the sun shines and the wind blows in the winter. Plus, batteries. Giant redox flow batteries are coming online now, sodium batteries, it's not like there aren't options for storage people are working on.
not sure if you meant to link a different chart - that chart is the electricity component of CPI-U (an inflation measure) meaning it's by definition not inflation adjusted - you'd probably want to find a chart of nominal values for electricity spending and adjust for inflation to see whether or not it's at an all time high. Which, recently it's run ahead of inflation because of the overall surge in demand and investment in upgrading aging infrastructure/adding new capacity, but long term it's pretty flat/down
It's cheaper and they're all subsidized. Solar is subsidized, oil and gas is subsidized, nuclear is subsidized. Fracking recieved significant subsidies to develop the market initially. Right now, solar/wind is cheaper than fossil fuels by a lot, which is why it's the largest percentage of newly added energy capacity worldwide. Cheaper, faster, and after installation, the power plant doesn't need to keep getting fed with a consumable (incredibly low variable costs are causing issues where markets are getting messed up and going into 0/negative prices since power plants are competing to dump energy it doesn't cost them anything to generate)
Electrification (converting things that ran on fossil fuels to electricity) is a net decrease in energy demand but massive increase in electricity demand if you're looking at electricity specifically. Past few years have also had a ton of growth in data centers and manufacturing investments. According to the EIA, they're projecting a decrease in energy demand over the next few years: https://www.eia.gov/todayinenergy/detail.php?id=65004
There's the anthropic paper someone else linked, but also it's pretty interesting to see the question framed as trying to understand what causes the hallucinations lol. It's a (very fancy) next word predictor - it's kind of amazing that it doesn't hallucinate! Like that paper showed that there were circuits that functionally actually do things resembling arithmetic and computation with lookup tables instead of just blindly 'guessing' a random number when asked what an arithmetic expression equals and that seems like the much more extraordinary thing that we want to figure out the cause of!
Spectroscopy is cool! Basically whenever light shines through a thing, the thing can absorb or emit light, and that happens at specific wavelengths depending on what the thing is made of. You can get small spectroscope pretty cheap and play around with a bunch of hone experiments - a common lab project in class is putting salts in a flame and based on the spectra figuring out what atoms are in the salt. It's how we discovered helium must exist in the sun before actually finding any on earth. So basically we just do that, except on a planet that's far away. When it's between it's star and us, we look at the spectra of the light that goes through the planets atmosphere
I think the point is that it's funny that the standard library has to use unsafe to implement a data structure that's like the second data structure you learn in an intro to CS class
I'm pretty sure that's replying directly to the comment about how c is close to assembly and that if you add that line of code somewhere you know there's a variable getting incremented. Doesn't really matter whether or not it's useful, the point is that the behavior isn't exactly what you wrote
I'm pretty sure a monthly payment would have been wildly wildly more profitable than ads lol. $3-10 per thousand impressions according to a quick search, versus even a couple dollars a month for 365 impressions over a year? The tone of that post sounded like the creator was also rather unhappy with the ad situation
Not sure how it is in all fields but at least in the molecular bio world there's a lot more replication than meets the eye. Especially methods, it's just that there won't be a replication paper, it'll be in the methods/supplementary info of the new research paper ("we used xyz techniques as described in researcher et al").
Null publishing bias still exists and obviously it's not remotely a perfect system. However, it's not some sort of completely broken system where every paper is either fraud or replication, and interventions that start from that sort of an assumption (people are very aware of the need to trust but verify) are in my opinion unlikely to be useful.
If the world is at a 60/100, you need to focus on things that get you from 60 to 80, not from 0 to 20
No, to the team. Science is collaborative. If you learn that the results you are working with and building on are fraudulent, what's even the point? Trying to replicate that team members results with it never workong, starting new projects designed based off of fake data, continuing to tolerate and endorse that fraud by keeping quiet means the entire group's work becomes meaningless. Fraud infects all the research that references it and builds on it. Would you want to spend 5 years of your life futilely trying to build on your collaborators work if you know at the start that their data is fake?
