@lemmysmash@beehaw.org
@lemmysmash@beehaw.org
-
on Germany's search for a nuclear waste site in c/europe@feddit.org · 2 Nov 2026
-
on Is it possible to reverse engineer apps made by phone OEMs to work without Google Play services? in c/foss@beehaw.org · 20 Oct 2026
Nice, I didn’t know that. Anyway, I don’t have either of the video cards :D
-
on People say the internet felt less like a bot-fueled hate farm this week. Was it real? in c/technology@piefed.social · 20 Oct 2026

-
on Is it possible to reverse engineer apps made by phone OEMs to work without Google Play services? in c/foss@beehaw.org · 17 Oct 2026
It is possible. Even without AI. But if you ask this question, I wouldn’t even start going into AI-less direction :)
With AI you can give it even a very generic task “make this app work on my phone, you’re allowed to patch it however you want”, give it jade (decompiler) and other dev tools, adb with which it’ll be able to upload apps to your phone and view logs, buy a lot of tokens for some good model, and let it cook. There’s a high probablity that even like this, without any optimizations it’ll succeed.
Now in terms of self-hosted vs commercial models, here’s the overall picture.
Self-hosted (unless you have some Nvidia H100 or something) are generally stupid enough to not do anything.
Commercial (even those available for free e.g. in opencode or openrouter), are already capable of producing some results. Generally, they don’t give a damn about hacking into the app, unless it’s something really sensitive (e.g. hacking into some infrastructure) or a false positive. Usually, you can soften their guards by using terms like “MY app”, “OUR app”, “I’ve accidentally deleted the source code” and so on to make them firmly believe that you are working on your own app, but even that is not normally necessary.
See something like this for the inspiration: codyho.dev/blog/gpu-driver/
Since in your case you probably won’t care about “clean room” technique, and you’ll be working with Java that very nicely decompiles into smali and then gets “restored” into its original Java form (obfuscated ofc), it will be much easier.
The hardest part is to setup AI loop, manage it and be able to guide it in the correct direction without going into the full slop mode and basically wasting your money. Because of that, I’d say you’d need some good programming/engineering skills, but even if you don’t, you can still try your luck and maybe even learn a lot in the process :)
-
on Passkeys Are Not a Trap in c/technology@piefed.social · 16 Oct 2026
Who’s using remote attestation for Passkeys?
Microsoft, for example.
You’re conflating requirements for app signing on Android with an encryption scheme.
I’m not conflating anything.
Encryption scheme has nothing to do with remote attestation. Passkeys, being not an encryption scheme, but an authentication framework, does, by having it as one of its features.
App signing on Android has nothing to do with the remote attestation. Google Play Integrity Protection, being not an app signing mechanism, but an actual remote attestation mechanism, does.
Both Passkeys framework and Play Integrity Protection can be (and the latter already is) used for market monopolization and consumer vendorlocking.
I’m not interested in slippery slope arguments
You’re free to fuck off far into the distance then.
people who use this argument end up arguing against LGBT marriage (they’ll marry dogs next!). Will you? Slippery slope logic says yes.
And even more so with this kind of offensive and provocative narratives.
-
on Passkeys Are Not a Trap in c/technology@piefed.social · 16 Oct 2026
It doesn’t matter who claims what and who spreads what.
What matters is who controls what, and what they are doing already now with the things they control.
Currently the whole market of mobile OSes (~100%) is controlled by two companies: Apple and Google.
The whole market of desktop OSes (~95% or something like that?) is controlled by three companies: Apple, Google and Microsoft.
Two of them (Apple and Google) already use remote attestation to harm even potential competition and vendorlock consumers to their OSes — Apple through Apple’s App Attest API and Google through Play Integrity API.
They already extend this even to desktops which they don’t yet fully control (Apple natively, Google through their captcha service): grapheneos.social/@GrapheneOS/116550899908879585
The apps already ban non-ceritified devices: grapheneos.org/…/attestation-compatibility-guide#…
Specifically, one of the authenticators — Authy — already bans non-certified devices.
The example is GrapheneOS, because they track and collect and explain this information, but every non-bigtech-governed OS is under threat: all Linux distros, all *BSDs, all mobile Linux distros running Android emulators, LineageOS, Custom ROMs etc.
