12 comments

[ 3.2 ms ] story [ 40.1 ms ] thread
I'm so glad this is out. Myself and a few friends tried to tackle this same problem as part of an open source project years ago but we eventually shelved the project due to lack of time.

This will be very useful if it gets integrated into Tox and other chat platforms.

> We implemented the ART protocol described in Section 5 in Java as a proof of concept.

Anyone have a link to the repo for this?

This is why I hate papers and love blog posts: they're less formal, to the point, and you bet that if they claim there's code, there will be a link.

That said, looking for "Asynchronous Ratcheting Tree" in Google, I find this paper and a bunch of irrelevant stuff. You'd think it'd be in the readme of the Github project if there was one. Looks like it's just academic.

Lol good luck finding rigorous and novel cryptography in blog posts. Different media are suited for different purposes.
Pdfs are suited for printing. That has nothing to do with rigor. You're conflating fashion with intellect.

Example: https://people.xiph.org/~greg/confidential_values.txt

Even that fairly technical document contains no formal definitions or proofs. It would also be much improved with actual mathematical notation.

Again, cryptography is a subtle science. It needs proper setup, proper explanation and proper contextualisation. That's not what blogs are good for.

Sure, write a blog to accompany a formal document like a paper, but don't replace one with the other.

Finally, PDFs are only the most popular format for paper writing because LaTeX does not compile well to HTML, and LaTeX is the best language for mathematics.

Not true. It does make formal definitions and proofs. Without being a PDF.
Where are these proofs? Where are the definitions? There's barely even pseudocode describing the algorithm. Honest question: have you read cryptography research papers before?
I'm one of the authors, happy to see this on HN. Just to let everyone know we're currently in the process of updating the paper and implementation, so there will be a more detailed version up in a couple weeks.

There's also an independent implementation in Go here[0] though I don't know how complete it is.

[0] https://github.com/bifurcation/treekeys