American regional grids aren’t strongly connected, you aren’t getting much electricity between America west (eg Washington state) and the mountain west grid (eg utah). There is one big connection between a coal plant in Utah and LA, everything else is just connected by lines with very small capacity. If the west coast somehow tripped…we have better chance of getting help from BC than Wyoming or Utah.
Fair comment, my apologies if I conveyed that they were.
There's enough weak connection to accomadate some slosh that helps to smooth things and enough long connections to have potential issues in geomagnetic storms - although these should be well and truly mitigated by now.
I just wanted to point out that while Texas is completely disconnected, the regional grids are still only partially connected, but I guess this is only relevant for the west where areas of sparse population densities make strong connections very difficult. This is relevant when someone complains about EVs on the west coast using coal powered electricity rather than hydro and renewables that makes up much of the west coast's energy mix. Technically kind of true since Utah is connected via socal (they are changing this to a renewable link though), but not really since the other connections are pretty small.
I’m in the opposite boat, having trouble instructing my colleagues on how to get the same success with AI coding that I’ve realized. The issue is that you spend effort “working” the AI to get things done, but at the end of it your only artifact is a bunch of CLI commands executed and…how are you going to describe that?
AI instructions for AI coding really need to be their own code somehow, so programmers can more successfully share their experiences.
> but at the end of it your only artifact is a bunch of CLI commands executed
That sounds like a failure of process. Executing the commands is supposed to result in a Git commit history, and in principle nothing prevents logging the agent session. I'm told that some users even prompt the AI afterwards to summarize what happened in the session, write .md files to record "lessons learned", etc.
That isn't what I meant, you could save the entire CLI session, and not have something that can be shared easily. You need to document things like "try this or that, it still isn't very sharable.
Many people don’t get it, it’s really expensive, even in countries with non broken healthcare systems (not the us) costs increase rapidly and no one is sure how the systems will remain solvent with the same level of care given today. The way things are currently done are entrenched but not sustainable, that’s when disruptions are apt to appear.
When I look at the US, the symptom -> diagnosis hypothesis is not anywhere near the most expensive bit. If you have a medical issue and AI works effectively for this then it saves you maybe one trip to your GP. Your insurance probably still requires your GP to provide a referral to a specialist. If insurance companies allow for AI to be used in place of a referral then you save this trip. But you still need all of the stuff to confirm a diagnosis. And you still need all of the treatment.
If you don't have a medical issue and an AI system tells you this then you save yourself a trip to a specialist and the associated diagnostic tests. Again, this saves a bit of money but is nowhere near the bulk of medical expenses. And it has to be able to do this without any diagnostic testing, just based off of your reported symptoms.
Even if AI diagnosis works flawlessly we save a bit of money but absolutely do not revolutionize the cost of the industry.
I mean, if we're talking Christensenian disruption, that happens in neglected markets rather than currently dysfunctional ones. There's no shortage of actors wronging money out of health care so there's not a disruptable space per se.
A lot of economic activity tanks in the USA if China is treated too harshly. Manufacturing has actually gone down since Trump took office largely because of his tariffs.
We must make America great at competing with Vietnam again! A great leap backward! Lower the value of the dollar so the American worker will be motivated to do menial work greatly and cheaply.
There are vast differences in obesity rates between rural and urban areas, and red states vs blue states. Someone in San Francisco California is going to see a lot less obesity than someone in Plano Texas. I’m sure it is similar in Europe, with the caveat that Europeans are healthier than Americans in general.
Is removing the human in the loop really the goal, or is the goal right now to make the human a lot more productive? Because...those are both very different things.
I don't know what the goal for OpenAI or Anthropic really is.
But the context of this thread is the idea that the user daxfohl launched that these companies will, in the next few years, launch an "engineering team replacement" program; and then the user eru claimed that this is indeed more doable in programming than other domains because you can have specs and tests for programs in a way that you can't for, say, an animated movie.
OK, so you successfully argued that replacing the entire engineering team is hard. But you can perhaps still shrink it by 99%. To the point where a sole founder can do the remaining tech role part time.
I have no idea what will happen in a few years, maybe LLM tech will hit a wall and humans will continue to be needed in the loop. But today humans are definitely needed in the loop in some way.
You can dislike a solution but admit that you can't think of a better solution, or specify that it is better than an even worse solution.
I can see why someone would have a issues with "a bunch of rich bankers appointed by politicians" controlling American monetary policy. But I can't really see a better way at least, until we can achieve a post-scarcity economy or something.
Bernanke had a strictly academic career before going into public service (and was/is probably one of the foremost experts on the Great Depression, something that was handy in 2008/9):
I think people over-estimate how many "rich bankers" are in the Fed, especially at the FOMC.
Bloomberg's Odd Lots podcast with some Fed members in recent years, especially the more obscure regional ones, about their work, and how they often go out and talk to local businesses about what's happening 'on the ground'.
I’m not denying that, but that’s not how I read the comment. That comment comes across as a relief that the Fed is under attack, but is more upset that the source of attack is the executive branch.
The vikings landed there, not Denmark, who were Norse, Erik the Red was from Norway (But was considered by then an Icelander exile?). Before Danish control Greenland was a Norwegian colony, this was the colony that died out.
Norse colonisation tended to reflect their origin e.g. the Norwegians colonised the north west of Scotland and Iceland, which were more similar to Norway; Danes went to England and Normandy which were more southerly, flatter and more fertile, much like Denmark; the Swedes with their long Baltic coastline turned their attentions eastward.
Denmark got the North Atlantic islands through the union with Norway, and retained them after Norway became independent.
I know, but that was much later and had a very different dynamic, due to climatic changes etc.
The earlier Norse colonisation of Greenland seemed to consist of farmers and independent settlers, mostly via Iceland. In some areas, they never interacted with Inuit, or rarely.
The later effort seems more focussed on Christian missions to the natives, and commercial whaling and sealing.
reply