Gold is just one of many commodities these days, mostly unconnected from most monetary systems for many decades. Treating it as the benchmark of value is really quite arbitrary, and I expect someone could compare the S&P to other random commodities and come up with completely different conclusions...
I'd definitely be curious to see the S&P valued in different commodities over time. With that said, gold certainly feels like a special indicator given its history as a universally recognized store of value.
> history as a universally recognized store of value.
History of what now? Gold is a volatile commodity. It has crashed, many times, often catastrophically, and had bear markets that dwarf anything you see in stocks.. A quick search tells me that inflation-adjusted gold prices dropped like 80% between 1979 and 2000.
And given its value right now, it's probably due for another.
I think it mostly just means a few hundred billion dollars of value wiped from the stock market - all the models that have been trained will still exist, as well as all the datacentres, even if the OpenAI entity itself and some of the other startups shut down and other companies else get their assets for pennies on the dollar.
But it might mean that LLMs don't really improve much from where they are today, since there won't be the billions of dollars to throw at training for small incremental improvements that consumers mostly don't care to pay anything for.
> You will have a switch at home (perhaps in your router) with more than two ports on it. At layer 1 or 2 how do you mediate your traffic, without CSMA? Take a single switch with n ports on it, where n>2. How do you mediate ethernet traffic without CSMA - its how the actual electrical signals are mediated?
CSMA/CD is specifically for a shared medium (shared collision domain in Ethernet terminology), putting a switch in it makes every port its own collision domain that are (in practice these days) always point-to-point. Especially for gigabit Ethernet, there was some info in the spec allowing for half-duplex operation with hubs but it was basically abandoned.
As others have said, different mechanisms are used to manage trying to send more data than a switch port can handle but not CSMA (because it's not doing any of it using Carrier Sense, and it's technically not Multiple Access on the individual segment, so CSMA isn't the mechanism being used).
> That's full duplex as opposed to half duplex.
No actually they're talking about something more complex, 100Mbps Ethernet had full duplex with separate transmit and receive pairs, but with 1000Base-T (and 10GBase-T etc.) the four pairs all simultaneously transmit and receive 250 Mbps (to add up to 1Gbps in each direction). Not that it's really relevant to the discussion but it is really cool and much more interesting than just being full duplex.
New coal data is out just a few days ago [1], it's plateaued globally and expected to start to decline.
China's consumption this year was about the same as last, and looking to drop a bit, so likely old coal plants were being retired at about the same rate as newer ones were built, and that will start to go the other way (more retired than built).
Grid forming inverters for providing virtual inertia are only going to get better and better, there's no reason that as those control systems improve why synthetic inertia won't be able to be basically identical to real spinning mass. In the meantime while that technology matures, synchronous condensers can provide grid inertia without needing nuclear or coal, we already have about four in Australia supporting our grid and will probably have another dozen or so built over the next decade or two.
I have one machine that runs Windows (apart from one Windows 11 VM on my Mac laptop I use for work), all this nonsense has got me to install Fedora on a separate M2 drive on it, and I haven't booted up Windows in a few days now. Will be an interesting experiment, I've run it before but more for fun, but will try to go as full time on that computer as possible.
I do agree, but there is still a valid logic behind what is shown because it's only using the pigments that there is direct evidence on the statue for - but to stop this confusion, maybe there should be three versions of each statue at these kind of exhibitions (assuming these are all replica castings and they're not re-painting the originals!) - a blank one to appreciate the unpainted form, the reconstruction of the base layer that only has the pigments found in the crevices (what this article is complaining about), and then an artists impression of what it probably looked like properly shaded (given that we have the evidence of painted statues as shown in the article).
Then you could still have the evidentially "pure" one, but also have a more likely rendering to reduce confusion.
As a data point, I had mostly only seen what the author is complaining about in the past, with articles having more of the "you won't believe what ancient statues actually looked like" angle and implying that it's just our taste that changed.
So I definitely feel that I was misled by what I had read and seen about painted statues (though I was always a bit sceptical), even though everything I'd seen was from secondary sources (news sites etc.), and not articles or papers written by the reconstructioninsts themselves, so I don't blame them directly.
reply