Guess I live somewhere in Asia according to my VPN from now on. Fuck this shit.
post
cause nobody ever bypassed play integrity
Ban age verification. It's an accessibility and privacy nightmare, besides the free speech implications.
So, let me summarize this, maybe i miss something:
a) This is a Module/Reference Implementation not the finished app b) It is in the earliest stages of development c) it is opensource, someone could provide alternate means for device/app verification via MR d) There is no mention about "THIS IS THE ONLY WAY WE WILL DO IT", that part is a straight out lie!
Nothing prevents nations to implement other verification methods the googles API.
So, what are we raging about here? People insulting developers for doing their work? People using an issue tracker like twitter? People, that do not read properly?
Yes, the italian implementation already is tied to google, and there should be some push back, but that is NOT the fault of this project.
Again, this is not an official app. It is a reference implementation.
You can not, and will never be able to, download this app from the App store.
Why must they have installed that? For what if there is no app?
Like what? Dude! There is no app you're can install, so there's no requirement for Google APIs.
I think you have no idea what you are talking about, what the issue is, and why the scope of the project matters.
Let me try to summarize this:
EU policies require the ability to verify COUNTRY SPECIFIC implementations to be verifiable. The reference implementation linked above uses a Google API for it.
Countries, that implement the actual apps are free to offer other verification methods then the Google API.
Italy has an app currently, and this app requires this stupid Google API, yes, but that is not the above project.
The websites will never communicate with the play services! They communicate with the APIs of the authorities issuing the state ID's. The App must be verifiable to protect the user! We have a state ID app in Germany. It is open source and I compiled it for my Linux desktop and graphene OS based phones. Guess what, yeah, it works.
What many are missing here: the purpose of the verification and the actual authentication workflow.
Also, let me just cite the readme of the project:
This is an initial version of the software, developed solely for the purpose of demonstrating the business flow of the solution. It is not intended for production use, and does not yet include the full set of functional, security, or integration features required for a live deployment.
But we are all going crazy because some minimal example code relays on Google APIs ....
Please stop pushing such utterly wrong and confusing bullshit.
I know this issue, yes? And now?
Wtf are you now talking about? The politicians already defined the digital rights act.
And politicians don't write software.
You are absolutely right about end-to-end encryption and VPN. Essentially everything that has to do work law enforcement.
End we should focus on those topics. They are political issues and have lasting consequences. That a reference implementation will require Google APIs in the future is not an issue worth rageing about beyond "That is a bad example and violates EU law".
There is no app. Read the readme from the beginning. Even cited it here. Show me the he app in the app stores! YOU CAN'T!
THIS IS A REFERENCE IMPLEMENTATION NOT AN PRODUCTION READY APP! THIS IS EXPLICITLY STATE ON THE PROJECT README!
Read the ducking docs
The Age Verification (AV) android app is part of the Age Verification Solution Toolbox and serves as a component that can be used by memberstates, if necessary, to develop a national solution and build upon the building blocks of the toolbox. [...] This is an initial version of the software, developed solely for the purpose of demonstrating the business flow of the solution. It is not intended for production use, and does not yet include the full set of functional, security, or integration features required for a live deployment.
The EU governing bodies are speaking out of both sides of their mouths if they claim that they want data sovereignty while simultaneously relying on an evil, American company to verify your "integrity" 🤡
You'll never be sovereign if you rely on a for-profit entity that makes money by spying on people and selling your data.
In digital age it should be understood as a personal liberty to not be compelled by state to use nonfree software in any shape or form. Just like court rulings must be public and legislation too (sadly this doesn't apply in EU).
What in the fuckety anti-anti-trust megacorp win is this???
Where is the petition to sign??
The main problem isn't the Google Play integration, but that this requires an Android or iOS device at all. This should be based on something like flutter or electron, and be easily portable with an agnostic build script for e.g. Linux, UBports, postmarketOS, and so on, as well. If only for the reason that most Android and iOS devices will effectively become unpatchable after the mandatory 5-ish years run out, while a standardized UEFI desktop platform will not. There are so many reasons not to have a "standard" smartphone nowadays. Also see here.
Is there anyone more familiar with this age verification process that can explain if and what data does this share with some UE body or government? Is the the system 100% client-side or is there any API or tie to other govt service that may be able to track when and where (website) you're trying to verify your age? Thanks.
Apparently they want everybody to get some sort of "EU wallet", that is, some digital signed identity which sounds super dystopian. But that's just what I read. It sounds like a complete disaster.
