centimeter

This will hit your company/industry eventually.

> Why wouldn't it be a fine choice to write a compiler in?

Javascript doesn't seem suited to compiler construction because it lacks lots of features that make compiler construction pleasant (e.g. strong rich types, algebraic data types, etc.)

It might be "fine" but it's not "good".

For those unfamiliar, John Baez is great. One of the few people who seems to have a good, highly connected understanding across most/all of modern physics. This is very rare even among physicists. He's also credited in a lot of excellent ([hard] science fiction) books by Greg Egan.

What is the proposed application of NeRFs? It seems implausible to me that they will offer better practical performance than other rendering techniques for something like video games.

One of the tricks of modernity is to constantly lower the standard of living, claim that standard of living is actually based on something like the number of pixels on your TV, and say "look how much cheaper things are getting, and the standard of living is going up!"

Durable buildings, good food, and robust societies are all capital-intensive to construct, and huge amounts of capital have been destroyed and/or transferred away from you by certain post-modern political movements, so it's no longer viable for everyone to live in a good solid building with plenty of space. Instead, you have to live in a cheap pod apartment with thin walls. But don't worry, they'll remind you it's more efficient, greener, etc. You may not have a good kitchen, but you can order a bug burger to be delivered by a subliterate third-world indentured servant. Very modern and eco-friendly!

That's not what exponential means.

Cities also rely on roads, just in a different way; they require road networks to bring them food and manufactured goods, transported at greater distances than similar products in rural areas.

The net direction of subsidy from rural to suburban to urban is complex to calculate and subject to dispute.

For example, urban centers are net importers (of food, processed goods, building materials, etc.), and benefit disproportionately from heavy freight on roads.

An example of a serious but less obvious negative externality imposed by urban centers is the bulk import of population, since urban centers always reproduce below replacement rate. The urban centers basically take most of the intelligent people from non-urban areas and then ensure they don't reproduce.

If something is true, I don't really care if you're going to try to attach a negative emotional connotation to it (at least if we're discussing things at arm's length).

If the device manufacturers could get their shit together and come up with some cross-manufacturer federated standard, we could easily dump the public telephony network once and for all. I’ve never been anywhere (rural or urban) where public telephone was better quality than some kind of VoIP. Even out in the boonies, data is usually more reliable than voice (at least with google fi).

> My initial assessment is that there isn’t any correlation between IQ and test scores

This is completely false - I'm not sure how you got that assessment.

Many test scores (e.g. ACT) are heavily IQ/g-predictive.

SAT used to be better at predicting IQ/g but it's still pretty good.

g-factor is the single largest predictor of academic success (which, these days, is mediated mostly by test scores).

That doesn't clash with my observations at all. Environmental activists of all strains, including climate enthusiasts, are overwhelmingly white.

It was really difficult for me to buy a new car last year. Customized orders are totally impossible right now. Whatever they might have in stock (or have arriving soon), that's what you can choose from.

It's confusing because they refer to "200nm" in the same article.

Beyond matrices and vectors, it would be nice to have generalized (typed) tensors. What you can do with matrices alone is really somewhat limited.

I have a local ZFS backup server which sends encrypted incremental snapshots to my rsync.net account, no problem. You can't mount the encrypted snapshots since freebsd ZFS doesn't support that yet, but I don't need that (and it would defeat the security point anyway).

> The ethereum chain is 639.43 GB.

False. This is for some sort of pruned SPV node or something. A full node takes multiple terabytes. Even the fastest implementations take days to sync, if you're lucky.

https://tjayrush.medium.com/building-your-own-ethereum-archi...

> Which means that no one can use that balance for weeks or months

You clearly have no idea how lightning works, so why do you keep commenting as if you do? Nothing about this claim is true.

I tried FreeBSD for the first time last year on a network appliance. Never going back to linux (for a server). It's so elegant and well-put-together, and "just works" to a much larger degree as long as you don't need a fancy GUI. Has lots of awesome features like boot-from-ZFS.

For a desktop/laptop computer, I do not use it.

There are two problems here: The first is that a lot of ethereum talk is just that; hot air with little substance.

The second is that PoS has significantly worse security properties than PoW.

The core reason is that "changing your mind" (rolling back the blockchain) is free with PoS, except for possible loss of value of the cryptocurrency, whereas it's extremely expensive with PoW. There are all sorts of band-aids you can slap on like ignoring block reorgs of a certain depth, but then you lose properties like network partition tolerance.

It doesn't sound like you're familiar with how lightning works. It doesn't have to be "synced with the chain" except during channel opening and closing, which occur infrequently (like on the scale of weeks or months).

> ethereum and bitcoin cash already have more transactions than bitcoin

The eth blockchain is untenably huge (good luck running an ethereum full node), and ethereum fees are often higher than Bitcoin fees.

I assume this is a bad-faith question, given how much it begs the question, but: there is no known solution to the double-spend problem that achieves the same security properties as proof-of-work without consuming energy.

> the overall collapse of sense-making

I suspect this is caused more by the collapse of (false, but easily comprehensible) narratives than an actual reduction in coherent reasoning. Most "normies", so to speak, have always believed a huge amount of bullshit, and the forces that stabilized schelling points in bullshit-space have been diminished.

Traditional east coast media is losing their grip on narrative formation and they're lashing out at intellectual figures outside of their social group, especially from the west coast indie-intellectual scene (e.g. Scott Alexander).

All of these "clever" energy recovery schemes are inherently stupid, because you'd almost always be better off just making the original system more efficient rather than recovering some negligible fraction of its losses.

This is an especially poor choice of example given that it's the canonical example for STM, and STM actually solves the problem.

Having two servers is a lot more than 2x as complicated and expensive as having 1 server.

I spent 2 years in London and 6 years in HK so I think I'm reasonably equipped to make that judgement

> You know there’s more to building long lived protest movements than street fighting, right

There's also more to being successful than being "long lived".

If you're going up against the Chinese government, you need some degree of violence in excess of that levied by the HK protestors.

> The doc itself lists many forms of effective action

The HK protest actions were not at all effective.

I think Ireland's lack of a martial tradition from the British side was compensated for by the healthy martial tradition on the Catholic side. (No value judgement about the IRA intended.)

(For context, I lived in Hong Kong from 2014 to late 2020)

I'm not sure why protestors would take tips from the HK protestors. After all, they were not at all successful.

I attribute this problem to two things:

1) Hong Kongers are culturally mostly British.

2) Modern British culture has no martial tradition whatsoever, and therefore no practical ability to mount a successful resistance to any kind of even moderately-equipped state actor that is willing to use force.

If you're dealing with a polite government that's strictly constrained by voter approval, then the HK strategy might work.

WebEx is by far the worst and least reliably video service I’ve used (compared to Cisco jabber, jitsi, google meet, etc.)