pancake

It seems to have plateaued and increasing more slowly. Combining data from Steam and Statcounter reveals this:

This time I've attempted to directly quantify the cost of living to income ratio in different countries. As the note states, this doesn't factor in anything that the state might provide for free, or that people in those countries might typically self-produce. So it's not a reliable indicator for very poor countries or those with extensive government programs covering basic needs. However, it does give an idea of "how much better off" people might be in a country versus another, at least for the higher and middle income ones.

Source: IEA

After some investigation and benchmarking, it looks like the best PIR protocol for this use case is YPIR+SP (from February). On a single compute- and network-constrained server, with users on constrained (and possibly metered) networks, this would amount to providing service to up to 1000 users while keeping latencies reasonable; by (quadratically) scaling the server(s) enough, that could become up to 100,000. That means this method of message routing could definitely work, although I look every day in case new protocols are published.

Edit: tl;dr: all tools seem to be capable of 28 nm process, except the lithography machines. However, the report (why?) doesn't include the most advanced lithography tools that are known to be manufactured.

Source

Today they published a new report in the series. To my surprise, it includes a second chart with revenue-based data. Enjoy!

I have been thinking about implementing this for quite some time, but I would like some feedback from people more knowledgeable than me on the matter.

There's been some great progress in the field of Private Information Retrieval (PIR) protocols. Recently, in a 2022 article, Lin et al. describe an "updateable DEPIR", with both read and write times that can be made sublinear to database size.

I wonder if one couldn't use a combination of this technique and regular public-key cryptography to provide fully anonymous message routing. One could write outgoing messages to a fixed address and issue private reads to their contacts' addresses, with the messages themselves being encrypted with the receiver's public key.

The benefit of this would be a messaging protocol wherein the server wouldn't just be oblivious to the content of all messages, but also the social graph itself, plus all message-sending operations becoming deniable as a side effect.

First of all, I'm sorry if my question could be easily answered by finding the right source. Overall I'd say I read very little theory written by contemporary comrades, and that's something I need to fix once I have the time.