Hacker News Reader

Comment by xchaotic | Hacker News Reader

For such generic statement of the problem, it's certainly unsolvable. But with encryption, timeouts and defaults it can be - as seen by many the distributed systems on the Internet

View on HN

> with encryption.

That has been accounted for.

" [Eve] does not, however, get to open the enveloped message."

> timeouts and defaults it can be

These time-outs and defaults would have to be negotiated between the generals over the postal system and wouldn't help them.

> as seen by many the distributed systems on the Internet

Those would have implemented a form of fault-tolerance for the more general "Byzantine Generals Problem."

Timeouts and defaults can be considered as part of the algorithm. The problem is that you may be unable to determine a reasonable timeout beforehand. The timeout as a practical solution is also described under 'engineering approaches': https://en.wikipedia.org/wiki/Two_Generals'_Problem#Engineer...

I cannot reconcile your claim with my understanding of the formal proof. We are asking for a guaranteed consensus, perhaps you are merely asking for arbitrarily high probability of consensus.

Can you point me at an algorithm that guarantees consensus? Your claim seems to be at odds with the formal proof in the article. I particular, the article already allows for encryption, timeouts, and defaults.

So I can't see how you are right, and would appreciate some additional information as to why you think the article is wrong.

The whole point of the problem is that you're not allowed to use timeouts and stuff to get a work around. Adding timeouts etc fundamentally changes the situation. The distributed systems on the Internet you see do not solve the problem - they work around it, basically by adding failure modes not present in the original.

I think I'm being more pragmatic and immediately jumping to what needs to change to make this situation treatable. An abstract statement of a problem like this is probably great for teaching people concepts about consensus, but hopefully no one knowingly designs stuff that way (except Mongo perhaps)

If you don't understand what's impossible, you run the risk of trying for something you'll never be able to accomplish. You'll never know why you don't succeed. It's precisely about teaching and learning the limits beyond which it's pointless to even try to go.

If it's pragmatic to dismiss impossibility results, perhaps revisit pragmatism.

Reply on news.ycombinator.com