(she/they)

  • 3 Posts
  • 218 Comments
Joined 1 year ago
cake
Cake day: September 25th, 2025

help-circle
  • Although i2p and tor have some similarities, I think tor is more designed for accessing the overall clearnet internet anonymously whereas i2p is more designed with the goal of separating websites and users from the clearnet internet as much as possible and thus creating an “invisible internet”.

    Again, they can essentially be used for both things these days, it’s just their original intent and thus their strengths are different. If the goal is to allow anyone on the internet to access the site, but also offer a more anonymized experience, then tor is probably better. If the goal is to keep the site entirely off the clearnet internet, then i2p is a better choice.


  • What’s the concern? The main issue with using biometrics is a person can force you to put your finger on the sensor or with face unlock can hold up the screen to your face (or some crappy tech just needs a picture of you).

    The fingerprint isn’t stored remotely, it’s used to generate authenticate to the app, not the account. That’s why you have to log in to the app the first time.

    Tech varies but at a high level that login returns a key that’s stored in a secure storage built in to the hardware on the device. It would be extremely difficult to intercept that key exchange or to access the key, even with longterm physical access to the device. This key is then and used to generate authentication tokens when you open the app. The key itself never leaves the device and typically isn’t even stored at the remote server, only another “public key” that can be used to validate the tokens are genuine but can’t be realistically used to reverse engineer a valid token much less the “private key” stored on your device.

    Of course it’s theoretically possible to reverse engineer a key or token, but as long as they use modern cryptographic standards given the current level of technology it would take the combined computing power of the world a very long time to do that for a single user. It is way easier to just guess or even brute force your password, which you should change periodically and should be long and complex enough and not reused to be hard to brute force or guess.

    I suppose there’s also some small risk that a hardware manufacturer (or their compromised LLMs acting as firmware coders) may build in a back door to their biometrics or a government may build in a back door to a new cryptographic standard that isn’t well known and force companies to use it, but again, it would be such a small pot of users it wouldn’t be worth it for anything but maybe a spy agency targeting a competing government’s leader. And something like that would be noticed long before it would affect the average user.

    So, I guess how paranoid are you. If you’re paranoid enough about remote access to your biometric based logins after that, then you shouldn’t use a phone at all. But again, a bad or reused and compromised password is exponentially more of a weak link.


  • If I ever had the option to unionize, or work towards organizing a union, I would have. It was never an option. It has always been a firing offense to even discus. And don’t say that’s not legal, because there’s story after story of people being fired for it in the US across many industries and the burden of proof that it was the reason for firing is on the employee and very expensive to hire lawyers to fight, especially in “right to work” states that don’t require evidence of reason for firing. Not to mention being essentially locked out working while you fight the case in civil court, so you need a heck of a lot of savings. So it’s a major risk. And the EBSA has been underfunded for decades before software was much of a thing. It’s hard to form a union, especially in the post-Regan-administration era.

    Being unhoused from not having income in the US also creates major barriers to reentering the workforce. Not to mention having no access to medical care while being exposed to the elements and hazardous living conditions is a major killer or at least makes finding work difficult. Having limited access to showers, clean professional clothes for interviews, phone number for call backs (especially before cell phones but even now it can be difficult to maintain in states with no phone assistance programs), permanent mailing address for background checks, hits to your credit score which can be used to deny employment, lack of nourishing food, etc. It can take many months to get basic government assistance after being fired, assuming your state or local government even offers any. I’ve been there, though not for union organizing, just because I lived in a small town with few jobs in the early 2000s when remote work wasn’t really a thing and relocating without savings was impossible.

    Trust me, it’s a minority of workers that don’t want unions. Even during the 90s boom, but definitely after the bubble burst.


  • Just my opinion, and not meant to be a 100% excuse, but I did at one point run a nonprofit encouraging ethical software practices with the goal of eventually empowering developers to refuse to write unethical things. But not long before Google dropped the “Do No Evil”, there was a huge loss of interest. They found out they didn’t need to care about their image anymore since they all have virtual monopolies/duopolies/etc. and the majority of people are willing or forced to sacrifice their morals for convenience or increasingly just to be part of society.

    Anyway, because of that, the lack of unions in the industry or really any protections or power of tech workers over mega corps, and the corporate-led crash and takeover of “Silicon Valley”. It’s generally difficult keeping a tech startup alive without a billionaire or corporate backer to protect you.

    So, there are very few well enough paying jobs for us in the software industry that involve the ability to say “no” to something obviously unethical. And there are so many of us unemployed for so long that it’s really difficult to change jobs, much less to one you won’t be blackballed at for “insubordination” in refusing to write the unethical code in the company you left. Then throw in the LLM problem and that they not only don’t understand the concept of ethics since they aren’t actually “AI”, but are designed or at least trained specifically in very unethical ways and on lots of unethical code and speech.

    So, not that it’s not that it’s not unethical, but that most of us have no choice if we want to be housed and have food and access to medical care. Just like it’s unethical to consume most anything in modern society without being extremely wealthy, including necessities like food that is unethically grown and sourced. (The show The Good Place honestly really helps make that clear if that’s in question.)

    Point is, don’t blame the software developers 100% (though some obviously really are on the side of genocide). Instead blame the corporate executives that not only allow it, but created an environment where being ethical in the software industry is nearly impossible these days.