My uneducated guess would be shared power infrastructure, easier licensing and permitting for one site, one large safety/fire extinguishment system instead of many, etc. The article mentions only 2 transformers needed to link that entire station to the grid. Also the controls, managing power delivery from thousands of small batteries versus one centralized facility. And if that particular wind or solar project doesn't go well for some reason the battery isn't tied to it, it's stabilizing the overall grid. Not sure though it is a good question, probably some technical or regulatory docs that go into details somewhere, I'd imagine this would have a ton of bureaucracy before getting built
Makes sense but the problem is that you're using words that already have specific meanings in the space, all related to creating one model with multiple functionalities. Merging meaning merging models into one model. Multi modal meaning one llm that handles multiple modes. The term you want is probably agent or framework or chain or something. Basically, what you describe is when it feels like you're only working with one model. What your title says is when you engineer specifically actually only one model, which is a distinct technical challenge.
If you care about putting up a fake identity this is still bad. Social media is all about being distinct and grabbing attention. Getting samified into a bland featureless identity isn't the same as as carefully crafting a persona to maximize clicks
If you want to pick an era of technological progress to make that point maybe don't pick the one that involves America becoming a superpower by putting a bomb invented by Jewish refugees on a rocket build by ex Nazi scientists after a physics revolution where be basically got to go and take all of Germany's top talent lol
That's a general rant against publically traded companies. I'm specifically asking about the original posters take that a company that makes speakers specifically shouldn't be publicly traded. Like it sounded like they thought a speaker company couldn't be big enough to be on NASDAQ or something?
At that point the enormously powerful central players like big tech and militaries and tax collectors will be more than incentivezed to use every remaining resource they have to bring everything back online and re-centralize power. And if they can't - society and power wasn't exactly more distributed in the distant past lol. Your local warlord/military dictator/whatever will probably not be supportive of nerds acting independently.
That proprietary format is owned by a company associated with folks who won two nobel prizes for AI related work this year and the employer at the time of the researchers who wrote the attention is all you need paper and also the owner of a search engine with access to like, all the data. Doesn't seem unreasonable lol
> where
Talk to current students/forum/visit etc. If you're gonna spend 3-6 years of your life there it's worth putting in the legwork. And figuring out the amount taught by adjuncts etc - you can just ask.
and on the percentage of classes under 20 - that's the exact metric I'm talking about being easy to game. You'd want to figure out what the sizes of the classes you actually take are.
If you really value faculty involvement etc, I might suggest focusing on finding out those statistics directly and getting a sense of what current student experiences are like. That's as opposed to relying on a proxy like student/faculty ratios, low rankings or being a state school. Each state runs its own state schools (and some cities like NY have their own) and have wildly different levels of funding and support for education. And being public does not make them immune from going the adjunct all the way route.
Student faculty ratios can be reported using "full time equivalent" faculty. Meaning, your class of less than 20 students might not have a more available professor, because they're actually also teaching at 2 other universities in the same system to make ends meet. It's not the ratio to the number of tenured faculty, adjuncts count.
For class size, you can have dozens of small classes nobody takes, and have giant lectures for the core intro classes everyone takes. Then the average class size of the classes offered is quite small, but the average class size of the classes you take is quite large.
The context of that comment is a parent comment asking about medical self experimentation. You're also not allowed to administer scheduled substances to yourself in many jurisdictions and plenty of other things like that but that's not what I was replying to.
It's already an importany part of diagnostics and treatment! A lot of therapies are approved not just for specific cancers but for specific cancers with specific mutations. As with everything cancer it turned out to be more complicated than expected (not surprising given that it's essentially an outcome from chaotic uncontrolled genetic mutation and evolution) but sequencing progressing at like an order of magnitude faster than Moore's law and better understanding of all the -omics is making a dent
There are a couple OVTs that are approved by the FDA and it's a rather active area of research. One of many in the recent explosion of biological cancer treatments (immunitherapies/cell therapies, etc etc). With any cancer treatment the exact same procedure will struggle to generalize well across cancer types and you have to regulate for safety and efficacy rather strongly. Especially when the treatment is lab grown measles or herpes virus strains that are genetically engineered.
The former is a reason why there actually is a lot of research into this already/couple drugs approved (cancer is usually very profitable), the latter is why all drugs, not just these, need extensive clinical trials and strong regulation. Because it's measles virus and herpes virus etc etc, and often genetically engineered, so the safety risks are ... obvious
You can do whatever you want to yourself. You can't hurt/encourage others, and the data may be no good, but certainly you have informed consent if you're the only subject