arbuge

Lots of comments here suggesting forwarding emails from the external account to Gmail as a workaround for this.

This isn't a complete workaround though. In particular, the option to delete the email from your server after retrieving it will not be replaceable. At least cPanel doesn't seem to offer the option to delete automatically after forwarding, and you could argue it shouldn't - with a push, you never know if the other end actually got it, unlike with a pull.

She

Are they allowed to rent them from server farms, datacenters, etc. located outside China that are able to procure them?

Re: the tax credit, note the income requirements here:

https://www.irs.gov/credits-deductions/used-clean-vehicle-cr...

I don't know anything about your situation, but I'm guessing quite a bit of the readers here on HN from the US won't qualify.

This kind of thing makes we wonder if the internet as a whole is heading for a Kessler Syndrome kind of situation.

Another example: here in the USA we have 50 states vying to each regulate AI; my understanding is that the plan was to have the OBBB put them off from doing this for at least 10 years, but that effort failed, leaving them free to each do their own thing.

Complying with all rules and regulations in all jurisdictions to which a website/service could be exposed (i.e. worldwide, by default) seems like it's becoming a well nigh impossible task these days.

I could see that happening... perhaps instead of plugging in the URL of the MCP server you'd like to use, you'd just put in the URL of their online documentation and trust your AI assistant of choice to go through all of it.

> Ice cream was important for morale. In 1942, the U.S.S. Lexington was slowly sinking from a Japanese torpedo attack, before the crew abandoned ship, they broke into the freezer and ate all the ice cream. Survivors describe scooping ice cream into their helmets and licking them clean before lowering themselves into the Pacific.

TIL...

"describe your resources for us to exploit them"

What you want to do is offer resources that make you money when they're "exploited".

This could be true... but only if there is an actual specific problem that they can put their finger on that requires the device to be AI-first. What is that problem exactly?

It's also not obvious to me that a concerted effort by Apple (unlike what we've seen so far, admittedly) wouldn't eventually be successful in converting the iPhone to something effectively indistinguishable from a platform designed from the ground up to be "AI-first".

Designing things from the ground up is hard by the way. It's not just the design itself; it's the ecosystems around them which are really hard to get going. Apple has the world's biggest flywheel in motion there already.

Those are the class A shares. Class B stock is below $600.

Well you do get admission to the Berkshire Hathaway annual meeting if you buy just a single share (and prepare your wallet for exorbitant Omaha hotel prices that weekend). Closest thing to that I can think of.

Correct.

I'm from Malta. It's dead center in the Mediterranean sea, just 60 miles off the coast of Sicily, and always has been a melting point of cultures from around the Mediterranean as a result. This is literally the first time in my life I've heard it described as remote.

Correct. In Europe and Mexico, South America etc. this is also the building pattern. The walls of the house go up to the sidewalk and all the yard space (if any) is at the back. And maybe a courtyard, as you said.

> language specific

Framework-specific in this case... beyond just language-specific.

> What, then, is an LLM vendor’s moat? Brand? Inertia? A better set of applications built on top of their core models? An ever-growing bonfire of cash that keeps its models a nose ahead of a hundred competitors?

Missed one I think... the expertise accumulated in building the prior generation models, that are not themselves that useful anymore.

Yes, it's true that will be lost if everybody leaves, a point he briefly mentions in the article. But presumably AWS would also be in trouble, sooner or later, if everybody who knows how things work left. Retaining at least some good employees is tablestakes for any successful company long-term.

Brand and inertia also don't quite capture the customer lock-in that happens with these models. It's not just that you have to rewrite the code to interface with a competitor's LLM; it's that that LLM might now behave very differently than the one you were using earlier, and give you unexpected (and undesirable) results.

It's good research but whatever you think about the value of the content on those sites, I think the word "fake" in the headline is misleading clickbait.

These sites are officially linked to from the parents (CNN and USA Today in this case), with whom they no doubt have a revenue-sharing agreement. They're very real in all senses of the word. The third party in question (Forbes marketplace) is not trying to fraudulently set up some parallel CNN and USA Today websites without their authorization.

For comparison that's almost 10 times the distance to the Andromeda galaxy (2.5m ly).

Better than shutting down entirely I think.

> The one area we have had some promising success is in advertising on the site. But the effort it takes to find advertisers and manage them, and my general desire not to flood Exercism with adverts, has meant that I feel this isn't a very sustainable strategy.

To me that begs the obvious question... what about trying some automated advertising like Google Ads? Probably not the most theoretically optimum use of their ad space, but minimal effort is involved.

Not sure about this one... according to this study the escape velocity of the Milky Way is around 550 km/s.

https://ui.adsabs.harvard.edu/abs/2021A%26A...649A.136K/abst...

The authors of that study cited above are pretty confident it's above 497 +/- 8 km/s, at least in the solar neighborhood.

1M Miles per hour works out to 446 km/s. Granted these are all rough numbers, and it depends exactly where you are in the Milky Way, but it looks like it may be a bit short of actually escaping.

From the article:

The researchers found that the rubbing of metal wheels and brakes on the rails releases metal particles, which appear in particular concentrations as the trains arrive at a station.

“We noticed that when a train arrives at the station, the concentration in the air shoots up, and about 15-20 seconds after it leaves, the concentration slowly comes down,” said Ghandehari. “This implies that it is the stuff that’s at the bottom of the tunnel that gets churned up and deteriorates the air.”

And they own Youtube...

That will be a massive benefit when it comes to training AI-powered robots I think. Conceivably, an AI that's seen every plumbing-related video on Youtube would be superior to any human plumber you can imagine, assuming a good mapping between the human hand movements in the video and its own robotic arms. And so on for other domains.

"Despite their large multi-seat public latrines with washing facilities, sewer systems, sanitation legislation, fountains and piped drinking water from aqueducts, we see the widespread presence of whipworm (Trichuris trichiura), roundworm (Ascaris lumbricoides) and Entamoeba histolytica that causes dysentery. This would suggest that the public sanitation measures were insufficient to protect the population from parasites spread by fecal contamination. Ectoparasites such as fleas, head lice, body lice, pubic lice and bed bugs were also present, and delousing combs have been found. The evidence fails to demonstrate that the Roman culture of regular bathing in the public baths reduced the prevalence of these parasites."

I would think that the communal nature of the latrines and baths would actually have contributed to increasing the incidence of those parasites, rather than decreasing them.

He appears to have a bit of unfortunate history with this kind of incident:

https://x.com/ArtemR/status/1814349423046209701

This is a problem in Texas also, although my understanding is that they're working on it: https://www.tdcj.texas.gov/ac/index.html

> Maybe you just slept badly. Coffee fixes all of those.

It doesn't fix sleeping badly for me anymore, at all. Perhaps it used to when I was younger, but after turning around 45, I increasingly found that the only fix for a poor night's sleep is a good night's sleep.

or perhaps poor nights of sleep just didn't affect me much back then.

Once the training is complete, one thing I didn't see mentioned in the paper was how they maintain the charge on the gate capacitors, which is analogous to the weights in a traditional neural network if I'm understanding this correctly. Any practical implementation will need to have some practical way to refresh that on a continuous basis so that the weights don't drift. Was this perhaps mentioned somewhere and I missed it?