I wonder if this is a flag that the AI bubble is primed and ready for something small to pop it. Curious if CDSs on other tech giants have similar behavior.

There is a level of truth to that, but a big part is also the way they treat the data. Many financial spreadsheets I have worked with over the years have few complex formulas and each step of a calculation is shown as a discrete cell. It makes it easy to verify data, even in printed form.

In the past I have used a few multi-million cell spreadsheets for engineering-- you get a csv file from a utility with 15-minute interval data for a few years and you start with over a half million cells, perform some basic operations on each row to make it usable, then a few columns for evaluating a scenario, and you hit a few million in under 20 minutes. (Then you export a pivot table to a new file to generate workable data and visualization, and the larger spreadsheet is likely to never be touched again.)

This is a poor application for programming or a database-- there is no institutional value to the information after first use, and while you could automate the 20 minutes of quick formulas, fill-down, and pivot the next project might get the information in a slightly different format.

Plenty of stories of DUI abuse in Blue states as well. One of the grant programs allowed "one and done" checkpoints, so one corrupt cop padded his retirement by doing DUI checkpoints, citing the first car through, closing the checkpoint (getting paid for 8 hours of OT), and reporting for an extra shift an hour later doing the same crap. He had over 40 days last year where he was paid for more than 26 hours!

My local PD purchased one a few years ago, but they are yet to actually put it to beneficial use.

Or if you want to do the work yourself, just look up the charity's form-990. It gives a lot of clarity to what the charity spends its money on. I was skeptical of an organization whose CEO was paid about $250k, but after looking through their financials it turned out to be a rather modest sum relative to the [program] assets they manage as well as their total assistance they provide.

Those are just different faces of the same coin. A gas turbine plant needs the physical equipment, the transmission capacity, a distribution network, gas pipelines, cooling water, land, and a whole host of other dependencies. When you design around one constraint the next one pops up.

You still need to get that solar power from "God's Country" to the Valley. SVP wasn't set up to own long distance transmission assets, and for private entities to build them to service their own needs has its own whole host of challenges.

So, you look at on-site generation options and ways to free up capacity from other users. General rooftop solar, storage batteries, etc work well to a point. Beyond that centralized generation is pretty much unavoidable.

Silicon Valley Power has always operated as a nonprofit, low-cost, major consumer focused utility. They have minimal generation resources and import the vast majority of their power. Their peak demand was ~500MW last time I checked. Adding 10% to that number with a single connection is a huge ask; traditionally customers that wanted a connection like that would be co-generating and only using the utility for load leveling and backup of critical loads. A new 70MW combined cycle plant has a lead time these days of around 5-8 years.

Which ultimately gets into why there is an opportunity for small modular reactors IMO. The customer power increments are just too big for conventional generation, but full scale reactors end up in the wrong place where transmission capacity starts to constrain a project.

I am very proactive with segregating IoT stuff (even sub divided into 4 different VLANs), but often there is no choice and no way to know until after purchase. I bought a cold plunge (Michael Phelps Chilly GOAT) that didn't have "smart" as a sales feature, but it turns out the heat pump can only have the temperature changed by unscrewing six screws on the side panel, or using the godawful TUYA app. This version of TUYA is cloud only.

I had a plan when buying though-- I could hook up to the modbus port on the control board and control it that way. Unfortunately that essentially requires complete disassembly of the unit's plumbing and removing the heat pump to get access to the control board. Warranty expires in a couple weeks so might go give it a shot then, but really diminishing returns. Biggest incentive to do it is that it keeps losing network access.

I avoid anything with TUYA, but lazy manufacturers keep putting this crap in.

For me it primarily comes down to rule complexity. I have 7 VLANs for my home, plus WAN. That works out to 168 paths that need to be evaluated with IPv6. For IPv4 I can eliminate about 75% of that with nat design as well as outliers which might otherwise force me to add additional VLANs. (Outliers like smart home devices that are generally blocked from external access but might need the ability to allow it for a firmware update.)

Then, once you have that many states you are tracking... it is easy to have something that unintentionally overrides something else.

That was my thinking as well-- what went wrong with IPv6-- too complex, focused on the wrong problems, or the firewall issues. For home I gave up on it before because my ISP din't give a subnettable allocation which made it not worth the hassle. I have changed ISPs though, might want to check again.

This particular plant is on the coast, between a refinery (with offshore tanker terminal) and a very large sewage treatment plant. It is a nearly ideal location for offshore wind turbines. Specific to LA DWP there is a need for a lot of excess capacity in this area because it is relatively landlocked for transmission capacity-- with Scattergood down it is hard to serve all the City of LA customers south and west of downtown.

Personally I think a small modular reactor at Scattergood makes more sense than burning hydrogen (rt efficiency of around 20%?).

If they fail it will knock $1.TB off NVIDIA's market cap, but unless they do it with a bunch of unpaid debts for capital projects it is hard for me to imagine them destroying the economy. If I was working on mega data center projects though I would tighten my payment terms for sure, as getting over extended there is what really kills a company.

The bubble burst will likely be more of a deflation IMO; as one company fails the remaining ones will have a brief period of opportunity and the cycle will repeat for a while. The fiscal circle jerk will take some time to unwind among the major (diversified) players, but it won't be the end of the world.

Very big assumption. I'm even skeptical with companies like Ubiquiti on their routers given how much responsibility they have contracted out.

Pit production at Los Alamos is about 1/year. A new facility should be ready by 2027 in Aiken, SC that can produce 30-50 pits per year.