beembeem

> 2,400 queries per order? That's not that crazy IMHO.

Isn't that off by at least an order of magnitude though? It forces them to operate a much larger cluster than should be necessary.

> my food still arrives cold af

Ha.

The first thing I noticed and you almost got to it in your summary: at 1.2MM/2300 = 520 qps per node, this isn't a wild setup. I'm wrapping my head around how they're generating that amount of load. Seems like an easy task for any database to handle.

As someone who was in the trenches on this - what were/are your thoughts on putting all your financial passwords in one place?

Zero Interest Rate Policy

Roofs should do more than just capture the energy of the sun, it needs to capture and store the water that hits it, because diffusing that water is enormously expensive at a city level. Hint: shingle roofs inject things into otherwise fresh water.

This is the case across much of the US. Seattle dumped 100 million gallons of untreated storm & wastewater in 2022.

This was prior to the introduction of transactions years ago. The versions stated in your comment are past EOL and not offered in Atlas today. Not relevant to today's reality.

Yes, you cherry-picked an under-regulated market with cheap land. Not to mention that cheap land is usually further from metro areas and thus requires more distribution costs. Either way, as I mentioned land acquisition cost does not fully encapsulate the environmental costs of covering large tracts of natural land with panels.

> Which costs are they able to externalize by building grid scale solar

The biggest one is transmission. Most utilities have an arrangement where they can receive a certain return from customers with new capex. No large-scale transmission is required for rooftop installations.

Thanks for the book recommendation. I'll have to check it out.

> "customer incentives"

Correct. Customers are incentivized to use as little as possible and therefore save money. The utilities are incentivized to get customers to pay for new projects to earn a guaranteed return. And if a customer overgenerates their credits in a net metering scenario? Typically this is free energy to the utility. No payout.

It is because utilities can externalize many costs with their current development practices. Every now and then the federal government jumps in with some more funding for them too (see: Diablo Canyon). Incentives aren't aligned between the utilities and their customers.

Thanks for the link.

Two issues I see missing from NREL's analysis: land use and distribution.

I don't see a cost attributed to land in there. It's free for residential because it's already put to use, but not so for utility-scale and probably hard to estimate broadly. Environmental impacts of that land use should also be accounted for in addition to direct land acquisition cost and/or leasing. Solar generation does not have land impacts when placed on a roof. That land is already "disrupted" and therefore allocated for human use.

Either way, labor is now the dominant piece of the cost for residential, and it's obvious that one-off small jobs in high-price metros are more expensive than a crew operating in rural areas. An accounting looking strictly and materials + labor is going to heavily favor utilities, but it doesn't capture the full picture.

A few comments:

Your comment assumes that the utility needs to be using up lots of land to power customers. Rooftop solar is far more efficient and there is no extra transmission required. Utilities don't need to consume vast tracts of land just to provide power to customers.

> retail price

Utilities in California (by law) already deduct fees out of this, so it's not reimbursed at full retail rate anyway under old NEM rules. There's also an absolutely massive gap between retail and wholesale rates that could be explored.

I would appreciate data to the contrary, but the accounting in your first claim doesn't seem right to me. Behind the meter rooftop solar should be the most cost-effective because no new distribution is required nor is any land use change required.

It would actually cost them a significant amount in lost revenue. It's not merely disrespect of users, its money vs user goodwill.

Correct, merely a money issue. Incentives are in the wrong place here.

Aha, to clarify for others - Vancouver WA is what you're referring to, not BC.

The problem you didn't touch on specifically is that it still makes sense to separate sewage from runoff, which is not cheap and becoming increasingly common in major US cities. Untreated sewage mixed with storm runoff makes for far worse water quality conditions than merely runoff.

I went down this rabbit hole on National Beer [1] day this year.

[1] https://en.wikipedia.org/wiki/National_Beer_Day_(United_Stat...

Bingo. Corrosion. Everything would need to be "marine-grade" aka expensive to plumb :)

That's really cool! Thanks for pointing this out.

I'm definitely going to try this, but oy javascript. It's probably the reason this can be written with so little code but I would prefer a native binary that doesn't have the npm dependency.

Thanks for linking those sources. I think this is the key part of the article:

> increased demand over the coming years, including data centres, cryptocurrency miners and greenhouses.

Price them out. Don't give sweetheart deals to uses that aren't beneficial to the local economy. Additionally, these uses don't necessitate baseload. Crypto should be using cheap, renewable variable power sources. I'm genuinely curious what greenhouses means in this case. Heating indoor crops in the winter? Geothermal would be more effective for this. [1]

[1] https://www.youtube.com/watch?v=ZD_3_gsgsnk

> how many years a installation in a non-geologically/thermally active area can run before it changes the temperature of its lithic heat sink

I would imagine this varies, but is a much smaller problem than e.g. heating up water habitats (data center cooling) or burning fossil fuels.

I present you the most amazing solution to this: https://www.youtube.com/watch?v=2z3ZyGlqUkA

Not to mention that the dams in Washington state do account for spawning/migration and slow down flows at certain points of the year.

> Right now there's no sustainable source for baseload power.

To push back on this, there is a centurys' old technology for this in hydro. There's also an emerging field of geothermal.

The Pacific Northwest (both sides of the border) has an abundance of hydropower that was installed decades ago and provides clean, reliable, baseload power.

Geothermal has its challenges, especially at utility scale where geographic location is important and not typically close to population centers. On a smaller scale it is not geographically a problem, but not as economically viable for individual homeowners. It makes more sense when a large corporate is investing in it [1].

[1] https://news.microsoft.com/2022/04/20/microsoft-redmond-camp...

Oh this is what you meant - making a living through permaculture. Can confirm that SelfSufficientMe strikes me as a great dude with engaging content.

At first I assumed this was the one near Belize, which is quite well known in the scuba diving community [1]. In fact, the picture on this post is probably of that one. However, this new one is over twice as deep!

[1] https://en.wikipedia.org/wiki/Great_Blue_Hole

It's a replicaset at least, correct? Hopefully not standalone

I don't claim to know too much about Arianespace, but I wonder if this has any connection to the Russia situation and not being able to launch Soyuz rockets due to sanctions.

https://spacenews.com/arianespace-and-oneweb-reach-settlemen...

yep, would be great to have this ultimate list but for Emacs *ducks for cover*