[Cryptography] Best Internet crypto clock ?

Henry Baker hbaker1 at pipeline.com
Wed Oct 1 08:45:24 EDT 2014

[I'm quite new to this mailing list, so I hope that the following question isn't embarrassingly trivial.]

In old B/W movies, when a person was kidnapped, the kidnapper sent a photo of the person together with a picture of the front page of today's newspaper to prove that he had the kidnapped person _on or after the date_ of the newspaper.

In this case, the newspaper headlines for that date are unknowable in advance, so the recipient of the photo can establish an earliest date bound for the photo.

In today's Internet world, one could presumably do the same thing with a crypto hash of the current contents of the NYTimes, but this is now quite difficult to check because the "front page of the NYTimes" is no longer very constant, various versions are served up to different viewers ("A/B testing"), and I doubt that anyone is keeping track of exactly what pages are being served up at exactly what times.

One could also hash the closing bid/ask prices for N stocks on the NYSE; since this information is kept for long periods of time, it could be far more reliable.  Unfortunately, these closing prices are available only once per day, 5 days a week.

Another possibility would be to capture a hash of a snapshot of the Bitcoin blockchain at a particular time.  However, I don't know how easily one can search backwards in the Bitcoin blockchain to check for when a particular crypto clock value occurred.

It's clear that any crypto clock would have to be a simple append-only, read-only database whose future values cannot be predicted.  It would be nice to be able to quickly search such a database, but since most searches would be querying about relatively recent events, even backwards linear searching wouldn't be too bad.

Since we are free to choose the format of our authenticated crypto clock value, we can easily include an index of what wall clock time it thinks it is, so a simple RAM access to the database will be able to check the value.  Assuming authenticated values, we can also trivially compare two such values to determine if one "time" precedes another "time".

I would imagine that the best agency to publish such a crypto clock value would be the National Bureau of Standards, using their existing time servers.

Does such a standardized "crypto clock" currently exist?

More information about the cryptography mailing list