Musk probably cancelled their hosting contract.

I just moderated you insightful, but then realized it was Tom Cotton, so redundant would have been more appropriate.

Sorry my moderation will go away because of this comment ;)

It is amazing, but your numbers are off.

50% is low. Perhaps really low.

The coders are some volunteers, and some companies that volunteer their employees. Still a community.

Some of the edge cases can get a bit scary. I tested a piece of iSCSI/ROCE code for a developer because he did not have the hardware to test it himself. It was a simple case, and the code worked, but the idea of a developer without a test-bed is at least "interesting".

So, still amazing, but like most things, more complicated.

The least disruptive is to just ditch vi and keep the current code. VS Code remove over ssh might be a good option. Unfortunately, the choice of editors is often more of a religious discussion that a technical one.

My math on this is 400 MW of solar, but with the typical over the day spread. The batteries can either charge or discharge up to 400 MW, but can hold 1600 MWh.

In that the panels and batteries are both at the same site, losses should be reasonable (and might already be built into the number). This balance should let the "combination" run pretty much one deep cycle per day most days:

In KWh:

1,600,000 x 300 x 10 = 4,800,000,000 or 4.8 billion KWh, which works out to $0.21/KWh.

Add in some incentives, plus the possibility to expand on the same site, and it makes a lot of sense, especially in that the power it sells is "peaker" power, not baseline like nuclear.

otherwise known as a "Hysenberg Compensator".

... looks like it was unanimous (imagine that). One commissioner was listed as "approving in part" and "concurring in part", so perhaps a small disagreement in the wording. Regardless, a "day of days".

I am not sure about this particular issue, but the "app" you speak of could be a bit of JavaScript on a web page you are visiting.

And the "encrypting data" could be your SSL handshake with your bank.

They are paying the premium because a) it is not that much in terms of the entire cost, and b) if some bad actor got hold of the satellite, it might give them the info they needed to take the entire constellation down.

The base A100 has an MSRP of $27,617. You can rent one in the cloud for a couple of dollars and hour, so perhaps the MSRP is a lot more than the big guys are paying.

more like a bug in the distro install, not in Linux at all. This looks to happen up-stream of the kernel. And it is "network boot" only. I suspect more than a few cloud providers care.

Whether a windows EUFI boot is impacted is probably more a matter of luck that design. Then again, how many "metal" servers in data centers network boot to windows anymore. If you really need windows in the cloud, it is likely in a virtual.

crypto miners do the opposite. They sit on the Texas grid and expect to get paid for not using power during shortages. The ultimate in an incentive system that has not been thought thru.

Based on some quick searches, if all cars were BEV, then 27% of the grid would be needed to charge them. In that less than 1% of registered vehicles are BEVs, the BEV usage should be somewhere around 0.3% of the grid.

I actually think that crypto mining is fine. Just restrict it's energy to times where the grid is in "curtailment". I am happy to heat up their GPUs with power that would otherwise be thrown away. Otherwise, their power use increases the electricity cost for everyone else.

The "stupid quick" Google search implies 40% to 80% efficiency for a rectenna, and the duty cycle is still near 100%, so 10% vs 3.75%. Regardless, the end to end cost is probably still way way too high.

25% of the energy density of sunlight. Solar panels are 15% efficient, and only get effective use 25% of the day. So 25% vs 3.75% (.15x.25). I am still skeptical of the costs though.