Former account: @Redjard@lemmy.dbzer0.com
Keyoxide: aspe:keyoxide.org:KI5WYVI3WGWSIGMOKOOOGF4JAE (think PGP key)
uptime should be handled by the kernel, so a kexec "soft-reboot" would still reset the uptime.
Flu and corona are both "common cold type" viruses defeating resistance in some way. For coronaviruses that method is stopping the body from building effective resistance by all means possible, so that is why vaccines tend to not work too well.For the flu it's the many variations and its tendency to change further and need new antibodies.So I don't think a specific flu strain is hard to make a very effective vaccine for, but ofc this doesn't yet solve the flu problem.The immense speed at which mRNA vaccines can be developed might improve that in the future, where this here could be one of many steps to get regulatory approval for blanket mrna and actually be permitted to change them at that pace.
In principle mRNA should let you crank you vaccines for new diseases/flu-strains in under a week. If this can fully stop the flu?... I doubt it. Whatever does solve it will probably make use of this tech though.
Headline sounds like they want them to relocate to Europe. Bummer.
Usually for typical conductors I think we can learn enough about the position and momentum of our electrons to keep track of them.Not overly precisely but enough to not confuse them with each other.A good way to see this is that we can simulate electrons moving through states in a conductor with good accuracy, without a need to go into full quantum mechanical descriptions, in an almost classical simulation.Of course the position and speed we are tracking there is not a typical blurred point, it will be a complicated wave spanning many neighbouring atoms in size, with different electrons being at different positions around those atoms.But you can know which electrons are in what loose region with what distribution, and follow them through interactions where they move to different regions or change the shape of their wave. Depending on your conductor the spread may even go down to single-atom-scales in some extremes.
Measuring all electrons in a real conductor enough to tell after some time which end of it any one electron ended up at, would probably change its properties slightly due to the measurements, but done correctly it should definitely still be behaving like a normal conductor.
Here is an example of a particularly low-speed localized electron in a typical material:
Every electron will have a patterned distribution like that. If you naively tried to measure electrons at some spot, there would be thousands strongly overlapping there, you would mess everything up. But there is no issue checking if this wide shape as shown has an electron occupying it or not.
They are usually very particle like.
The extreme is semiconductors and band fuckery. There you have a lot of stuff that only makes sense with electron particles jumping around accelerating and colliding. You can talk about mean free path in conductors etc.
They do however interact a lot via their charge, so electricity propagating is usually a pressure wave mostly visible in the electric field, with electrons only doing very little, moving slowly. This field is however driven by the electrons, so it really is electrons doing that.
Resistance, diodes, even heat conduction (/"resistance" to heat transfer) directly follows from electrons as particles interacting with atoms or themselves.
I think you loose these particle effects in (type 1) superconductivity. Your cooper pairs are so smooth and non-interacting they might aswell be clouds if you don't zoom in enough (quantum circuits sometimes have singular amounts of those charge carriers running in circles in a tiny superconductor loop).However, when you look into why superconductivity is a thing and at what temperature you see it etc., that once again derives directly from the properties of electrons and the layout of your material.
Or maybe you meam do electrons have dimensions and physical properties? That they don't.They have a charge, speed, momentum, ... but interaction is via the charge. That is what touching means, there is no further collision (technically there is electroweak force stuff).They are not points either because qm eave blurriness stuff.
For your specific electron movement questions. I think electrons in metals under normal current where the metal heats only a litte move at snails pace, literally. In semiconductors it's a lot faster, about bullet speeds if you push some current. Metals could probably also get to bullet speeds but they would vaporize and explode due to the electric and magnetic fields. Faster than that isn't really feasible. Btw. the random motion at room temperature is also at that scale, though due to quantum effects it's not a bouncing around.
So when you have circuits operate near light speed, that is an electron pressure wave not electrons. The electrons react via quantum fuckery, but changing between speeds and positions at the slow pace described. The reaction is carried on via the electric field. The speed of electricity is the speed of interaction between electrons, not the speed of electrons.
Without the electrons reacting and causing their own electric field changes, the electric field still carries the effects but it quickly dilutes. This is called radio (depending on the timescale of your change), and while it does transmit electricity it does so very weakly, nothing compared to the "continuously electron reinforced short range interaction" wire.
Might be an (almost) uniquely american meme for the circumcized?
The reason is that before all this only 20% of oil passed the strait and 8% is being redirected through pipelines built long ago anticipating this.12% of supply decrease will double prices, because turns out you don't have much choice when you design everything to depend on oil, but eventually everyone figures out to use 12% less.
Try to log the stdout of your services, I dare you.
openrc is just missing some pretty essential things. I'm not saying to copy journalctl, but at least dump stdout into some tmpfs file by default.
To have some sane basic logging on hand if a service breaks weirdly or is misbehavingy you'd need to edit that specific service file and restart. And most of the time look up the spec of the specific service command to remove log supression.
Unlogability alone makes openrc quite a nightmare for a lot of setups. I've wasted hours repeadedly that would have been 5min had I gotten the log upfront.
