• 0 Posts
  • 84 Comments
Joined 3 years ago
cake
Cake day: July 2nd, 2023

help-circle






  • Tetsuo@jlai.lutoMildly Infuriating@lemmy.worldWelcome to New York City
    link
    fedilink
    English
    arrow-up
    23
    arrow-down
    1
    ·
    3 months ago

    With so many cops why would it rely on the honour system ?

    I mean it should be the cops kicking out people off bike lanes… We have assholes like that in the EU but they sometimes get a fine for it and eventually some stop doing that kind of shit.

    The problem is that it’s not enforced by your police forces.



  • You fell for the trap of thinking that you can only protect childrens by accepting to sacrifice privacy.

    My post was clearly pushing for doing both :

    Proving anonymously that you are an adult without giving your identity.

    Again I have to repeat I’m not against preventing childrens from accessing social networks.

    Where we disagree is that you think you have to give up on some privacy to achieve that goal. I don’t.

    Now let’s agree that we disagree and I will let you give your faceid to shady businesses and see them leak two weeks later.






  • I was expecting skepticism about that in my comment and choose purposefully the word “referendum” for that reason.

    I dont think electronic voting is ready for important electoral votes but it may be sufficiently secure for referendums that are fully digital. In France, you can sign petition to force certain questions to be asked to the government. A cryptographic mean to sign that petition would be very useful and in this case prevent people from attempting to sign multiple times.

    We already have a hardcopy ID to travel internationally or ID to drive vehicles. The risks of an electronic ID being easily copied would have to get me visiting other solar systems or driving a time machine before I considered it.

    I think the opposite. Standard papers are and always have been fraudulently copied by criminals. Cryptography can make it impossible to copy an ID card the same way it makes it impossible to clone your credit card.

    The main risk is for governments to try to emit fake ID cards for imaginary citizens, but they were already doing it mostly for their “spies”.

    I honestly think it will completely stop criminals from copying papers or creating fake ones. At least as long as the government takes security of the ID card emitting body is taken seriously and audited transparently.

    Really a lot of your criticism of digital means applies very much to older paper based systems. I personally participated in counting votes in my local elections and despite many counts by multiple people, we didn’t get the same count everytime. So while mass fraud is very unlikely, the count of all elections is not really accurate.

    But again my original point was for this digital ID card to be used only for referendums and not electoral voting. Anonymity is not necessary when citizens petition for a cause to be heard.



  • Here is how I wish we would solve this problem :

    Every adult citizen gets an electronic ID card. Said smartcard let’s you authenticate on government websites (FIDO2). And maybe it could sign your vote for referendums if that’s possible in your country. Every smartcard would be secured with a PIN code.

    It also can create an anonymous “adult cryptographic token” to register on adult sites, gambling sites and social networks.

    Parents could create a token for their childs but they take responsibility for any illegal use, for example in this case using social networks below 16 years old.

    Edit : This seems to already be a project in the EU with the EUDI Wallet. It is supposed to start deployment before the end of 2026. Too bad the UK left the EU, this EUDI wallet has features to prove anonymously that you are an adult…


    Now what really will happen is politics will say “think of the children” and use that opportunity for more surveillance of their population.

    It’s too bad technical problems are answered by politicians and not engineers. We are also very late in term of digital ID cards. Everybody should have for free the means to authenticate online and do so anonymously when needed. We live in a digital world, we need the tools to evolve in it.




  • Either Linux’s built-in display drivers are black magic or microslop is incompetent

    Why not both :) ?

    I just realized that since switching to cachyos 4 months ago I have never had to install any driver or driver updates (outside of just running the system update).

    Even during initial setup I don’t think I installed anything driver related.

    It’s really a step up from Microslop. Last time I installed my W10 I had to prepare all the necessary drivers and collect them on each manufacturers websites.

    It is so much more streamlined on Linux.


  • They could also put a checking tool into CachyOS Hello, which is shipped and pops up by default.

    What would this “checking tool” look like? What would it check?

    I personally have deactivated the opening cachyos Hello a long time ago. Why would I need that popup once I setup everything?

    And I’ve definitely gotten “urgent” text notifications that all-but-required manual action through pacman.

    Pacman has no idea if it is installing something malicious. It notifies you only on functional actions that are required.

    Basically, none of the suggestions you make would have avoided the AUR attack to work. Nor a future one?

    The only thing I would maybe agree is for some notification system that let’s the cachyos maintainers send an urgent message but that would mean they would have to sign that message in some way. If that signature verification ever fails someone could send malicious notifications to all cachyos users and that would create another threat.

    And even then if the malicious package is noticed after a few days, if you already installed/updated it, it’s too late. You could receive a notification giving guidelines to cleanup but that’s too late. The infection could disable these notifications or worse.

    And if you have an emergency notifications systems, is it a “pull” or “push” notification? Is it your computer that checks if there is a notification? How long between pulls? If that’s a push then the notification servers basically has a full list of cachyos IPs which would suck too.

    Sorry if I look nitpicky but I just want to illustrate that this is a very very complex problem to solve while respecting user privacy and “sovereignty” over their system. Supply chain attacks are extremely difficult to defend against and open source projects have increasingly numerous dependencies…