Skip Navigation

Posts
16
Comments
1020
Joined
3 yr. ago

  • I don't think that"s the reason for the absird lobbying, which is why it's getting through.

    Governments aren't that competent to get this through at scale, but corporations can and have a very good reasob to do so.

    I believe that it's a lot more probable that the reason is profit - social networks have a problem with bots. Advertisers don't want to pay if 90% of impressions are bots, and you can't really solve that problem easily.

    Age verification by ID solves this, and if they even can lobby hard enough to outsource the costs to mandatory OS level veryfication, so they just call one OS api, it's even better.

  • They should make the API call for apps to query that value a per-system/boot randomly generated signature, so it's impossible to use while also complying with the law.

  • I'm mostly interested in how will they handle giving the info to apps. If it'd let me to block or fake the request depending on what I currently need (just prompt me every time an app asks, and let me choose the bracket), I'm good.

    Tbh, most sites that are slowly getting targeted by age verification laws are things I'm kind of addicted to and have been trying to drop for a long time. A "scan your face or id" dialog would be a good reminder to finally cold turkey it. It's one of the things I hate more than however much I need their platforms.

  • I really hope there'll be an option to disable / block this when I'm not a California resident, and an option that allows faking the result per-app and response (so I can i.e give randomized results most of the time, with an option to switch to "random 18+" when I actually need it).

    I also really hope that most distros will block California users from using it, or rather say that they can't.

  • Yes, they should in fact just state that Californian users are not allowed to use it.

  • Would it be possible to add a layer on top that shows you when an app requests it, and shows you a checkbox about what you want it to report? Or just block the call as not supported.

    Faking it to be child (or just random with each request) until you need a higher number could mess up with advertisers and in general fingerprinting.

  • I mean, it does spread awareness about it harmful effects and how to recognize them.

    I'd say thats a good to tshirt to have.

  • You should probably add /s to that, hah.

  • I've just started reading "Working Effectively with Legacy Code", and so far it has been great.

    While I'm having troubles with it mostly because I work in gamedev, and a lot of the TDD approaches are difficult to apply (or actually merge into the codebase, since it's simply not an industry practice and I'm nowhere senior enough to be able to push as big change in workflow), I've still learned a lot, especially cool tricks about how to add/fix features or refactor.

    It should be recommended reading for anyone who deals with codebases.

  • It's a for profit ad company making a "privacy first browser".

    Thinking for literaly a second about that sentence should tell you all you need to know.

  • This is exactly the reason why "but you can disable the AI features easily" isn't a valid argument. They moved resources from features that people want to focus on AI bullshit.

    I simply canceled my subscription after many years and will be staying on the perpetual fallbavk license of a version before they gave in to the AI craze.

    It's a shame, but itdoes save some money.

  • Deleted

    Permanently Deleted

    Jump
  • I remember seeing an amazingly visualised website that was showing a result of an experiment that made people talk with strangers, and it had mostly positive result for most participants, even if they felt uncomfortable or scared at the beginning, and it had gopd results even if people with differenting opinions about i.e politics talked. Unfortunately I can't seem to find it again.

    I don't know if it was part of the same research, but there was also a part where they had people talk to strangers on a bus ride, and it also went well in most cases.

    Sp, talking to strangers is mostly recommended and should be mostly positive, at least statiatically speaking.

  • I mostly work in gamedev where they aren't that much feasible so I don't have much real experience working with them and I might be wrong but from when I looked into it a while back, it's basically just a docker container that you specify in a .devcontainer file (at least for VSCode, but other IDEs probably have something similar) and when you need to develop, compile or run your code, it runs it in the container. It also doesn't have to run locally on your machine, if you can run docker somewhere else (i.e on a more powerful shared server).

    I can see several advantages (but I never really tested it in practice, so I'm mostly guessing) - containers are usually quick to start, you have the same and stable and replicable dev/build environment for all devs (since you just commit .devcontainers), so there aren't some hidden dependencies and "works on my machine" shouldn't happen too often. It also helps you keep your OS clean, so you don't end up with 5 versions of python, 3 JDKs and 20gb of random NPM packages installed in your OS after 5 years of development - which is the most important advantage for me.

  • Devcontainers are awesome once you set them up properly, no need to run a VM.

  • If you don't use any identifiable information for the account (i.e email, post any photos with your face, or real name, that you use for anythinh else), use a VPN, an anti-fingerprinting browser like Mullvad, and most importantly use IG just for DM and nothing else, you should be pretty ok. Just the people you talk to (and what you talk about) will give them plenty of data, tho.

    One way to avoid having to use the website or their apps at all is to run your own Matrix server with meta-bridge, which can bridge Messenger (and I think even IG DMs, never tried that tho), so you minimize the contact surface you have with their site, because you are chatting through an unrelated app.

    If you don't do most of that, Meta already has a shadow profile on you anyway, since they track stuff across websites based on numerous fingerprinting methods. I never really looked into it, but AFAIK most of websites have the "FB like" widget that Meta uses to track people across the internet, and I'd guess that Meta is pretty good at working with that data.

    Not sure about directly poisoning the analytics, but you could just run something like a VM with a pyton script with Selenium that just randomly browses web or IG.

  • Managing centralized security and device management correctly on multiple OSes must be a nightmare. From EDRs to app and device provisioning.

    You should do dev work in devcontainers anyway.

    Not that it's an excuse or that I'm happy with that, but I can totally understand why companies do that, and tbh I'd rather see a properly secured than have the option to run Linux.

    But I'm biased, because I used to do Red Teamings, and the things I've seen...

  • Someone check on the Bohemia Interactive devs.

  • I used to be an EVE player for a few years, and also tried the blockchain thing during some closed tests, and you're right.

    Ngl, I have to admint the technology behind it is pretty cool, if you get past the blockchain hate, and I do think as a developer that it's pretty interesting idea, especially the bit about modding and the idea of open source frontend (which I don't know if they ever got to, didn't really keep up to date with the project).

    It's a shame it didn't work out well.

  • They don't want to deal with the costs, just get the data. At least that's what I got from the text, as a reason why they were pushing to make social networks extempt and keep it on app/OS level.

    Fuckers.

  • So, like Farmville.