Hacker Newsnew | past | comments | ask | show | jobs | submit | more AstralStorm's commentslogin

Still waiting for someone to make a tiny token sized phone. Unfortunately the smallest around, Unihertz Atom, is both outdated and too low resolution for some apps to work.


Now that is more of a problem than a bank. Which is why someone beeds to integrate OTP tokens into ID cards, closing the issue.


Uh, banks still provide separate tokens and one time pad cards last I've heard.

If yours doesn't, pick one that does.


The larger point here isn't whether they do, but that they'd rather not. They want you to rely on their app, and have been pushing people to it for years now (some more intensely than others).


Warrant canary depends on agreed upon inaction, which shields it somewhat. You cannot exactly compel speech by a gag order.

This, being an active process, if found out, is violating a gag order by direct action.


Warrant canaries depend on action, the removal or altering of the canary document. It’s too clever but no more clever than what Israel is requiring here.


the canary notification method is a lack of updates, not a specific update.

you update your canary to say that nothing has changed, at a known cadence.

if you ever dont make the update, readers know that the canary has expired, and so you have been served a gag order warrant.

changing or removing the canary in response to a warrant is illegal. not changing it is legal.

for an equivalent cloudwatch setup, its checking the flag for "alarm when there's no points"


Yes, the equivalent of a warning canary would be that Google pays the Israeli government a set of payment every month such as 3100 shekels (for +31, NL) and then suddenly November 2025 they stop issuing it. That would mean there's a legal investigation targeting Google by the Dutch prosecutor (OM) involving Israeli data.

I suspect they didn't go for this route as it is too slow.


I would think to stopping doing something is equally an action as to do something, in regards to warrant canaries and gag orders. You had to take make some change to your process, or if automated take an actual action to disable. In either case, there was a cognizant choice that was made


The legal theory is that in the US the first amendment prevents the government from forcing you to make a false update. I don’t know if it’s ever been tested.

As I understand, this theory wouldn’t even hold up in other countries where you could be compelled to make such a false update.


What if I, sometimes, annually paint a canvas with an artistic interpretation of a canary bird? Can a government compel me to make an artistic expression with specific content, at my own expense? What if I'm just not in the right kind of creative mood to make it a good painting?

Or maybe I can bill the government for the compelled artwork -- I'm afraid I'm tremendously expensive as an artist.


Yes but the theory, at least in the US, is that the government cannot compel you to say something. That is, they can't make you put up a notice.


More specifically, the theory is that cannot compel you to lie, there are all kinds of cases where businesses are compelled to share specific messages.


As far as I've seen, the examples of that have always been things like health warnings and ingredients lists, where showing that message is a condition of being in that (licensed) business, and applies equally to any company.

Do you have a more custom example in mind?


All sorts of consent decrees, a huge amount of union and workplace law requires things to be posted for employees.


For employee things, I can understand being required to notify parties in agreements the company has entered into. As far as I understand, consent degrees are settlements and as such a mutually-agreed mechanism for ending a lawsuit early; their terms are whatever the parties negotiate and do not come from the government.


To be more precise, the law requires employees to publish the nlrb notice in well trafficked or otherwise conspicuous locations.

I think there are other places where "government mandated corporation inform people of their rights" is a thing, especially with things like data use and sharing.

In terms of consent decrees, that was the wrong example. But lots of judgements do involve various notification requirements.


Car manufacturer warranty recall letters are probably a good example. I get them even though I've never done business with the car manufacturer -- I bought the vehicle from a private party.

But that still sort of connects (at least in my mind) to health warnings etc.

What do you think of this angle: https://news.ycombinator.com/item?id=45892680


Ah, that was confusing to me. Thank you.


yea, I get that, but my gut tells me this doesn't pass the sniff test

It's a choice you make and action you take either way, be it not updating a canary or sending a covert financial transaction

That it has not been tested in court is why it's still a "theory" (hypothesis?)

My hope is that a jury of our peers would stay closer to the spirit than the letter of the law


Inaction is not action.


The choice to cease perform an act, when you have been consistently doing it, is itself an action


No, making a choice to do nothing is not considered action by any legal definition.


And this would be why warrant canaries aren't seen as a proven legal shield yet.


>Warrant canaries depend on action, the removal or altering of the canary document.

No, they can simply not publish a warrant canary in the future, which will tip people off if they've been publishing it regularly in the past.


Right - the whole premise is that the government cannot compel speech (in the US). So if you publish something every week that says, “we’ve never been subpoenaed as of this week” and then receive a subpoena, the government can’t force you to lie and publish the same note afterwards. The lack of it being published is the canary here.


Whether you can be compelled to lie under these circumstances or not is not a resolved question of law. Although it seems fairly likely that compelling speech in this way is unconstitutional, if it has been tested in court, the proceedings are not public.


Good thing no one is doing anything unconstitutional right now?


TPM itself is a simple data container with slow encryption/decryption capabilities. It cannot hide anything really.

You might have mistaken it for say Intel ME and the AMD equivalent.


No, these should exist in the TPM and highly volatile memory like CPU cache. This including the decryption code. This can be achieved using mechanisms similar to what Coreboot does before RAM is initialized.

No need for the keys or decryption to touch easily intercepted and rowhammered RAM.


Yes, I think we’re saying the same thing. A TPM is a Secure Enclave.


Don't worry, you're going to be searching for logic vs requirements mismatches instead if the thing provides proofs.

That means, you have to understand if it is even proving the properties you require for the software to work.

It's very easy to write a proof akin to a test that does not test anything useful...


No, that misunderstands what a proof is. It is very easy to write a SPEC that does not specify anything useful. A proof does exactly what it is supposed to do.


No, a proof proves what it proves. It does not prove what the designer of the proof intended it to prove unless the intention and the proof align. Proving that is outside of the realm of software.


Yes, indeed, a proof proves what it proves.

You confuse spec and proof.


You might have just filtered off all the nutrients and have yourself a dietary deficiency. Oops.

And your supplements might well be contaminated...


I suppose you wouldn't be eating these oats regularly? More like a couple times, then test the levels and maybe repeat after a while?


Thing is, because the whole design is closed as well as firmware, the security of it is near zero, even for sealing firmware device images (e.g. option ROM), much less bootloaders. Multiple security holes have been found.

There's no issue booting a boot rootkit with the standard Windows bootloader unless you manually seal the image with command line or group policy, and even then it's possible to bypass by installing a fresh bootloader because the images are identical and will boot after a wipe.


>Thing is, because the whole design is closed as well as firmware, the security of it is near zero, even for sealing firmware device images (e.g. option ROM), much less bootloaders. Multiple security holes have been found.

This. It is secure only for MS, AMD or Intel.


Yeah right. Because building a freezer that goes to -30 C is as cheap as going to -18 C. It's much beefier hardware with a lot more insulation.

Likewise a heat pump can only boost so much.

This, like other environment related changes never happen by market forces. Not once. And small tweaks even on large scale produce small effects, insufficient for our needs.


Most normal home freezers have a way of setting temperature, e.g. mine can go from -16 to -24

So maybe -30 is difficult but it wouldn't be that hard to have the existing temperature range on new models be dynamic based on electricity pricing


> Because building a freezer that goes to -30 C is as cheap as going to -18 C.

For small sizes, yes it is.

But also, capex vs opex. Even if it's twice the cost, you only pay it once.


Guidelines | FAQ | Lists | API | Security | Legal | Apply to YC | Contact

Search: