Comment by rickydroll
Comment by rickydroll 6 days ago
1) Revisiting one of my past projects, using Hashcash as a rate limiter. The original one I worked with was Camram/2 Penny Blue, which embedded proof-of-work tokens in an email message as an anti-spam method. It was the wrong solution, and proof-of-work tokens should be embedded in the protocol itself, which would enable dynamic pricing based on reputation.
Why I'm not doing it: I'm not sure it would be accepted. I took a lot of lumps for trying to use Hashcash in email, and I'm not sure I want to go through that again, but it does have value. Embedding proof-of-work puzzles in a protocol is a great way to limit abusive requests and patterns. SMTP, HTTP, HTTPS are easy to modify and probably could be done via a proxy. I'm not sure how easy it is to change the SSH protocol, but that would be useful as well.
2) Low-income living space electrification.
I tested this idea out on a friend who works for a Housing Authority, and their eyes lit up. However, they warned me that it would take a few years for everyone to sign off on it.
The original idea was to provide a kit, a bag of parts, that an affordable housing authority could use to improve living quarters and housing for low-income people, and eliminate/reduce the use of gas.
a) Replacing gas stoves with a set of three induction plates. The significant challenges are filling the void created by the original stove, ensuring sufficient power to operate the induction plates, and addressing how to handle the absence of an oven.
b) Filling the hole is easy. This is something a halfway decent carpenter could do, or we could provide an adjustable-size box that fits in such a space, not quite an IKEA flat-pack but roughly similar.
c) Power is a little more difficult. One company is solving this problem by putting in a battery to handle the load. This is possible, but the baseline cost would now be approximately $2,000, just for the parts.
d) window mount heat pump. New York City has funded in-window heat pumps as part of a design project. The problem is they run around $4,000 to $6,000, but an ordinary handyman could install them.
3) Recycling car batteries from crashed vehicles into home power banks.
This project is a bit of a stretch for me. I know people are doing this, but not in the States as far as I can tell. The off-grid solar community has a variety of inverters and solar chargers that may be suitable for this kind of situation, but I don't have enough knowledge.
4) Ad hoc virtual power plants
Many people have rooftop solar. The grid gets overfull on bright sunny days. People who can't have solar often have space for batteries. Work out the instrumentation and accounting so that solar producers can charge batteries, and everybody gets compensated when the grid demands the battery's power.
It seems to me that this would be a great application of distributed system concepts, providing a win for the local community and grid resiliency.
I need a second life to make progress on these ideas
Interestingly, just yesterday I found out 4 already exists, but it's inverse of what you're thinking: top up excess battery capacity overnight, when grid prices are low, and then resell during the day when purchase prices are high.
I think the economic problem is that while there /may/ be overproduction during the day, the day is the only time other than early-mid evening when there is significant demand.
You would effectively be targeting the early-mid evening demand, assuming there was overproduction during the day, and with the current cost of batteries, their installation, and their replacement, I can't see the numbers working out.
I would love for battery prices to come down enough to make something like this possible though.