    And “I was just following orders” isn’t an excuse I the military (yes I was in the military as well) because in most countries it’s legal and legally protected for a soldier to refuse an unethical order. That’s not the case in the corporate world.


  • Yeah, I mean in get that they need to verify your identity to avoid liability for giving out your personal information. But they don’t care if they actually give out your information improperly, its only a liability issues for them.

    First, there really needs to be a way for you to verify the caller before they verify your identity. But also, since these callers are often not employees of the company, just random call center company employees, they probably don’t even have a way since that would require the company build something into their websites or pay for a service to allow the contractor to trigger it. And I sure as hell wouldn’t call them back since that would mean waiting on hold since the call center employee will just get automatically pushed to another automatically dialed call (possibly even for a totally different company) as soon as they hang up with you.

    So, unfortunately there’s not an easy solution since the companies sure as heck aren’t going to pay for it. This is why I almost never answer calls. If they can’t leave a message and I can’t do something online, and they refuse to pay for enough employees to do it not online, then I probably will avoid using their services entirely unless there’s a really good value proposition. It’s yet another cost for me for using them.



  • Yikes, that’s pricey. But I suppose that’s the way of the world with LLM hoarding. I pit it on an off brand mini-pc with an n100 processor 4 2.5Gbit ports, 32GB RAM, and 512GB SD for under $500 just 5 years ago. Works great. And if you know Cisco, OPNSense is a breeze in comparison. I don’t use the CLI much, but it’s nice to have a more standard OS behind it than Cisco stuff so you can do a lot more if you wanted to. Though I put Proxmox as the base OS on the bare metal so I could install other stuff even more easily. I’ve found the documentation to be OK, but yeah not the best. Fortunately there are a lot of places to find more information, unfortunately a lot is on Reddit who is about to lock everything down behind accounts and I refuse to go back there after being a huge reddior up until the IPO fiascos and continued decline. Hopefully more people will move here.


  • I use a an older n100 mini-pc with 32 GB RAM, 4 Ethernet ports, and a decent SD drive with Proxmox to run opnsense, pihole, some arr apps and UniFi for my AP. Works great for me, but it does mean needing to configure the networking and such properly and it is another layer that will need to be updated and rebooted periodically. I reserve 16GB of RAM for OPNSense but I also use a ram drive for logs and cache files and such to reduce wear on the drive since anything important gets mirrored to my NAS.





  • In the US, health insurance generally depended on the condition, but generally if you were treated for or a doctor was aware of a condition existing you then lost coverage for a period of time,say when you became and no longer qualified for your parent’s insurance, and you then for new insurance and tried to get that conditioning treated, the insurance could deny those claims and you’d be 100% liable for payment. Could be a heart condition, could be depression, could be any chronic condition they had evidence of you knowing existed prior to any gap in coverage regardless of age or how long it had been since you were last treated for that condition. I was top young for life insurance and I wonk in the healthcare industry, so I don’t know the life insurance side as well, but I know even now ADHD and Autism can disqualify you from coverage, since I have those and can’t get coverage outside of employer group coverage or some high cost types of coverage. Also disqualifying for long-term care coverage.


  • And that’s just the way the parents are using the data, imagine having all that medical data so when it becomes legal again to exclude people from health insurance for “preexisting conditions” as “Obamacare”/the Affordable Care Act is ever being whiddled away, the insurance companies will have all that evidence of all those conditions right from the data brokers. Just one of many examples of course.


  • I host a mailcow instance. Had some issues getting the IP address the VPS assigned me removed from some blacklists, and some companies whitelist email domains and only let you get their emails if you use outlook, gmail, etc., but those are few. Unfortunately it’s often medical stuff that thinks they’re being “secure” by only sending mail to systems that steal your private data for profit, but not much you can do. And that happens with any email software if you use your own domain as your email address.


  • First avoid the major social apps. They are designed to give their companies access to your private information for advertising and selling, so of course these have lots of ways for nefarious attacks without user intervention since they aren’t likely to spend money keeping them out. Also, wouldn’t hurt to move to an OS that better isolates apps (like GrapheneOS) and avoid background services with administrative privileges like iOS or Google Services. GrapheneOS isolates Google services a bit or allows never installing them. IPhone has no real way to mitigate risk outside of the basics like encrypting your device and online data, but the admin services still have access. Not sure how easily they are hacked these days since they haven’t really been promoting privacy anymore, but have also started wanting to profit from your data. Still slightly better than google, but way less control over the device which means you’re at their mercy. I want even get started on Samsung administrative services…and I don’t know the others well.


  • Also for Android devices I find Rethink in DNS only mode with locally downloaded block lists works pretty well and is pretty private. Only issue is it takes your VPN slot, though it itself can connect to a WireGuard VPN as well, but if you need to enable and disable VPNs for various circumstances or need other types of VPNs that require nonstandard protocols or software to connect to then it becomes a pain. But this when when you leave your network DNS behind, you still have some protection and it’s all done locally.


  • Glad Meta locked me out of my account assumedly because I mentioned that their excluding anti-trans speech from hate speech used horribly flawed “logic” and giving examples of how that would translate if they used it against several other minority groups. Didn’t even get banned so I would have been able to appeal, just locked out with no recourse like many others at that time. It was a huge favor to me since I never went back. LOL.