I feel like a productive way to address this would be to make a child mode mandatory for all operating systems, as some EU countries already did, and then to give parents a better incentive to actually enable it. For example, all end-user devices could be pressured into prominently showing an option to enable it when first booted up (without forcing your hand either way) so that it's hard to miss. There are so many other ways to improve this situation.
Here is the source that they want to make e.g. Youtube, Netflix, ... rely on this new app: https://www.mlex.com/mlex/articles/2368265/online-services-get-up-to-12-months-to-apply-age-verification-eu-guidelines-say
I agree with most concerns here but as a professional prototypist... people do not seem to understand here and on related issues what "reference implementation" means.
This is NOT supposed to be used! By anybody! This is basically a technical demonstration that shows how it can be done at all.
Think of this as a test suite rather than software proper.
Again, this does not mean it's OK to even suggest that Google and Apple are in any way acceptable bottleneck. I do believe those are terrible choices. I do also believe relying on them just to do a proof of concept or technical demonstration is quite "lazy" but I also bet that this was necessary due to the scope of the project, e.g. "deliver us an app that works in 6 months on an average mobile phone". I really don't think they had discussion on accessibility, inclusion, etc.
So... yes, do keep track and be concerned but also don't conflate a proof of concept with a maintained app that will be required to be used on all EU citizen mobile phones next year.
There's a big difference between a reference implementation and a proof of concept. A proof of concept just shows it's possible at all, but a reference implementation is meant as a reference for "you should do it this way". Expect most companies to just directly copy the reference because they'll feel it's a waste of time developing their own system that's in compliance.
This is definitely going to be copy&pasted as a foundation in many EU states. Therefore, that it requires Android and iOS at all, let alone Google Play, is a fundamental error. Some people avoid smartphones for good reasons, yet still access parts of the internet that may apparently soon be gatekept by this new age verification mechanism. Also see here.
Sure it's beyond a proof of concept and others will definitely heavily rely on it yet my point still stand, i.e no one is supposed to install this built app directly. As I also said yes it's wrong to rely on Google and Apple in general but even more so with talks of EU sovereignty so I'm not giving them any slack for that. What I'm still insisting on is that this repository is not the app people will have to use.
This verification efforts were kicked off earlier this month; this app hasn't really launched yet, has it? I think proper implementation after a test phase will maybe come next year. I think it is too early to complain that aftermarket OS's are being excluded. It seems to me that nobody has tackled that problem yet rather than this being a willful exclusion. And while the EU lawmakers thought it was okay to put the Googles of the world in a position where they get to be judge, jury, and executioner for the right to be forgotten, I have a feeling that GDPR and the general vibe within the EU will not allow this to only work with the help of one American corporation on the continent's most used OS. We need to be watchful but not despairing just yet.
Graphene OS can give apps access to the Google Play Integrity API.
It'd better to check whether the bootloader is unlocked. If banks can do it, then this app can also do it.
There is a disclaimer though, which kind of brings hope:
This is an initial version of the software, developed solely for the purpose of demonstrating the business flow of the solution. It is not intended for production use, and does not yet include the full set of functional, security, or integration features required for a live deployment.
This makes me to thinkgof this solition as of a risk for now (high one, though, with modern approach of shipping PoCs)
Yes, but it also says that the integrity check through google will be added in the future. Reason why it's not yet operative.
The way I understand it is that the app is not yet ready because integrity checks through google are not ready, not the other way around.
It also says that integrity check through google is a "key feature".
Honestly I'm starting to understand hermits. I have to accept that I've become a digital hermit.
Whats better than enshittification? Government powered enshittification!
If the EU is going to make software then I think they'd have to consider themselves a gatekeeper requiring people to make a third party authentication app using their API.
I'm honestly surprised by the negative sentiment here. Government certified digital identity is the obvious solution to spam & online disinformation. In the short term bot farms will lose the ability to influence social media. In the long term it'll allow us to distinguish real people from AI agents, true scientists from pseudo-scientists, actual journalists from state-sponsored fake-news spreaders. It's a key technology that fediverse needs to survive in a world flooded by AI content. Finally it's something that can save democracy from literally falling apart. Remember that democracy only works if citizens are well informed. Otherwise those who control the information are the true rulers.
It's also complete bullshit that third party ROMs can't get proper attestations. It's a statement literally made up by the OP. Third part ROMs can get the keys. They just need to go through the attestation process (which BTW can be fully automated) and publish signed images for users.
Don't like the existing integrity providers (Google, Apple)? Make your own. The system is designed to be flexible and not rely on single integrity providers. Or tell us how else you'd like to solve these serious and growing problems.
Sorry for the rant-y comment. I'm tired of people who only complain and nitpick.
all 36 comments