computerdork

I agree, great coding is art too, and not just the code itself, but very much the problem solving process. It’s sad the highest levels of cognitive ability humans can do are no longer needed (or not nearly as much)

Agreed, really enjoy solving tough problems. Like having some interesting enhancement that needs to be carefully and surgical added into a complex codebase. Or writing an algorithm that grabs a bunch of info, sorts it in multiple ways, and uses this info to generate a useful result for our API (am thinking of specific past situations that have done this).

What is not fun is the repetitive stuff, where you have to search a code base for the same change over and over again, but it's not something you can just do a search and replace. Am happy to have AI do this (although, even this, you still have to read through every change).

Think some of it this as we don’t have a strong safety net in the US. But also read that is because cities are just so expensive. Yeah, for people making something like $40k/yr it’s much easier to fall into major debt. Had a couple of roommates who went through this, staying financially solvent in the city was a huge deal. It seemed like they were constantly treading water.

But, agree that we as a country can strengthen to safety net, probably good for us as a whole.

Agreed, and probably has already started with the high usage of the internet and YouTube

Agreed, it’s incredible and scary how natural and human it sounds. It’s gonna be pretty much impossible for almost anyone to tell if they’re talking to AI

Agreed:) no one’s gonna mate and continue the our human lineage, haha

Actually, used to live in SF. And yeah the tenderloin is pretty bad. Saw drug use and and a lot incredible things walking through those areas.

But wondering if that is also just a city thing? LA has Skid Row and just looked it up online, many metros have a Skid-Row type area.

Not saying that lenient crime policies don't make them worse, but it also might just be a feature of cities?

Think the Philly DA situation is a bit more unique and interesting. Yeah, agree, he was one of the first progressive DA's elected in recent times, and he is in many ways light on crime. But he just got re-elected by a 75% in 2025 for a third term. That's incredible and think there is a reason for this.

Am no expert on him, but as we're all talking about here, the legal system is really messed up. Yeah, there are a lot of terrible processes and laws that get in the way of justice, and a lot of the things his office did made a lot of sense.

Probably one of the biggest was the "Elimination of Cash Bail for Low-Level Offenses." The cash bail system for minor crimes really screws poor people, who can't make bail for things like shoplifting, and have to stay in jail until their pre-trial. (it's often something like $1000, which for middle-class people is not a big deal at all). But imagine a teen who had to stay out of school for a month or even two just because they stole something for $20 and his family can't put together $1k. The teen gets behind in class and stops trying, and this becomes the start of getting further and further behind in life. And a similar situation happens with poor working-class people for a minor crime, they can't pay bail, so they are in jail and can't work. They are often fired or just get into debt because they're not drawing a salary.

I don't know about anything about open drug markets, but think overall what the Philly DA seemed sensible (and just looked briefly and it seems like the crime statistics turned out well in Philly, but that could be from a number of things).

Makes a lot of sense:)

Yeah, as mentioned in my comment, there are uses other than for this rocket-jump maneuver they did that thrusters would be useful for.

Well, think you're only seeing the most splashy uses. For instance, at high speeds, could use the thrusters to first, increase acceleration and speed, and second, have them also able to do a bit of thrust vectoring to help push the car's tires on to the pavement for better traction. And then on turns, have thrusters firing to help push the car into the turn.

Also, maybe the car is able to hover for short to medium distances, which for a military and offroad vehicles, could be very useful to cover all types of terrain

Well, the first thought is that they borrowed the thrust technology and software from spacex. This makes sense, because in starship, they used to have a lot of pneumatic powered actuators for things like thrust vectoring control, then they switched to electric motors, because they are instaneously responsive for one thing (and I think more reliable as they aren't fluid based). From what remember, they originally used electric motors from Tesla, so makes sense that Telsa would in turn borrow from Spacex.

Wow, that is a great story! Really like how you used tech to treat a health issue. Keep up the good work!

Yeah, didn't find it painful either

Seems like it was due to a couple of factors:

1) Very old injuries, almost all when I was a child or teen playing sports or doing reckless things, which caused neuralgia (nerve pain). Two were on the head (one from skiing, one from playing baseball and getting hit with a bat on the head by accident). Another on the tailbone that am focusing on now

2) Very bad posture related to these injuries - had horrible posture as a teen, and thinking back, this is very probably due to my tailbone pain. It caused a ton of pain when sitting and also caused me to twist up my pelvis and hips to avoid it. This terrible twisting on top of the other pain destroyed my posture and clenched up my neck, and for the past 30 years, have been trying to undo those postural problems and muscle clenching. Have done a lot of physical therapy and exercises, but it's much better now (although, it's still a work in progress and will probably never be finished).

