you link to a page with convicted police numbering in the tens in a nation of 340 million people, with a police force on the order of a million. I wouldn't believe you in a second if you said that the police commit crimes at a rate of 0.0001 per capita. That's absurd. You're basically verifying the claim that the police are not held accountable for breaking the law. Great work. If that was your intent, please do more than post a link, and elucidate your opinion in the future please, if it wasn't your intent, well, next time just please don't post, it's not a useful contribution to the discussion in this forum.
Those are the ones that were high profile enough to warrant a Wikipedia page, it's not exhaustive. Here's a more comprehensive database: https://policecrime.bgsu.edu/
You're still citing arrested, not charged and convicted though. Those are all different, with no guarantee of the officers facing repercussions beyond a brief arrest. While those are still consequences, they have to be consistently applied (which they don't seem to be for police officers in America) or have consequences for consistently poorly behaved officers
The comment I responded to didn't have any data. You don't have any data, not for the US and not for any imaginary country that is better than the US at holding police accountable.
> police commit crimes at a rate of 0.0001 per capita. That's absurd
Once again, Care to share comparative data with other countries? I will.
I am amazed, though not entirely surprised, that these models keep getting smaller while the quality and effectiveness increases. z image turbo is wild, I'm looking forward to trying this one out.
There are probably some more subtle tipping points that small models hit too. One of the challenges of a 100GB model is that there is non-trivial difficulty in downloading and running the thing that a 4GB model doesn't face. At 4GB I think it might be reasonable to assume that most devs can just try it and see what it does.
Quality is increasing, but these small models have very little knowledge compared to their big brothers (Qwen Image/Full size Flux 2). As in characters, artists, specific items, etc.
Agreed - given what Tongyi-MAI Lab was able to accomplish with a 6b model - I would love to see what they could do with something larger. Somewhere in the range of 15-20b, between these smaller models (ZiT, Klein) and the significantly larger models (Flux.2 dev).
I hate that excuse. I want the model to know who the Paw Patrol is without either finding a lora (which probably won't exist because they're mostly porn) or needing to make a dataset, tag it, and then train it myself.
Is there a theoritical minimum for params for a given output? I saw news about GPT 3.5, then Deepseek training models at a fraction of that cost, then laptops running a model that beats 3.5. When does it stop?
Houses have windows, and bricks are widely available, locks don't protect you, they exist as a social contract: "don't go in here," if people want to go in there, they will, but we have a mostly civil society (fair to argue against this), and locksmiths exist, so this is just a tool, and a fun project.
Almost every digital lock I've seen (I own about 5) have a keyway hidden under a rubber tab or plastic tab. I haven't built an EMP device yet, but supposedly that works on a lot of them too.
I am pretty sure anyone building a lock picking robot is at least aware of some lock picking basics, and that's clearly not the point of a project like this. They could get a Lishi, a rake, a comb, almost anything, and we could also argue that this won't work with pin-in-pin dimple locks, tubular locks, and disc detainers, but again, obviously not the point.
This is a fantastic idea for a community challenge! Would have to make some ground rules though or else the winning bot is just a diamond coated drill or a gallium squirt gun.
Thanks. Yeah, I was talking to a lock picker and his hourly rate is more than my attorney. He tells me, that is why he stopped being an attorney. Badda boom.
Then get more? Sounds like a fantastic way to waste military resources. I have no clue why this mythical US military might and efficiency idea persists after so many failed interventions.
I'm genuinely interested in learning more about the shortcomings of meshtastic if you have a link to share. Groups like the Anarchist Black Cross seem really supportive of the tech for disaster situations. Even Benn Jordan claimed it played an important role during the floods in NC
My understanding is that it relates to the flood routing in meshtastic. I haven't heard a real-world failure example, but another comment on this post mentioned defcon being a case (I don't know anything about that).
I'd really like to learn more about Reticulum. Are you using a specific app on top of it? Do you have a meaningful user network you communicate with, or is it still more just proving out the system?
Triangulation is damn easy. If the US can put on bomb on a suspect satellite phone user back in the 2000's (and they did!), they can certainly send a bomb on that today.
Sat phones during the second gulf war (maybe even the first) became a liability. The transmission lit them up like a god damn beacon saying, "Bomb goes here!".
Triangulation, the math isn't the hard part. Where exactly on the continental United States are you proposing dropping ordinance? MOVE in 1985 was controversial even back then.
It can have challenges, but triangulation can be done with signals that have recognizable patterns or features -- even in a sea of other co-channel noise sources.
If you can observe the signal strength of your neighbor's home router while standing next to your own even if the signals differ in strength by some orders of magnitude (which is easy on Android; no idea bout iOS), then anyone else can also do the same.
The intervention part is an administrative problem the military isn't designed for. For the core mission of collecting intelligence, eliminating targets, and occupying land, the US has an unrivaled track record over the last 85 years.
I don't think it's going to be military killing a
Americans. As of now it more looks like federal government.
Nevertheless, sure, in the rural areas, but less so in the cities, reflections and bending of the waves make it much harder, and a single repeater with solar panel and battery could plausibly be made under $50.
They are being made. I have a four node network already in my suburb. There is a software project that is written in Python that essentially turns lorawan nodes into BBSs similar to briar.
Why would they bother? Super low bandwidth unencrypted communication they can triangulate if they really need to sounds like a perfect thing to let keep running and just monitor. Then you can triangulate just the "seditious" people who incriminate themselves.
I guess if you were relying on the meshtastic network as a backbone network replacement, which I'm not sure much of anyone is even currently setup for and I've heard isnt really feasible with the naive meshtastic toy implementation, you could be sending encrypted traffic. But then you have to have pre-shared encryption keys for participants and it will significantly lose it's usefulness for adding new adopters.
They're incredibly easy to build and even disguise as lawn ornaments as Benn Jordan showed in a recent video. When it costs us less money and time to build them than it costs the gov't to find/destroy them it's a worthy investment
Maybe ham repeaters but when we are talking lorawan they will have a hell of a time taking the networks down that are already established. Just in my suburb we have more than 6000 nodes because of the helium network.
Renee Good was killed after dropping off her six-year-old child at school. I agree with you, but people like her have children and are not trying to die in the street just for looking at somebody the wrong way. And it's one thing to open carry, it's another thing to become a trained and confident marksmen.
And as someone who has had half a dozen police officers simultaneously pointing guns at my head, mistaking me for someone else in public, once you're in that situation, escalation is only going to lead to death. Out here, police shoot you if your hand goes anywhere near your waist.
Alexander Hamilton explains his definition of what "well-regulated" is - and the purpose of a citizen militia - in contrast to the standing army in the Federalist Papers, No. 29. Most of the idea has become much more federalized than intended with the National Guard, but it has long since been misused for its intended purpose.
I think the most you can say is that they recognized that many male propertied white protestant Americans are American. Maybe some more qualifiers are necessary.
