emin_gun_sirer

Very interesting autopsy. But there are many uses for a virtual notary besides those times when "sh#t hits the fan."

Open source distributions, for instance, are currently distributed along with a hash, supposedly for protection, but we all know that those hashes are essentially worthless. This is a way to get integrity protection for free.

Employment status is another case where an independent certificate that you were employed at that instant would be valuable for the recipient.

Raffles and so forth are entirely untrustworthy at the moment; participants need to trust the reputation of the raffle holder, and while there are some laws, they are just not enforced.

In all these cases, VN can provide a useful service far before anything hits the fan. CertTime looks like it was a great attempt at a very specific instance of the problem. Virtual Notary is attempting to tackle a more general version of the attestation problem, so I'm hoping that there are more savvy users to draw on.

We've been running the service for many months now, and it costs 0.005 BTC per day, which is a small price to pay for a useful service. Just curious about what caused you to shut the service down.

Yes, quite possibly, if there is sufficient interest.

Great idea. We'll add this to the to-do list.

A git factoid plugin is a great idea. What exactly is the attestation that you have in mind? You provide the name of a repo plus a tag, and the VN issues a certificate of the hash?

It's an API-compatible drop in replacement for LevelDB, so anyone else who currently uses LevelDB should be able to switch fairly easily.

This article would have been stronger if it had contained a bit of introspection. Why did it require a UK paper to uncover this operation?

I can answer the opposite question: we intend to keep in sync with upstream. Since HyperLevelDB is a drop-in replacement for LevelDB, it should be fairly easy for LevelDB users to pick it up.

Because each coin transaction entails a 40KB zero knowledge proof. So a Satoshi will require 40*10^8 KB transferred for a 1BTC transaction.

Technically, this is very cool work. But one thing the paper overlooks is divisibility. How does one make change with zerocoin? It appears that the trapdoors allow only a whole coin to be spent, with no recourse for spending a partial coin. Needless to say, non-divisibility will make a practical deployment difficult.

Don't forget that the short squeeze also feeds the runaway price of BTC and props it even higher.

For the record: not Oakleys, not at a beach, but yes, looking to the side.

If we're done picking on looks and personal issues, something I tried to avoid very carefully in the article, perhaps we can focus on content and ideas?