Have had terrible pain and muscle clenching in my body since I was a teen (am 52 now) and for the past 15+ years, have been intensely working with the medical system to figure out the cause (have gotten a lot of great relief!). Agree with everything said in this blog in working with the medical system. To put my spin on it:

- You are the least-constrained team member in your healthcare: At an appointment, your doctors only get to spend ~30 min on your problem and it's unrealistic to expect them to be as deeply involved as you are on your healthcare. Yeah, you yourself can spend days or even weeks researching you health problems and planning your healthcare.

- Research your health problems: you don't need to be an expert to read about your health problem, learn the basic anatomy involved and the high and mid-level details of the surgery or treatment. You'll have to look up a lot of terms, but it's not that bad and the articles aren't that difficult to understand at a high-level.

- You are also the manager of your healthcare: Planning and seeing the big picture is critical. You can optimize a lot of the planning of appointments, exams and surgeries buy thoughtful planning. And, you can definitely catch a lot of opportunties and mistakes in decisions that are made. Just like in a tech project, a couple of good random ideas can lead you to some fantastic outcomes.

There are many more, but yeah, working with the healthcare system is an skill that can be studied and learned. Really useful.

It's an over generalization but think this person has a point. The US health system has it problems (cost!!), but have read that since it's money driven instead of a government service, compared to other healthcare systems, it's often got a shorter wait time to see a doctor. This makes sense, a lot of students want to become doctors in the US, because they can make a lot of money

Completely agree. Imitation is a great way to learn. But yeah, then typically you need to extend with your own ideas

Interesting. There is of course be a line between condoned use and unwanted stealing. And now that you’re mentioning it, think there is a higher tolerance for humans using a person‘s ideas versus AI. Think this is not only because a human is a real person and typically is also associated with music who has studied it to some degree, but also because when AI uses your ideas, you’re basically giving free access to them for any of the millions of users of the model. Hmm, Useful to think about

Agreed:)

Agreed... and am realizing, especially with themes/melodies, sometimes it's not always easy to modify them enough to the point that it's unrecognizable, but even then, you could be said to quoting their theme - although, am of course not talking about quoting popular music melodies, no one wants a lawsuit:)

Actually, didn't hear about this, but makes sense, Brahms clearly builds off of Beethoven.

But, have had heard the same thing that was said in the page you linked, that one of the reasons Brahms took so long in his life before he wrote a symphony is that he had to compete with Beethoven, and didn't feel it was ready until much older. Yeah, what an immense weight he must have felt to have to follow up what Beethoven had done!

Am an aspiring composer. We often say we have no qualms about "stealing" from other composers. Yeah, if we see an idea or technique used in piece of music we like, we won't hesitate to try it out in our own pieces. Of course, we don't actually steal to the point of plagarism, but it's pretty common to do, and the ideas we take are often modified for your own needs or just taken in parts.

Seems similar to what is happening with software developers.

Very useful info. Didn't realize model releases often reused the same pretrained base-model and just enhanced the post-training RL. And also that contractors are creating tasks and solutions for this post-training. Super interesting and makes a lot of sense.

... And yeah, agree with your statement that this is not emergence. But to me, it's still impressive, because the more post-training tasks they create, the more the model can do, and this set of abilities is only going to keep growing and growing.

... Also, if you wouldn't mind one last question, have read that the AI companies have started to do self-training, where the models are starting to train themselves. Might this lead to even faster and faster increase in model capabilities? ... although, based on what you're saying, maybe it just leads to a model reaching a its max plateau for existing capabilities rather than an explosion in new ones. Because, as you mentioned, for a model to learn an ability, it still needs a human to create a training dataset for it. Is this your understanding?

haha! Speak for yourself, but software never judges you or cancels on you at the last minute when you're supposed to meet up at the bar. "Friends," sheesh! :)

Also interesting. Do you happen to know if once they've used those thousands of contractors to teach the model something like Blender (or some similar app) using RL, when they training a new model, do they need to use same contractors again to teach that new model the same behaviors?

The reason I ask is even though it can take hundreds or thousands of contractors to teach a model a certain behavior, wonder if they really only need to do it once for each desired behavior? (of course, future models might expand and refine this previous training) Because if that is the case, then wow, then future models can really expand their capabilities really very fast.

...Am wondering if they somehow record a training session so they can play it back whenever they need to train a new model with the same info? Or maybe the new models can just use distillation from the old model to relearn the old behaviors?

Interesting. Have heard about the need to create synthetic data for LLM's, but didn't realize it was such an important factor. Although, how big a factor synthetic data is the next question, but guess there is no way to truly verify how much difference it makes with these closed models.

Ah, I see, LLM's aren't able to do incredible things and you were just commenting on that. Sounds reasonable

eh, they are also used extensively in medical research using alpha fold, and in math, and in legal for one thing. And this is of course only the beginning, and IMHO, it'd take a lack of foresight to not start putting layers and layers of safeguards now.

Yeah, to me the hugging face hacking is a harbinger of what's to come, not just in the world of tech but eventually to other fields too.