DickingAround

I run thousands of robots in production. We can get a very high success rate but only for the task they're designed for. Production robots can't pick up stuff they drop yet. And this '80%' level is not actually acceptable or even state of art for just pick-and-place, but it's compelling for a robot that also knows how to do other things with equal quality (if JEPA does that).

I think actually you could do that if you wanted to; look up what notes mean, write some little program to make a sound if you had to. You could do it in a week if it was your only job.

We should not under-estimate the need for speed in supply chains. Predicting future demand is hard. To be more specific, we're talking about predicting ~100M unique products (the order of magnitude that moves on the pacific) and some of them have very lumpy demand (e.g. invent a new product, but it depends on 100 other obscure products).

I think the core of the 'improved productivity' question will be ultimately impossible to answer. We would want to know if productivity was improved over the lifetime of a society; perhaps hundreds of years. We will have no clear A/B test from which to draw causal relationships.

This is a good point; a man or woman sitting behind a desk doing correlation analysis are going to look very similar in their function to a business. But they probably physically look pretty distinct to an x-ray picture.

Just because a whole industry is bullshit doesn't mean I should force it to not exist. I don't like musicals. I don't understand or care anything about their culture. But it has a right to exist. Some people are into musicals. Their existence or non-existence isn't my problem and it isn't my business. We cannot and should not try to engineer the world around what we personally find valuable and ignore what others find valuable, even if they got their opinions form an ad, or their parents did and they inherited it.

TINA = There is no alternative... to stocks. Lot of people lost money on long term bonds. Short term bonds are at least better than cash. Stocks are supposed to be about investing in an asset. I guess there is kind of an alternative out there.

Humans can do a lot of useful things with electrical energy. It's good to see our most powerful entities (companies) directly getting involved in getting the energy they need and still trying to do that in a climate-friendly way.

Sadly, that's still very non-causal. A lot of things in China were happening at the same time. We just need to be honest that we don't have causal data and we have to make decisions on correlation data (and not try to over-claim causal testing).

At a practical level, this is NetBSD story is a story of a machine that just does it's job and doesn't change. For example, there is no active experimentation, no machine learning added on, etc. It's like a wood-mill; good that it make 2x4s but also not a highly dynamic business. Probably not a business that's seen a lot of competition or having to adapt to changing markets and products.

There was no pre-mining in Bitcoin. There were early adopters, but I think we're all early adopters still at this stage. Really, it might not be CPU mine-able and $1/coin but it's still nowhere near final value if it managed to actually survive and win dominance over fiat. Clearly no one centrally controls it. Satoshi is gone and his money never seems to move.

In what way? There's a limited number of bitcoin (I get there's unlimited altcoins, but they're easy to distinguish). I've never heard of anyone being able to censor a transaction on the network (hard to imagine how you can unless you can cut off internet access). I'm making an honest try to understand how it could fail either of those, but I don't see it and need to be shown a bit more directly what you're claiming here.

Block rewards get replaced with fees. Part of the 'blocksize wars' was the decision to not just increase block sizes or (or course) increase issuance of bitcoin but to be honest and recognize that there will be constraints on what's in the core chain and it's going to cost money to get stuff in the core chain. There are fees now and eventually fees will replace inflation. But it's going to be hard to 51% attack it because there will be still people mining chasing those fees.

The constant combining of altcoins and bitcoin is not helpful. There's a use case that is 'money' in which the value of the object used as money is always deeply disconnected from the underlying asset value (i.e. we never trade factories and cars as money. We still don't even trade stocks as money). Bitcoin is attacking that use case head on. It has features like scarcity, no ability to stop a transaction, etc. as features over other forms of money.

Then there's a huge list of altcoins promising things that are different, are not actually different or helpful, and are really just still targeting being a money like bitcoin or the dollar except the founder of the new currency wants to be the one that centrally controls it. Thus 'crypto' is not a helpful space, but bitcoin could be.

It does seem to be almost some kind of stigma or classism with buses rather than an actual functional difference in technology. We have electric overhead-line buses and they're just the same as streetcars except you don't need dedicated lanes or putting in rails in the road (which are expensive, limit expansion, and present a real hazard to biking).