carpenecopinum

Given the recent gripe that Bun/Anthropic indicated regarding compile times with Zig (i.e. that their vibe-coded 4x compilation speedup PR wasn't accepted), it appears to me as an "interesting" move to switch to a language that probably delivers 4x longer compilations than even vanilla Zig.

This is honestly the first thing I look for with anything new claiming "CAD".

Roughly every other week there is a new "The (programmable) CAD that fixes everything!" post on the front page, just for me to open them up excitedly and noticing that they use a mesh kernel and will thus never be able to provide fillets and chamfers painlessly (for the user). All while they are absolutely essential for a lot of designs, especially in 3D-printing, a well-placed fillet/chamfer can make the difference between an object that breaks upon looking at it funny and one that can bear significant load.

Just tried it out with GrapheneOS on my Pixel 7 and the eBay app works fine.

Generally I think the marketing of GraphenOS is interesting. They are usually positioned as the "Absolute security. No compromises." ROM, but in my personal experience, they are the "It must work. No compromises. Then make the rest as safe as possible"-option, given aspects like using the "real" Google Play Services (if desired), but sandboxed, instead of MicroG or unrestricted Google Play, which pretty much all other ROMs roll with.

I mean, GrapheneOS hits at least 2/3 of your demands pretty well. The Play services are "regular" apps with permissions that you can take away. For contacts and files you get "scopes", i.e. you decide what the app can see, while the app is left to believe that it can see everything there is.

That said, I think the marketing of GrapheneOS could be better. Every introduction of GrapheneOS I've seen paints the image of Graphene being "Absolute security, no compromises", whereas in reality GrapheneOS is the most "Things need to work, no compromises. Then make the rest as safe as possible" custom ROM that I've used thus far (in particular regarding them allowing you to install Google Play, rather than using MicroG).

There is the option to register the signing key of the ROM with the bootloader and then relocking it, thereby making those apps happy again.

The biggest issue is that there is a different way to do this for every device, so most custom ROMs don't bother. It's relatively simple and automatable for Pixel devices, so the GrapheneOS installer takes care of it. e/OS/, which is based on Lineage, allows this for some devices, iirc.

That's pretty much my impression of the post, too.

It has a spark of profound thought "Generative AI will obscure the path to truth" combined with a lot of conspiracy-theory-grade, flimsy analysis.

There is no legislation here. CUII is a private organization that generates lists of domains that contain copyright violations. ISPs voluntarily choose to block those.

Besides being a great project, this device reminds me a lot of the Technifant (https://technifant.de) which, under the hood, is similarly simple. The "hats" that you put on the Technifant contain a cheap USB thumb drive with the contacts soldered to a magnetic connector. All the audio is stored locally on the hat itself, and you can even add your own MP3 files with a cable provided by the manufacturer.

For me personally at least a few aspects about this are efficiency and control:

The number of CPU cycles my current android phone burns through just to boot and get ready to accept my "first useful input" is probably in the same order of magnitude as or higher than my old N900 would use for the entire day (600MHz single core vs. 8 cores at several GHz). Yet somehow the N900 could easily run quite a lot of things in parallel and would still react quickly to inputs, while I decided to get rid of my previous (still several times more powerful) phone because it would regularly hang for 10 more seconds without any good reason (also there were no more OS updates).

Also with the N900, I had control over every aspect of the system, I could easily script things in python without installing a huge app for it, which the OS would decide to randomly kill to save battery, etc.. Closest thing you can do on Android is root your phone and now every second app complains what a horrible person you are for wanting a bit more control over your own hardware.

That being said, I too eventually buckled to the fact that all the software you need to make a smartphone useful/entertaining is pretty much only available for Android and iOS. And the most realistic way to get "Android-compatibility" to a Linux phone is to just ship an entire Android build with it, due to how interwoven things are on Android.

I have a OnePlus 6 becoming "free" soon and I will definitely give PostmarketOS a shot (I had a glance at their compatibility list and noticed the OP6 is on there). Thanks for bringing this to my attention!

Now imagine that knife stabbings became so common that almost everyone started wearing body armor and you start selling body armor defeating knives explicitly. I can honestly see why most people would be upset about that.

In a roughly two-week window of not having adblock enabled, I have seen a ton of "politics-adjacent" ads in Germany. They'd usually start off with something that will get right-wing nuts excited ("This Green Party politician ruined their career, learn more.") and will then pivot to some crypto-scam.

I'm wondering if those will be affected by the new ban as well, or if crypto scams aren't political enough to apply.

1) I'm reasonably confident that this issue is not an accident. Better compatibility / better specs won't help here, I'm afraid.

2) A reference implementation for browser-features is an insanely complex project. Already there are effectively only two entities on the entire planet who can produce a browser that is reasonably close to the current spec. If you forced a reference implementation to exist, it'd probably just end up being Chrome(ium), which is arguably an even worse situation than where we are now.

