While it's likely due to other factors (i.e. like maybe the stock indices have just completely de-coupled from reality or are just being helped by AI-hype?), the fact that US job openings seemingly de-coupled from S&P 500 in Nov/Dec 2022 when ChatGPT was publicly released (as a web app) is pretty interesting.
So much coincidentally happened in 2022 and I'd feel like I'm wearing tinfoil if I tried to connect them. Google/Apple changed how ads worked, S178 was deleted, we were at the tail end of a global pandemic that we're still feeling to this day, and of course ChatGPT started to take the world by storm. And those are just off the top of my head.
There was more also going on in that time-frame: several interest rate hikes, no fix for section 174 changes by the end of 2022. Maybe someone will pinpoint whatever had the largest impact in a detailed study.
I still think some sort of shift in trends. Maybe it could have been COVID totalitarianism ending and as such shifting trend in leadership to reduce investments. Lot of headcount decision might not be actual needs, but what in general looks good for stock market. And if there is no more growth expected from WFH, well less hiring and reducing labour force looks better.
ChatGpT was used by probably 0.01% of people back then. I highly doubt it would have a discernable impact back then. Even in 2023, it would be way too early. Early adopters were the main users, not the mainstream public
No, it’s because $15T (80% of the money supply) was printed during COVID and while that money initially circulates through the system it eventually pools up into certain places like stocks and housing
It might not matter hugely to most people, that's true, but as someone who's used eSIMs while abroad in both Australia and Canada earlier this year (from Airalo and Nomad - they seemed at the time to be fairly well regarded), I was surprised to see my traffic routed through Hong Kong in both cases.
Google and Duck Duck Go both on the phone assumed I was in Hong Kong when searching, even though I was in Sydney and Vancouver respectively, which did make searching for local places a tiny bit more frustrating.
When the selected here its are using the worst lowest bar providers that are reseller of lowest cost network with the absolute lowest quality, In this case roaming probably Three HK and Plus Poland are the "norm"
These are some of the most "slop" provider, which is mostly ads and affiliate links unfortunately. It's same reputation as nordvpn whereas the best you could say is it's well known
Is this supposed to signal confidence in the chips already available on China's domestic chip market, or is it primarily aimed at boosting that market to make it ready?
How big a deal is it to be on the cutting edge with this? Given that models seem to be flattening out because they can't get any more data, the answer is "not as much as you would think".
Consequently, a generation or 2 behind is annoying, but not fatal. In addition, if you pump the memory up, you can paper over a lot of performance loss. Look at how many people bought amped up Macs because the unified memory was large even though the processing units were underpowered relative to NVIDIA or AMD.
The biggest problem is software. And China has a lot of people to throw at software. The entire RISC-V ecosystem basically only exists because Chinese grad students have been porting everything in the universe over to it.
So, the signal is to everybody around this that the Chinese government is going to pump money at this. And that's a big deal.
People always seem to forget that Moore's Law is a self-fulfilling prophecy, but doesn't just happen out of thin air. It happens because a lot of companies pump a lot of money at the engineering because falling off the semiconductor hamster wheel is death. The US started the domestic hamster wheel with things like VHSIC. TSMC was a direct result of the government pumping money at it. China can absolutely kickstart this for themselves if the money goes where it should.
I'm really torn about this. On the one hand, I hate what China does on many, many political fronts. On the other hand, tech monopolies are pillaging us all and, with no anti-trust action anywhere in the West, the only way to thwart them seems to be by China coming along and ripping them apart.
Arguably, the leading RISC-V IP is from US firms like SiFive; it has also caught major traction with NVDA in custom products, and US Govt for various industries, and even Redhat is now supporting RH on Risc-V
Microchip Inc partnered w US Govt on the aerospace angle and funded Canonical for linux ports. Their Polarfires and now Euro aerospace like Gaisler are heading in the same direction. US Govt/DARPA and others have been funding risc-v ports for years, to include mainly automated porting.
There are big differences between lowend profile-challenged SBCs and the work of NVDA, Microchip Inc, and the US Govt in the much more highend GPU related, and safety critical industries.
With Heavyweights like IBM/Redhat now on risc-v joining canonical and others, the SW side is definitely improving
I think those are both the case: they’re telling Chinese companies to invest in domestic hardware–implicitly also saying things like being prepared to stop using CUDA–and that means the hardware vendors know not to skimp on getting there (a nicer version of burning the landing boats on the beach).
It’s also an interesting signal to the rest of the world that they’re going to be an option. American tech companies should be looking at what BYD is doing to Tesla, but they’re also dealing with a change in government to be more like Chinese levels of control but with less maturity.
My cynical view is that it's mostly trade war and nationalism. If you follow the official PRC position, the chips are already made in China because TW is CN... Practically buying TW chips is boosting it's economy and hence funding it's military so from that perspective that makes sense. From long term development perspective this will absolutely boost national market... however that will take an insane amount of time. If you buy into AI is going to change everything hype, this move is a huge handicap and hence a boon to external economies. And I am probably missing a ton of viewpoints... politics meh
I mean, progress is already getting slow in llm development space and their qwen models are, well, good enough for time being. Meanwhile, its good for the world that they are working on their own chips, that way nVidia will have to stop being comfortable.
This is step in good direction for everyone except nvidia and its chinese distribution network
In 24 months, US hyperscalers will be training models on GPUs/XPUs with 16A process technology and HBM4E. The gap between the raw processing power of US and Chinese AI hardware will be widening.
> Does anyone know why they are still in widespread use?
Because of a lack of things compelling people to change them until it causes a breakage. And then when it does cause a breakage, most people would rather move heaven and earth to complain, research workarounds, etc. rather than just change it. (Institutional structures can also make "just" changing it far harder than that should be.)