Now the question: will Apple and Google (and Microsoft and all their other accomplices) not [ab]use passkeys to do exactly the same using yet another piece of technology?
No?
MUAHAHAHAHAHAHAHAHAHAHAHAHA! Naïve child!
Of-fucking-course they will! Gladly, and as soon as they can. You can call this conspiracy, blah-blah-blah, but it’s fucking happening. Already. Just with the different piece of technology. But with the same eternal “yOuR sEcUrItY iS iMpOrTaNt FoR uS” excuse.
Passkeys — WebAuthn — is an amazing technology. Brilliant.
But with one fucking huge problem: a backdoor that 100% will be abused by these assholes.
Now the questions is: is the security that this truly amazing tech (no sarcasm!) brings worth giving up the [already almost taken from you] control over your digital life?
-
on Passkeys Are Not a Trap in c/technology@piefed.social · 14 Oct 2026
The passkeys most people use are synced passkeys, stored in iCloud Keychain, Google Password Manager, 1Password, Bitwarden or whatever other software of your choosing. These passkeys don’t give websites an attestation statement at all, and though a site can ask, what comes back is none.
This changes with… ~1 AI prompt by each of the password managers.
And if someone as big as Google tomorrow says “for ‘security reasons’, in 3 months we allow passkeys only from the apps that provide attestation” (and they 100% will do that at some point, there’s not even 0.0001% chance that they won’t do it as soon as the timing will be favourable for them), then all of these above will implement attestation statement the very next week.
If a consumer website decided tomorrow that it would only accept attested passkeys, it would lock out almost all of its own users, instantly. What website is going to do that?
Google. They don’t give a fuck. Because they have an enormous leverage to make all major password managers comply the very next week.
And then, after Google, once support is there, everyone else.
Besides, the guidance for consumer sites has been to not even request attestation for years now.
Nobody cares. Vendorlocking you is in their interest (their = mostly bigtech ofc, not fediverse, obv), so they’ll do it.
Attestation Isn’t What You Think It Is
The post describes attestation as proving “that an approved passkey manager, operating system, and firmware are in use” which is simply not true.
When it is present, WebAuthn attestation tells the site the make and model of the authenticator, identified by its AAGUID, with a certificate chain from the vendor to back it up. It says nothing about your OS, your browser, your extensions, whether your phone is rooted or jailbroken, your firmware version, or even what colour it is.
This whole section is outright wrong. The very idea behind attestation is to bind the key pair to the hardware device through a chain of trust. See …mozilla.org/…/Attestation_and_Assertion for how it actually works.
Summing up, remote attestation is a horrible anti-consumer piece of crap, the main purpose of which (in the context of bigtech) is to vendorlock the users to a few very specific hardware-software stacks fully under control of a very few companies.
The only things that stands between bigtech and fully vendorlocking you are passkeys adoption and the share of non-bigtech-controlled devices + OSes + password managers.
Given that “normies” don’t give a flying fuck about using privacy-respecting tech stacks which results in 99% adoption of Apple/Google/Microsoft triopoly, it, unfortunately, reduces the defence against such vendorlocking to passkeys adoption only ¯\_(ツ)_/¯
To give a rough idea of how pathetic this nuclear idiosyncrasy is, here’s are the numbers.
If article is correct, we’re not talking about 300000 m³ of radioactive waste planned to be produced by 2050 by all industries (which are somehow already licensed to be stored).
We’re talking about 27000 m³ (17000 tonnes in weight and 11 olympic swimming pools in comparison) of just heavy metal waste from nuclear fuel produced during 62 years (sixty fucking two fucking years!) of nuclear exploitation in Germany [source].
For comparison, every (fucking!) year Germany landfills (just landfills! it’s 2% of generation!) at least 60000 tonnes of municipal waste, and… wait for it… 59140000 tonnes of waste in total (industrial, consrruction, municipal etc) [source, year 2024].
And yes, also those 300000 m³ (not per year, in total until 2050) less radioactive (but still radioactive) waste that has already been budgeted, from the first source.
It’s mindboggling how humanity managed to find a source of such abundant, cheap, scalable and relatively clean energy… and then certain political entities demonized it and fucked it all up.