Exploring Time and Order in Distributed Systems

voidpapers.substack.com

49 points·flyingsky··2 comments

It's a bit of a weak sauce when it comes to reviewing a paper. I mean it's nice to be made aware of a paper you might care about, but it doesn't take the problem head on with solutions - maybe the paper didn't go that far?

Even the summary isn't a summary.

This is a subject that I found particularly interesting because I explored it a lot when I tried to make a new PoW (proof of work) system that wouldn't burn the planet down; which I named proof of time, in which I explored the idea of proving that time had passed, via server latency rather than computer riddles...

Sadly I didn't go further than theoretical aspect, because at this point, anyone writing a "whitepaper" related to crypto falls into the bag of unfunny clown jokes.

You might not be aware that this is one of, if not the, seminal paper in distributed systems.

Lamport clocks, described briefly in the review, are an answer to the general problem of establishing an order between events that happen on different machines that respects causality. They’re still practical to this day, over 50 years after this paper was published.

There are lots of in-depth descriptions of the paper and its contributions elsewhere. I’d totally recommend finding them - it’s a great paper and not a hard read at all.

Reply on news.ycombinator.com