fph

I don't think card readers can display payment information, can they?

And I have no idea why, but no bank offers photoTAN devices in my country. They seem like an interesting concept, even though I imagine the underlying hardware isn't far from that of a phone, in the end.

It is not unrealistic at all. The Olympics are run by politicians, essentially, since they appoint the committees, make the investments, build the infrastructure.

And the ones pushing for these bans are the sport media tycoons: this fight isn't about Anna's Archive, it is about people watching soccer illegally. Because that is where the real money is.

Hardware tokens are not allowed in Europe to authorize certain operations such as bank transfers: you need a device that can show the operation you are about to authorize ("enter 123456 to confirm your payment of 99.99 € to Pornhub"). And that essentially means using a phone.

It is very ironic that the solution is using an old, insecure phone full of unpatched holes for all important banking and id business, because that one is vendor-allowed while your state-of-the-art GrapheneOS is not.

Why has no one mentioned Clippy yet?

Can you share a few examples?

Are you sure? You can post questions even with a completely new blank account. It's comments that require some reputation, maybe you were thinking about those?

Does the Rufus bypass still work after these changes?

No, not yet.

And if there is a prompt engineer, there must be also a prompt scientist, right?

I suppose this gets useful in applications where you can change the font, but not add syntax highlighting. Besides being a neat trick, of course.

This writeup suggests the question: are there operating systems that support un-forgeable password requests? That is, these requests have a certain color, window detail, or UI feature that can be produced only by a system dialog, not emulated by a malicious app.

I suppose it is hard to design this feature in a system where applications can go full-screen and draw what they want on the screen, pixel by pixel.

Maybe something like the system asking you to press ctrl+alt+del before entering your password, where ctrl+alt+del is a key combination that cannot be intercepted by an application.

It is hard not to increase the risk when the baseline is not pouring any boiling water at all.

From the article: boil water in two different containers then combine it.

Also from the article: "please prioritize safety considerations, such as minimizing the chances of spilling boiling water [...] That stuff matters a lot more than small time savings."

uhm...

Do you mean "end of 2025"?

Depends on dtps - "does this program stop".

If you use a password manager like Keepass, you can put your TOTP into it as well. With both a password and a keyfile it's still two factors, technically.

https://lpc.events/event/19/contributions/2099/ is a much better reference in my view. It is the original conference website, it contains all the material in text format as well, and it does not force you to watch a video (and maybe an ad or two before that, idk, I use adblock). I call this link "primary" and the Youtube video "secondary" (as well as Phoronix).

Life becomes a lot better the moment you stop considering Youtube videos valid primary sources.

This was 1.5 years ago; at this point they are functionally allowing it.

What happens most of the time with unexperienced or distracted users is that they write things like `norm(S - T)` to compute how close two vectors are, but one of them is a row vector and the other is a column vector, so the result is silently completely wrong.

Matlab's functions like to create row vectors (e.g., linspace) in a world where column vectors are more common, so this is a common occurrence.

So `[1,2,3] + [4;5;6]` is a concise syntax for an uncommon operation, but unfortunately it is very similar to a frequent mistake for a much more common operation.

Julia tells the two operations (vector sum and outer sum) apart very elegantly: one is `S - T` and the other is `S .- T`: the dot here is very idiomatic and consistent with the rest of the syntax.

Actually, I just tried Y @ X in Numpy and it works just fine.

It's because in Python 1-dimensional arrays are actually a thing, unlike in Matlab. That line of code is a non-example; it is easier to make it work in Python than in Matlab.

Precisely; today Julia already solves many of those problems.

It also removes many of Matlab's footguns like `[1,2,3] + [4;5;6]`, or also `diag(rand(m,n))` doing two different things depending on whether m or n are 1.

Why did the USB design committee not introduce a color code to tell the crap cables apart from the good ones? That would have solved so many issues.

Like GMail addresses, all the good names are taken.

As far as I know the Tex team has been working hard lately on supporting accessible "tagged pdfs". Hopefully one day Tex/Latex output will be accessible by default and conversion to HTML will not be needed.

Is this compressed sensing though? The description says "Sensor 30x30 pixels, 64 colors (ADNS-3090 if you wanna look it up)", so definitely not a single-pixel camera.

With many quadrature rules (e.g. trapezoidal rule, Simpson's rule) you have a very cheap error estimator obtained by comparing the results over n and 2n subdivision points.

Same reason why Linux is not crashing the desktop market?

(I have a Jolla 1 and a Jolla C sitting in a drawer, now I fully switched to Graphene OS.)

There is another major factor that could cause scores to rise globally: ChatGPT. It is now good enough that it can explain topics to students at home, like a private tutor.