In most videos, there'll be a whistle-blow that marks the point where the life guard in the video reacts to the drowning person. Upon clicking the right location, you are "graded" based on how quickly relative to the life guard you reacted. You get the "answer" by looking at who is getting rescued.

In my case, it started with a video that appears to be broken (?) the video just ended without any life guard reaction, but you can hit "Play Another Video" which appears afterwards.

I like the general idea, but the page you linked explicitly states that it isn't meant to be used to hide (user) IDs, or actually encrypt data.

I have to admit that those dark patterns on Amazon already saved me a good amount of money. For every order, I have to confirm that I don't want Prime (an estimated) 3 times now. Usually between the second and third click I'm reminded that I should shop around and to look for a better deal elsewhere (and usually there is one).

In that sense, thanks for your greed, Jeff.

Which basically means the same thing, as the main use of crypto right now is to avoid financial regulation.

TikTok and Instagram both limit comments to a single layer of "answering".

I guess that's a "good" alternative if your goal is to prevent the evolution of deep discussion and instead get people to move on and scroll over more ads.

The big issue here (provided that I'm reading the PR correctly) is that it's purely for arrays of the primitive number types. Whereas most real applications (that I've seen) will be sorting objects by some (potentially computed) property, be it an ID, a timestamp or a name. For all of these cases, the linked pull request won't be doing anything useful. The most useful application for this (outside of getting nicer numbers on a benchmark) that I see is computing the median/quantiles of some property on a bunch of objects that aren't already sorted by the interesting property.

I brew it just like any regular herbal tea and I'm pretty happy with it. The whole ritual with the pipe and gourd is a bonus, but never required.

If you dislike the more bitter flavors about it, try steeping it at a lower temperature (like 70-80C) and only for a short duration (2 or 3 minutes). That should make it a lot more bearable to you. You can also get packages that have less or no stems or powder, which both add to the bitterness.

There is still a difference between leaving out special cases to give more succinct advice and going out of your way to repeat the words "never"/"ever" three times to imply the absolute absense of any exceptions.

That's a big thing with Julia actually. A lot of people first getting into it criticise it for its use of 1-based indexing (I'm not the biggest fan of it either tbh), but in reality, there are very few cases in which you actually directly interact with array indices at all. And even if you do, the idiomatic way to iterate over the indices of an array is to use the `indices`-function, which has the benefit of working with multidimensional arrays, or arrays that start with an index other than 1.

> Cut the AI off humans' work, let it interact with the real world and see what it produces. It will be nothing.

That "experiment" could just as well be done on humans, though, cut them off of any work that any human has done before and you may get simple cave paintings, if you're lucky.

> ...was based on countries’ short-term goals for the next decade

What else though? There is no "long-term" for politicians that have to face another election in less than 4 years.

"This will happen in 10+ years" is just a thinly-veiled "We aren't going to do it.". If we can't get any short term commitments and hold people accountable for it, nothing substantial will get done.

Nah, it's definitely true. I have seen plenty of (especially larger-format) TVs where, with "game mode" (or the respective equivalent) disabled, it's unbearable to even do latency-forgiving tasks like office work on them.

> But they are turns of a circle of a given radius. > Radians (circle fractions)

It's the other way around actually. A turn is a turn, no matter the radius of the circle. Radians are the length of the line you need to draw a fraction of a circle with a radius of 1.

(In the end, both are just ways to describe angles and thus independent of any radii... the only effective difference between them is a constant factor of tau or two pi.)

The standard library of Julia, especially with the possibility to just enter `@code someFunction(params)` in the REPL to see the source of the method that the call would be dispatched to, the source can basically function as a second layer of documentation (the first layer being docstrings that are part of the source as well, but are usually accessed either via the IDE, or the help-mdoe of the REPL).

In principle, yes. With webGPU (but also WebGL2) we get access to compute shaders, which makes it feasible/easier to evaluate a model like Stable Diffusion.

The biggest issue I see is that those models (or rather their trained parameters) are usually pretty large (several GiB). So it'll take a while to set up in the browser before the evaluation can actually start. It'll also require a lot of bandwidth on both ends.

A lot of those things should already be doable with the fragment shaders we get from WebGL and a lot of hackery, like clever (ab)use of textures. So the actual issue that we're not seeing a lot of this is probably not due to it being impossible right now...

Considering that the water would add a significant amount of weight, a chair that works in reverse would be kind of interesting: Arrives as a flat, light piece and once you placed it in the room you need it, you sprinkle it with water to make it unfold.

Might even help against back-issues for office workers as they'll have to interrupt their work to re-moisten their seats every so often.

Each community decides among themselves which technologies will help them strengthen their community rather than weaken it and thus be allowed to use.

That's how you end up with things like horse-drawn carriages with electric lighting or even Amish that use computers for work.