• 0 Posts
  • 38 Comments
Joined 10 months ago
cake
Cake day: June 4th, 2025

help-circle
  • This is fundamentally not how Signal works, but you are generally correct in that a phone number has been shown to provide a lot of context for a person (or a device, at least). But Signal (the app) only uses a phone number for initial verification of an account. You have a lot of options to break that association with you - use a landline and get a call verification code, use a VoIP number (assuming you trust the provider), use a burner SIM, etc.

    Once you have an account, you can choose to identify yourself on the network solely via username so the registration number is not presented to other users. The Signal protocol itself is well-audited and generally secure.

    If your issue is with Signal the American company, use an open source fork like Molly with your own UnifiedPush instance. Then you’re only trusting them with transport of your encrypted messages, which again have shown to be secure at least in public audits.


  • In the US, I largely agree with you. Or use a website from a mobile browser. Different story in different countries where a smartphone might be the only compute the average person has, or where state services are tied to a mobile ID or bank app.

    Not saying that should be the case, but if the choice is between running niche FOSS apps and removing yourself from societal benefits structures, I know what most people will pick. That’s the real danger of allowing one company to own an entire ecosystem and have enough power that they have conversations directly with governments about their people instead of with their people.



  • LLMs specifically are great for intermediate use cases. You had a campaign in mind, but needed help with visuals. I was designing a piece of jewelry and had a series of reference images. Fed all those into a VLM and got something closer to my imagination, but still worked with a jeweler to realize the final product.

    These tools are best when you have a foundation of knowledge and need a little extra guidance, but fall off when you get to deep expertise. I’ve used them to troubleshoot my server but I already had a basic understanding of how a config should look. I also wouldn’t trust an LLM to properly configure something like crypto for it.

    To me, the biggest ethical concerns surround the training and creation of LLMs - stealing artists’ work to train them, energy usage, etc. I suppose in using the models I’m creating ongoing demand for them, so I’m not sure the answer. The best I’ve seen so far is what Anthropic used to espouse, no new frontier models until we can guarantee safety. And I’d throw in “utility”. Train new models when people are actually using them and clamoring for new use cases, not because a bunch of private equity shows line go up.









  • What about cases where the move is only temporary? Should people sell every time and hope there is a place to live when they return?

    In the private lender case, do you see that as different from someone who starts their own company and manages the property themselves while the renter pays them directly?

    The earning equity piece isn’t necessarily incorrect, what the owner is losing is potentially the opportunity to move at all. This assumes they can afford a mortgage + whatever it costs to live somewhere else.


  • The economists’ answer is that renting exists for the people in this situation. You may be moving to another country for a year or two. Are you going to buy a new house every time you move? Renting gives flexibility in that regard.

    Likewise for refugees, putting them up in a rental is a more efficient solution than building new housing for each family.

    That said, the model provides an inherently exploitative market and needs some kind of overlay to function efficiently, which in most US cities it doesn’t at all.


  • 100% it’s always a question of your resources vs theirs, but you’re dead on to make it harder.

    I’ll just add to also turn it off, pull the SIM, and show in a Faraday bag on your way back too. If the recent reporting about ICE buying location data from ad networks shows anything, it’s that they are interested in a capability of following people to and from protests. Graphene should obviate this by disabling Google Play services by default anyway.

    You should leave your regular phone at home, go to another place, power on your other device, speak your part in public, then travel another location and power off. This provides no consistent start or end location to work with for a particular device.


  • Alright, I already “umm, ackshually’d” someone in this thread but this post in particular hit a nerve with me. The Tor security model is based on 3 hops but does not guarantee 3 different jurisdictions. Their circuit building only takes into account “jurisdiction” in the way we’re using it here if you use guard nodes or specific cases when you cannot access the network directly or look like you’re exiting from a Tor node.

    That said, it’s still a very strong project and security model. And everything you said about spreading out your providers without a single point of failure (or pressure) applies.




  • You can plus this up into a vodka cream sauce fairly easily. Throw some garlic in with that onion once it’s starting to look translucent. Cook the tomato paste for a bit until it starts taking on some color then add like 1.5 tbl of vodka and let it cook a bit to lose some of the alcohol flavor.

    Throw in some butter and Parmesan cheese at the end if you have it. Also can incorporate different meats if they’re cheap that week and scales up real well.

    Look up Brian Lagerstrom vodka pasta on YouTube or Piped for the recipe I’m referring to. He also gives options for shelf stable replacements so you can stock up.