jpk · HackerTrans

jpk

no profile record

comments

hace 2 mesesdiscuss

"Communal ownership of the means of production" evokes an image of a hippy co-op trying to buy pallets of GPUs, or something, which is probably why it sounds unattainable. But if you reorient that to something more along the lines of, "the Mullvad of hosted llama.cpp", then it actually doesn't seem that far out of reach.

hace 8 mesesdiscuss

The comment you're replying to made use of a simile, which is a figure of speech using "like" or "as" that constructs a non-literal comparison for rhetorical effect.

hace 9 mesesdiscuss

If only. If ICE arrests and deports someone without due process, there is no court case.

hace 9 mesesdiscuss

There's no such thing. The house always wins or the business collapses.

hace 10 mesesdiscuss

> A few individuals can completely takeover the government

That's not what's happening.

When most people serving in positions of government do so in good faith, most forms of government work, including the American one. When most people serve in bad faith, most forms of government do not work, including the American one.

The American system has checks in place to keep what is happening from happening, but those checks aren't working because those who would exercise them aren't doing so, as withholding those checks benefits them personally, at least in the short term. The underlying theory of the American system is that if you distribute power enough, one or a few bad actors can't seize total power.

But, there are just too many people in elected office right now who did not take their oath to uphold the Constitution in good faith. Namely, in Congress which has simultaneously demonstrated that it is unwilling to effectively wield the impeachment check, and is unable to do effective legislative work, leading to a latent desire for a stronger executive. In this circumstance, no form of government will hold up without a correction towards replacing all the bad-faith actors.

hace 12 mesesdiscuss

The point is you can't reliably tell if someone's choice of vehicle is wasteful unless you get to know them a bit. Snap-judging someone's entire lifestyle in the second it takes to recognize a make and model isn't constructive.

el año pasadodiscuss

I think part of this is because people often don't appeal to local government unless they've got an axe to grind. Nobody goes to the city council meeting to comment on how everything is great and things are fine the way they are. So when someone shows up to complain about ice cream truck music, the people who are pleased, or at least indifferent about it, don't show up to oppose the complainer, and the signal the council members get is that it's a problem and a city ordinance or whatever is required. There are typically opportunities in the local law-making process to allow someone to oppose the complainer, and it does happen, but few will match the complainer's level of effort. Then if a law makes it on the books, local LEOs become the complainer-class's customer service representatives, and you get what you're describing.

Ultimately, local civic engagement is often what matters most to your day-to-day life, which is good. I think effective and durable self-governance must start at the local level. But we get blasted by media related to national politics at every time and season, to the point that the thought of trying to stay dialed into local government is a non-starter for many. If all the attention we can bear to allocate to politics is monopolized by the national wedge issues of the day, who will muster the volition to save the ice cream truck music?

el año pasadodiscuss

Yes. I pine for the ATX of phones.

el año pasadodiscuss

It's not so perplexing when you understand that Python has long had the best ecosystem of libraries for data science and ML, from which the current wave of AI stuff was born. There are plenty of reasons to dunk on Python, but the reality is lots of people were getting real work done with it in the run up to where we are today.

el año pasadodiscuss

Just a drive-by thought, but: What you're describing sounds a lot like Temporal.io. I guess the difference is the "workflow" of an agent might take different paths depending on what it was asked to accomplish and the approach it ends up taking to get there, and that's what you're interested in persisting, replaying, etc. Whereas a Temporal workflow is typically a more rigid thing, akin to writing a state machine that models a business process -- but all the challenges around persistence, replay, etc, sound similar.

Edit: Heh, I noticed after writing this that some sibling comments also mention Temporal.

el año pasadodiscuss

I mean, it probably was a branch that several people contributed commits to that was squashed prior to merge into mainline. Folks sometimes have thoughts about whether there's value in squashing or not, but it's a pretty common and sensible workflow.

el año pasadodiscuss

For sure, but on balance, during that second: - A smart little girl aced her math test. - A loving father smiled at his kid. - A grandma blew out her 80th birthday candles.

Lifetimes in progress, building their own memories and experiences. So, two people may have died in that second, but 8 billion people lived.

el año pasadodiscuss

If you actually want to offer substantive contrast, then I'd suggest elaborating on what you find interesting.

el año pasadodiscuss

Your entire comment is threat modeling, which is great! But it demonstrates my point: using an old, insecure thing is sometimes obscure enough to fly under the radar of one class of threats, but presents a juicy attack surface for another class of threats. And one's own personal risk for any given class of threats will vary depending on one's circumstances. So the parent comment's unqualified statement of, "security is not a consideration for such devices," isn't quite right.

el año pasadodiscuss

> Security is not a consideration for such devices, because of their very limited number. Nobody is going to crack into your internet-connected Amiga except maybe some of your friends, as a prank.

This depends pretty heavily on your threat model. You're right that a device like this is exceedingly unlikely to get exploited by attackers casting a wide net against common vulnerabilities. But an attacker targeting you-in-particular would love to learn you've put ancient hardware and/or software on the network.

el año pasadodiscuss

The basis for my understanding is a convo with a Google engineer who was working on self-driving stuff around 10-15 years ago -- not sure exactly when, and things have probably changed since then.

At the time they used just a single roof-mounted lidar unit. I remember him saying the one they were using produced point cloud data on the order of Tbps, and they needed custom hardware to process it. So I guess the point cloud data isn't necessarily harder to process than video, but if the sensor's angular resolution and sample rate are high enough, it's just the volume of data that makes it challenging.

el año pasadodiscuss

My (tenuous) understanding is that the challenge with lidar isn't necessarily the cost of the sensor(s) but the bandwidth and compute required to meaningfully process the point cloud the sensors produce, at a rate/latency acceptable for driving. So the sensors themselves can be a few hundred bucks but what other parts of the system also need to be more expensive?

hace 3 añosdiscuss

Fun fact: The Adobe enterprise slack workspace is not only open to everyone to add emojis, but (AFAIK) every slack-using company that Adobe has ever acquired has had their slack workspace subsumed into the Adobe one, which includes bringing in all of the acquiree's emojis. :-)