30min streaming from a PC with dongle. 3 hours charge time. It's just pretty lame tbh. Damn :(

Is it really that bad? The hardware processing and storage is effectively a S24 which retails for $800. Add the additional displays, cameras and controllers, an extra $250 doesn’t seem that out there.

Whether it's a brutal price is separate from where it's fair for the hardware costs that went into it.

I don't want to buy that high end cell phone. First because I already have one (it would be very nice if I could have my phone handle compute from the back of my head or my pocket), second because I mostly want the headset to use with my desktop computer. $250 or $500 for a dumb headset with these features would be great.

Also charging 2-3x slower than you use the battery is very lame.

Eh, I don't think that's how the maths works, especially for products that are in completely different categories. For starters, display costs relatively much more on a VR headset compared to a 6" phone screen, the unit cost is much lower for a phone that is sold in many millions, the marketing cost is different etc. There will likely be actual breakdowns that provide some perspective into this.

Sure, it's not one to one, so the fact that the headset is so close to the retail price of a presumably much cheaper per unit flagship phone is why I don't really think the pricing is that bad for what it is.

I played it on my Quest 2 so that's not really the price of admission

How much of Google's revenue is just scams?

Search for "datadog" and the first result is a sponsored link for Datadog. I click that and datadog had to pay google. scroll a bit down and there's the "real" natural link.

Fugazi!

Otherwise their competitor gets top billing for their very own brand name. See this on Apple’s App Store too. Feels very unfair.

They’re selling a limited ad-space - again, instead of to their competitor.

In that context: suggest a fairer way.

Now, it may just be you don’t like advertising in this context - a valid take. But a different one!

The fairer thing is what Google _used_ to do: organic results in the main column, ads in the sidebar. Searchers get to see competitors without confusing them for the real thing.

If someone searches for a specific product, you could still have competitor ads in a relatively fair way, but you have to put them after the correct link. Putting them first is going to mislead so many people.

And you could advertise related products, or unrelated products. So many options there.

Mandating Google choose the “one right” link is the failure point, here. (Not because I disagree.)

Similar problems with mandating that they track what’s considered a competitor.

> Mandating Google choose the “one right” link is the failure point, here. (Not because I disagree.)

I'm talking about the actual search results. They already do that part.

> Similar problems with mandating that they track what’s considered a competitor.

Well I said fair not mandatory, but they make enough money off these ads that they could handle it being mandatory.

Besides “I’m Feeling Lucky” and new AI offerings, Google specifically does not purport to tell you the one true link!

It ranks based on likelihood of what you’re search for - and yes, that’s different!

Feels unfair. Guess humans could be hired to try to prevent the situation in cases where it was on the obvious side of what was happening. Would increase the cost of ads or reduce ad network profits, and feel more fair?

Major loss for react native community at large with Skia and Flashlist dying. :(

Legend List is the new hotness in RN.

iOS video playback is pretty highly optimised and hardware-assisted, though, right?

Not sure you can really extrapolate non-video battery life from video battery life. I have an ageing iPad and I can get a couple of hours of video when the battery is at 30%, but it would be gone in 45 minutes if I was browsing.

Depends on which video codec. If it’s the most efficient one that has dedicated silicone embedded in the chip, it’s going to work great.

Web browsing is typically significantly harder on power. Plus these benchmarks usually mean in a perfectly dark room with the backlight at close to zero.

I had no idea, I thought video was the most expensive thing to burn battery on since it's video, full screen motion, and loud sound. Interesting!

Apple quotes 30 hours of mixed use or 45 hours of video playback. So video playback is almost 50% more efficient than general use.

It is gaming if you want to burn through your phone battery.

It's like electric car ranges. Their claim will be 45 hours under a very specific testbench. Real life results will be lower

I only lasted two turns before I had to close the tab. I can't stand anthropic models.

Does tailwind even make sense in the age of agentic engineering?

Tailwind was created to mitigate HUMAN problems, and HUMAN risks. Are these same problems experienced by an agentic workflow? I'm not so sure. The calculus is different.

Knowing Facebook's behavior around user data why would anyone with memory of their behavior use this product?

Why would you give them access to your accounts, your email, your purchase habits, your preferences, your communication style and more.

How do they wash this stink off?

And on top of that, we've all seen how confidently LLM's get things wrong. I've seen them get better, but even the current SOTA models fumble. Why would you want to send a bot a message "sure, reschedule my flight" and have a bot potentially fumble that in all sorts of ways? I really really do not see the appeal. My calendar and my emails and my things to do are all things I want to be on top of. Not have it half-assed or potentially messed up by a chatbot made to harvest all your data.

Or worse, take money to steer you to a specific airline.

"DO NOT COMPRESS ANY DATA WITH THIS PROGRAM UNLESS YOU ARE PREPARED TO ACCEPT THE POSSIBILITY, HOWEVER SMALL, THAT THE DATA WILL NOT BE RECOVERABLE."

So why would anybody, hobbyist or enterprise, use this? Or does something older like 7zip also have this caveat that I've never experienced.

Tool author here. bzip2 also has this clause, in fact it has been lifted from its dist tarball README verbatim. So does lzma, xz, or in practice any open source program that you use.

Hey, I am the creator of https://www.photopea.com and if a user decides to publish their work, it compresses PSD files (inside a browser) and sends them to the server for storing. Right now, we have about 2 TB of files. I use the DEFLATE compression which I fully implemented myself (compressor and decompressor, inside a library called UZIP.js).

I wish I could use something better :D but it must be implemented in Javascript to run inside a browser, and the library should not be too large (mine is about 8 kB ZIPped). I would love to switch to ZSTD once someone implements a compressor in Javascript that is under 40 - 60 kB. Or do you think that bzip3 could be re-implemented easily in JS using some AI?

I really admire you for being so smart at such a young age :) I wish you best of luck in your scientific career! I visited Warsaw this year, but I am not sure if that is where you are from :D

Hi Ivan,

Please send me an e-mail! Perhaps we can figure something out.

Is that all that different than the standard MIT License:

> THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE....

Practically, no. But that specific disclaimer could lead on to think that there may have been some observed data corruption in practice that isn't disclosed. I have no idea if there is, but I'm not keen to discover it myself.

Also practically, this isn't MIT. It is LGPL 3, which I believe includes the warranty terms of GPL 3, i.e., no warranty. So we're in the same place anyway.

This big warning gave me pause, too. Why not have a compression option that automatically checks the data after compression (by decompressing and checking against a hash of the original data), which would reduce the probability of undetected errors to that of a hash collision? i.e. like `7z a` followed by `7z t`, but in one command.

Whatever you do, don't download nzget, and don't download sonarr/radarr, and don't download plex media server, and don't use AI to search for the best device that support "emby direct playback x265/AV1", and don't connect that to your media server, and whatever you do don't invite friends and family to your server, and do not install a media request app so they can just request things to automatically pull from The Web.

Definitely, definitely do NOT have an AI assist you doing all this in 1 hour on your computer.

Shhh, do not talk about *senet!

You can get usene* from TorBox in the pro plan but I've never had a problem with the cheapest plan. What's the advantage?

my bad you're right. edited.

is there anything decent actually which is not on the internet? I thought it's an outdated place

It’s not that the content is not available on the Internet it’s that it’s a better experience. Indexed high quality content that can be retrieved automatically and at high speed without the need to seed or worry about copy strikes or deal with ever changing, poor quality streaming sites.