l33tman

It's an approach to a hypothetical other type of universe's gravity..

The optical nerve has an information density of around 10 MBit/s (ref https://pmc.ncbi.nlm.nih.gov/articles/PMC1564115/) Concentrating on only the symbolic thinking speed seems to be unnecessarily restrictive..

I think I know which you mean, but TBH that paper read as an AI auto-generated troll paper that any physics undergrad should be able to dissect :) It was a bit fun to watch though and sure sometimes you need to provide some fun content as well!

TBH you need to take the youtube influencer Sabine Hoffenfelder with a bigger grain of salt. She has converted to mainly posting clickbait youtube stuff over the last years (unfortunately, she was interesting to listen to earlier).

The RCS is a common benchmark with no practical value, as is stated several times in the blog announcement as well. It's used because if a quantum computer can't do that, it can't do any other calculation either.

The main contribution here seems to be what they indeed put first, which is the error correction scaling.

How does the Macbook check the ambient light to determine the screen brightness, does it have a separate ambient light detector buried under the screen somewhere or inside the camera? (if the camera is not used for this I mean which would have been the best thing to do, but would require the camera to grab frames now and then without the LED flashing)

Maybe that says more about the QC field than pop media though? Shor's algorithm truly implies both an algorithmic and practical breakthrough (if possible to implement, the small if). But what other algos are there? It's this, quantum fourier transform, and Grover's like you say. What else can be cool and useful? What is the bulk of the active development, except error correction codes? What should someone start with now if they just enter the QC field?

The analog computer is a quantum system (of atoms for example), and your argument rests in that the complexity of simulating a system of atoms is not high and that nature (i.e. the "analog computer") isn't performing something better from a complexity standpoint.

While I agree that there are pretty efficient algorithms to simulate systems of atoms behaving like Newtonian bodies (like molecular dynamics) it quickly falls apart when you need to include more "quantum like" effects.

The digital quantum computer is itself built out of atoms, so if you could simulate an analog computer (which derives its properties from QM) efficiently, you could then also simulate the digital quantum computer efficiently.

I guess it's a question of semantics - like what properties of nature are an "analog computer" allowed to tap. My point is, I think, that you kind of downplay the capabilities of what an analog computer is capable of by saying it's "just" an analog computer. But any small molecule whizzing around is in some sense a very difficult to simulate analog computer from another point of view.

Thanks.. but that github program only lists the same info you get if you command-click on the BT icon in the menu bar. It basically only shows the device name.

I guess filtering the streaming log entries in the Console app gives some info.

As they seem to have removed Bluetooth Explorer and all ways to get diagnostic info about the bluetooth system and/or change codecs and settings, does anybody know any good cmdline ways in later mac osxes to do the same?

For example I'm having a problem that comes and goes now and then where Bluetooth audio is 300 ms delayed compared to the video playback everywhere except in Youtube on Safari, very strange. It's good for a few months then suddenly it becomes unusable, then back to zero sync delay after a few months.

I was thinking this might be related to CODEC selections etc or some hidden setting that might get changed which we normally aren't allowed to determine :)

(btw I know there is a difference between latency and synchronization - latency might be unavoidable but video sync should always be able to compensate - I got curious on how exactly that works, where in the app / SDK / OS pipeline does the a/v sync happen on a Mac?)

Wondering the same. The same column lists a bunch of Windows-only CVEs where an unprivileged user can do stuff, so there has to be some difference between those (CVE‑2024‑0117 - CVE‑2024‑0121) and the headliner CVE‑2024‑0126

They mention hypervisor breaches further below, so could the CVE 0126 imply that a local root user on a shared GPU machine of some sort can break out of the virtualization?

Lattice QCD is a rabbit hole (like all other advanced subjects) but it's IMO a somewhat conceptually more accessible rabbit hole than the canonical QFT formalism with the Feynman diagrams and perturbation theory. LQCD is built around a "clean" approach that is at least possible to visualize - you go back to the basics, which is that you need to do a sum over every possible combination of field configurations (as this is what it seems the universe does). This is obviously intractable even for a small array, but the concept is easy and the QCD sector of the SM is very clean in interactions.

The sum is intractable not because it's big but because it can't be sanely approximated by lesser sums, due to the nature of what you're summing - an exponent of i * the integral of the Lagrangian over each spacetime field configuration. This vary likes crazy so if you skip the some parts you might get a completely different result.

The trick to anyhow get anything out of this is to restrict yourself to observables that can be calculated from the ground state, which turn out to be accessible by doing the field integrals over imaginary time instead. This is a trick of calculus, it's not really "non-physical", although it is a pretty crazy method :) This has allowed practitioners to calculate hadron masses for example, it's getting better and better over the decades.

I think the point is that if your data is valuable enough for you, you can't really trust that option in the backup tool to work - maybe you misunderstood some config option and the test now isn't really run, the tool is broken, or is run only on some of the backup files or dirs etc... or your original config might have missed a folder because it was mounted through some other filesystem (happened to me with Borg actually, and my whole /home/user dir especially wasn't backed up for the first 6 months I ran it :).

Seems to be good to have another tool that you either manually or automatically can setup to run regularly that tries to locate random files from your existing file system in the backups? Something like that.. though that other tool might be broken as well of course... :/

What happened to that other research team who used amateur radio network packet metadata that had been saved, to track airplanes including MH370 and showed tracks that also matched the satellite pings? Can't remember the name of that group though...