There’s definitely inertia but I think it’s also that the US/ names match official usage: nobody, not even residents, says the time zone is New York because the official name is Eastern time.
Some of this is surely just muscle memory or intertia as well. I remember random config values from when I was trying out linux boxes back in high school that I replicated into files that just don't get touched for decades afterwards.
When was the last time you rebuilt your company's postgres config from scratch?
> When was the last time you rebuilt your company's postgres config from scratch?
Last year, when we upgrade to version 17.
I looked at the example/template configuration, diffed it with our configuration from PG15, and for every change decided whether to keep our version or the new setting.
I didn't use it, but Debian/APT has had a tool to do this sort of comparison for any software upgrade for as long as I can remember.
Do other people just copy the old config and shout "YOLO!"?
When your configuration value from PG15 was different from the old default and the new default on something like the statement timeout, and the statement timeout was previously working fine, and obstensibly set for a reason.... I dunno, are you going to question the value that much?
And you might say "well I know how a statement timeout works" and I agree! I would also generally agree that something like a timezone setting would generally be something I'd expect to be fairly stable.
That's what I meant about rebuilding the config from scratch. Rebuilding from scratch would almost involve _not even looking at the existing configuration_ and then doing first principles footwork to figure out what is needed.
I think the diffing flow you went through is the right way to do it, but I believe that flow might lead to some values getting less scrutiny than others. Still perfectly reasonable though
Still typing "nano -w filename" each and every time since back around y2k when I was working on Linux for the first time I was told that bad things could happen if I didn't...
I suspect some of it will be because the legacy form is a bit more intuitive than the standard form. You don’t really use continents and cities as a reference to time zones normally, countries and local subdivisions makes more sense, but as other people note, it brings up POLITICS.
You don't use them normally in the US, I've been referring to europe/amsterdam or europe/paris all my life in Linux installers and various equipment. I've never ever encountered netherlands/amsterdam or something like that.
It’s the same in Europe as it is in the US. Normal people refer to Europe/Paris as CET, just like normal people refer to America/New_York as Eastern Time.
Likewise with the dot com bubble and the web - it was a bubble and it was overhyped, but it was still transformative if you look back 20 years as to how things are different in terms of media, and commerce.
Both smartphones (and tablets, and smart watches) and the internet went through the hype cycle [0], and the sentiments I've been reading lately indicate AI is in the "trough of disillusionment" right now. That said, I don't believe AI will ever reach the heights (i.e. the measurable ones, how much it penetrates our lives, how much money goes into it) as either smartphones or the internet had. Probably higher than VR / AR, but nowhere near the other ones.
It will of they can actually make it think better than we do. Whether they ever will is hard to say, but it feels pretty clear that throwing more money at LLMs isn't going to get us there.
Huge offshore wind farms with hundreds of turbines are counted as single counts in this map though, so it's not really a compareable thing I don't think.
What would be the point of that? It seems to be by complex. You would not count every single generator separately at a hydro plant, would you? Every single solar panel?
What's interesting to me along these lines is I assume most of the companies funding the research are targeting the "creative" media in terms of image generation, music generation, avatars, speach, etc.
I can understand it's very interesting from a researcher's point-of-view (I'm a software dev who's worked adjacent to some ML researchers doing pipeline stuff to integrate models into software), but at the same time: Where are the robots to do menial work like clean toilets, kitchens, homes, etc?
I assume the funding isn't there? Or maybe it's much less exciting to research diffusion networks for image generation that working out algorithms for the best way to clean toilets :)
There are companies out there working on those problems as well. How the funding climate for them are. I don't know. But the market for smart robots, should be gigantic. So there must be some.
Keep in mind that what is easy, and hard for a human, which is the result of billions of years of evolution. Isn't necessary the same things that are hard or easy for our technologies.
Or replacing CEOs, investors, bankers? I would have thought those would be easier to replace than creating robots to clean or replacing artists, or even developers. Maybe I am wrong?
All these jobs are more who you know not what you know. The social network of these people is often an integral part of the work, so they are in a sense much safer than programmers, accountants and artists.
robotics is difficult and since transformers are just next word predictors they can't actually help us design those robots
:)
also the billionaires have help so they don't give a shit if the menial stuff is automated or not. throw in a little misogyny by and large too; I saw a LinkedIn Lunatic in the wild (some C-level) saying laundry is already automated because laundry machines exist
fucking.. tell me you don't ever do the laundry without telling me. That guy's poor wife.
The Parent poster is arguing that the "recommended" way documented on the Rust website to install rustup is using curl bash, and you're saying "it's possible to do things manually".
How is that helpful to the vast majority of the people on Mac/Linux trying to install Rust from scratch and reading the instructions on the website?
> ... to install the rust development toolchain (because it changes too rapidly to effectively be in any repos)
Rust toolchain is installed using rustup, not curl bash. It's rustup that's installed using curl bash. And while the site does recommend it, installing rustup alone securely is far easier than the entire toolchain.
> How is that helpful to the vast majority of the people on Mac/Linux trying to install Rust from scratch and reading the instructions on the website?
If you're concerned about running a remote script, just check how much work the script actually does. If it's not much, it may be worth exploring the alternative ways for it. For example, the rustup package in Arch Linux [1] does the same thing as what you get from curl bash.
I have mise installed - another package which recommends installation using curl bash. But I don't use it, because it's really easy to install it manually. And when some other tool recommends curl bash, I check if it's supported by mise. As it turns out, rustup can be installed using mise [2].
same parent had said "It's crazy that a language that prioritizes security so highly in it's design itself is only compiled through such insecure methods."
reply