If you say so*steals @username_1@programming.dev 's computer*
Keine mehr? Gar keine Fundis mehr? Zwei noch? Boah, die werden jetzt ultramäßig gefickt, jaSieht man auch nicht allen Tagen
There was a confusing name change, and it doesn't help that ecdsa/ed25519 has two names, but the number 25519 is specific to this fixed version. Funnily if you quote search nsa and ec25519, this thread is the only result besides one ycom thread (which also is in context of them being safe).
ec25519 is not a typical name for it used in any software afaik, only in writing.
Edit: Historically ecdsa used to refer to the backdoored one. Since it has fallen so much out of use, ecdsa now means ed25519 since it's usually imcorrecly called ecdsa and also changed to ed25519. It is of course better to specify 25519.
Speaking of which
and intentionally put vulnerabilities into Ec25519
25519 is the fixed one. It is also not backdoored. Please fix that aswell. It is only Dual_EC_DRBG that is affected, not RSA nor ECDSA/ED25519
NSA has long since broken RSA
This is clearly referring to the algorithm. You don't "break" a company.
There is also little reason to bring up the RSA company at all, it is for all intents completely irrelevant.
Please just edit your root message to talk about the EC (Dual_EC_DRBG) that is not really in use anywhere but at least real and something security people know of.
If you say the nsa has broken rsa, you are making a lot of sysadmins sweat for no reason.
You linked the article I was talking about.
There are two, different, unrelated things:
RSA, Rivest–Shamir–Adleman, an asymmetric encryption, that comes in sizes like rsa2048 and rsa4096. It is now, having largely been replaced by ecdsa, which is using elliptic curves, a different kind of mathematics. The main benefit of EC is smaller key sizes.If you have old ssh keys, they are likely id_rsa. New ones are likely id_ecdsa.
The NSA tried to backdoor elliptic curves, long after rsa the encryption was already around (rsa encryption dates back to the 70s). This presumably nsa-backdoored EC implementation is quite famous, and what your article is talking about on the technical side. This EC has been largely abandoned. An ssh key named id_ecdsa or id_ed25519 will be using a known secure EC using different safe seed values.
Now, RSA encryption and EC encryption are two separate categories, an asymmetric encryption algorithm is either RSA or EC (or something else), but never both.
Enter stage left the company "RSA", RSA Security LLC.This is a company originally founded to market rsa encryption, hence the name. It has long been owned by another company within which it now deals with many different encryption algorithms and related tech.It does not own the rsa algorithm, and it of course has no influence over it. The algorithm is set in stone and has been for decades. If you try to change it you are making something new with a different name.
This company was naturally dealing with the hot new encryption tech of 2014, called EC cryptography. Which, as you may recall, is mutually exclusive to being the rsa algorithm.
RSA Security LLC was apparenlty influenced by the nsa to adopt their broken EC cryptography. This of course makes the company, their products, etc., all suspect.
Now stay with me here. The company RSA Security LLC, which is suspect, is not related to the algorithm called RSA. If the company is suspect, this does not call the RSA algorithm into question, which has been subject of cryptographic analysis for decades and predates RSA Security LLC by a number of years.
The suspect thing is a special EC crptographic implementation, which excludes the rsa algorithm being involved.
Now let's read the article:
[...] Dual_EC_DRBG, was ratified by the National Institute of Standards and Technology (NIST) in 2007 and is attracting a lot of attention for having a potential backdoor. This is the algorithm into which the NSA allegedly inserted a backdoor and then paid RSA to use.
An EC algorithm. Meaning not RSA."paid RSA". Since this is definitely not RSA encryption, it must be RSA Security LLC."paid RSA". You cannot pay an algorithm, only a company. Thus, this is RSA Security LLC.
My dude, rsa is fine. This article is talking about a company called rsa, not rsa encryption.I have never heard of doubt about rsa's security, given enough size. The main issue with raa is that it needs to be thousands of bits in size due to not being very efficient. And of course it is not post quantum.
GTT (gpu swap) is handled by the gpu driver, so only nvidia can see if they can add it to their closed source drivers. radv is the amd vulkan driver.
This means continued security patches until at least end of 2027, so at the earliest these will be out of support in 2028.
Realistically this will be much longer into the future, as the LTS window of multiple LTS releases is likely to be extended more.I'd be surprised if these go out of support before 2032
Molly supports unified push
Notification logging is usually done by some other part of android as far as I know. GMS is the typical way to deliver notifications and is a far more serious privacy concern, since it also directly passes googles servers and is not encrypted. However as others mentioned, signal does not send contents there, message notifications with the message contents stay on device.
Don’t You Know Who I Am? @lemmy.world Spend some more time on Reddit like the grown ups
Männer haben keine Gefühle und fragen nicht um Hilfe.Pass auf dass deine Kleidung keine Farben zwischen Rot und Lila hat, sonst denkt noch jemand du wärst ein Mädchen.