taminka

yeah messenger still has the pin code thingy, i'm curious why they do it at all that way, can't you just have your keys on fb servers encrypted with another set of keys derived from your password, which is much stronger than a 4-6 digit key?

oh, i see, is it the same for facebook messenger and instagram, imessage, etc?

you can (https://faq.whatsapp.com/1046791737425017/?cms_platform=andr...)

they even have it on fb messenger and instagram (though they recently removed e2ee completely from instagram lol)

they do have encryption, just not e2ee, and in fairness to them, it doesn't make sense to have e2ee on a channel or a group with 100k ppl in it, also device sync is possible with e2ee, it's just a slower

whatsapp, facebook messenger, imessage all support multi-device and it's pretty convenient, in fairness to telegram they launched a bit before double ratched was invented, but still, they've had over a decade to switch to it...

it's really unfortunate that telegram doesn't do e2ee, bc it's hands down the best messenger otherwise :(

it's unironically just react lmao, virtually every popular react app has an insane number of accidental rerenders triggered by virtually everything, causing it to lag a lot

you do/can control all the VPN nodes in this setup (most often just a single one) since your traffic doesn't actually go through the website you're masking under

and the nature of the protocol makes it extremely difficult to detect and thus get server IP banned, i got one server banned, but after that i implemented some practices (including directly connecting to websites that are inside Russia) and it's been working fine since then

you need secure channels of communication (and preferably a connection to the outside world) to solve any problem

vpn protocols we use here nowadays are way more advanced than this, they mimic a TLS handshake with a legitimate (non blocked site, like google.com) and looks essentially like regular https traffic to that site

it looks like they are basically impossible to detect, given the failure to block them, outside of timing attacks (seeing if a request crosses Russia's border and comes back quickly after), however that is fully mitigated by just having having the vpn "disconnect" and route traffic directly to Russian unblocked sites, which would otherwise be able to perform such a timing attack detection

pretty interesting stuff, there are several versions of this system, and even the ones that have existed for a while work pretty well

perhaps, there's still hope i think:

- roskomnadzor just not being competent enough to implement the block fully

- they'll reserse the block, since it will likely completely cripple everything that relies on the internet (which is basically everything nowadays)

- they won't go through with the ban completely, since if they do, their job is sort of done, and they want to continue to exist to make money off of the digital infrastructure required to implement the block, and they'll just continue playing this game of cat and mouse

- outside internet connectivity will likely remain to some degree, it'll just be very slow and probably expensive, but i really struggle to see a country like Russia being completely cut off from the internet in the year of our lord 2026

i could be wrong, who knows, after all this whole situation is unprecedented, and human ingenuity sort of always finds a way

and in a somewhat positive note, mobile internet has come back today and the blocks are bypassable with a regular vpn now, even ones that aren't being hosted on whitelisted subnets

read the post please, the precise problem is that this may soon not work

nobody actually likes it, it's just macos is still the least terrible to use option

what do you mean by that?

i swear if someone starts another single header vs other options debate in this comment section i'm gonna explode

most of the traffic is probably from open weights, just seed those, host private ones as is

actually you may be right, according to project zero by google [1], ~50% is use after free and only ~20% for out of bounds errors, however, this is for errors that resulted in major exploits, i'm not sure what the overall data is

[1] https://projectzero.google/2022/04/the-more-you-know-more-yo...

logic errors aren't memory errors, unless you have some complex piece of logic for deallocating resources, which, yeah, is always tricky and should just generally be avoided

this is silly, we already have an algorithm for generating very efficient assembly/machine code from source code, this is like saying maybe one day llms will be able to replace sin() or an os kernel (vaguely remember someone prominent claiming this absurdity), like yes, maybe it could, but it will be super slow and inefficient, we already know a very (most?) efficient algorithm, what are we doing?

this is amazing, counter to what most ppl think, majority of memory bugs are from out of bounds access, not stuff like forgetting to free a pointer or some such

presumably you'd just use unique one time codes derived from the eID

really cool website, what's your colour palette?

wasm isn't meant to supersede html/css/js (unfortunately) and it's regularly used for high performance applications in the browser, web-based cad software, figma, youtube (i think they use wasm for codec fallback when support is spotty) etc

there is also games, stuff to do with video (ffmpeg built for wasm), ml applications (mlc), in fact it's currently impossible to use wasm w/o js to load the wasm binary

as a result, the web stack is a bit upside down now, w/o the seemingly "low level" and "high performance" parts over the slow bits (javascript)

i wonder if ppl's electricity consumption habits will change in response to this, idk like turning the heat way up during the day or using high power appliances more during the day

so they're ordered, which i didn't dispute, but alignment is implementation defined, so it could be aligned to the biggest field (like in the article), or packed in whatever (sequential) order the particular platform demands, which was my initial point

i was talking abt padding/alignment, not ordering, that's indeed not allowed you're right

struct field alignment/padding isn't part of the C spec iirc (at least not in the way mentioned in the article), but it's almost always done that way, which is important for having a stable abi

also, if performance is critical to you, profile stuff and compare outputted assembly, more often than not you'll find that llvm just outputs the same thing in both cases

first time i used tahoe to help a friend w/ their laptop i legit thought it was like a knockoff macos or something, genuinely the ugliest macos version and even in the brief time that i've used it, i've encountered annoying bugs, QC at apple is dead lowkey

that's just called having a complete project in a stable language lol, not everything needs a change every week to function well...

unfortunately just an inherent consequence of treating software as needing continuous improvements and having yearly release targets, you can't just say that this settings menu is already perfect as is, you have to change it, therefore everyone perpetually shuffles around ui and adds features that nobody wants