For example the entire atomic composition of the stars in the observable universe depends exquisitely on the expansion parameters at the big bang. The ratios can be traced back through the expansion to the quark-gluon soup stage. Changing the expansion rates changes the delicate balance between the particles that form at that stage, and when the various particle fractions "freeze out" during expansion when the temperature cools (btw we're talking about seconds from the big bang here :) which can be subsequently observed in stars all over the universe by spectroscopy. It's pretty beautiful.

There are so many intricate dependencies between these pathways that it's pretty unrealistic to postulate anything else than a big bang + cooldown process IMHO.

Same here, been using the nvidia binary drivers on a dozen computers with various other HW and distros for decades with never any problems whatsoever.

That's not true for the most well-known models. For example Meta's LLAMA training and architecture was predicated on the observation that training cost is a drop in the well compared to the inference cost for a model's lifetime.

As someone who previously did research in MD and simulating protein kinematics and binding, may I ask what you who are in the field today think about the complement so to speak - doing synthetic structures from 3D templates? Like David Baker's lab (I haven't been keeping track, to me it seemed like he was the only one in the world going all-in on this).

Maybe unrelated to this, but I noticed fairly recently that my backups from my macbook now backup seemingly randomly modified pdf and txt files all over the disk. My guess is that whenever I search for something, it decides to touch a couple of hundred files (but not ALL pdf/txt files for some reason).

To be fair, this is already the case on all the platforms, as you can easily put together a game with free assets from the assetstores (or pay a few dollars for pretty high quality assets). For every standard game genre you can imagine I'm sure there are thousands of generic games released every year on every playform (don't have any real numbers but I get that feeling...)

Rendering the assets by AI or buying them from the asset store is not going to change the number of generic games put out there I think, maybe AI gen can make some of them a bit more unique at best.

Already discussed here a few days ago

https://news.ycombinator.com/item?id=40820063

I have a single 18.04 machine that is stuck on that, because it has 18.04 386, and as 20.04 doesn't support 386 anymore, apparently there is not a simple upgrade path to 20.04 64-bit (without doing extensive surgery). Very annoying...

I asked the CTO/R&D manager at a company I worked for once why he didn't bother getting talent or retaining talent, he said (paraphrasing): "Because I can't scale talent I might as well don't bother with it"

Many large corps are fine with being on the thick part of the Bell curve, their managers can easily scale if the board says they should output more, they are essentially linearizing their whole company around a point and they can scale up/down around it.

It works until someone else innovates and beats them, and then the loop repeats.

It's in some sense good that CxO's behave like this, because it leaves opportunities for smaller companies all the time.

It's just that from a more complete point-of-view, there are many other areas than rotation curves that depend on dark matter and has been modelled to really require the specific properties of DM that the lambda-CDM model has, and it's just a bit worrisome every time a blog post talks exclusively about rotation curves. Sounds good that he writes about more of the issues elsewhere! The real picture might require a new blend of both DM and MOND of course..

Consider also that the author of the linked blog post seems to be a (pretty strong) advocate for MOND, it did read as somewhat biased

Please explain. I've only ever seen handwaving in pop-sci books like Sean Carrolls suggesting this (something about that the Born-rule was the "only possible equation" which seemed somewhat tautological to me), and IIRC this wasn't very encouraging either because it didn't come with any experiments that could be done to show a difference to the standard methodology and then you tend to pick the simplest method which is the one everybody else uses every day.

Would be cool if there were of course. I'm not saying it isn't interesting, I'm just saying I completely understand why most physicists don't bother as it doesn't bring anything new to their table - there are plenty of other practical quantum effects that require research as well.

Well, there are no experiments that have been done or even suggested that requires this. Except maybe the Wigner's Friend thought-experiments? They seem pretty interesting and strikes to the center of the evolution discontinuities issues.

But for all normal practical purposes you don't need to solve that and can ignore it (which is both a blessing and a curse of course)

All of the complications and complexity in QM/QFT are from doing calculations that map to the extremely coarse-grained boring low-energy reality we spend all of our lives in. If you ignore that, the underlying math is exceedingly trivially formulated. It's like 2 powerpoint slides to show what CUDA arrays and ops are needed to underpin a quantum field theoretic calculation of quarks and gluons inside a proton for example.

Unfortunately you're probably not going to be satisfied with that, at some point you need to connect these simulations to real measurable parameters and this is exceedingly non-trivial :/

Starting from the other end, pure old-school undergrad quantum mechanics is really the easiest way to connect to measurable quantities and this is why it's used in all of engineering, quantum optics etc. It just provides the right amount of detail.

You do have to assume at some point that if you fire individual electrons through the dual-slits and record their individual results, that the results are independent. If you can assume this, you're drawn to the current model which is that the electron is in a superposition (i.e. "interferes with itself") that when resolved gives the observed statistical outcome of your independent results. This seems also to be the simplest model (there has been 100 years of trying to find simpler models that reproduce these results).

It's because the Copenhagen interpretation comes with the standard method of actually calculating stuff. You might call it a cheap shortcut, but it works (so far).

There is no real resistance I think, it's just that the various interpretations are not very relevant in practice. All of engineering that depends on quantum mechanics make do with the standard treatment that has been taught since 100 years now.

The MWI doesn't offer any other way to calculate anything (as it's an interpretation, not a theory). It's a good mental image of what happens though.

I guess at some point there will be some experiment that can be performed that requires a more advanced theoretical method than what current QM provides and then maybe the MWI (or other interpretations) get a chance.