Safe Gossip protocol

Thank you … specifically though what I am attempting to clarify is the legal status of the current implementation of the gossip protocol.

So are you meaning to say or imply that maidsafe’s current implementation of gossip cannot be patented by anyone as it has been around for many years [or is from another open-source repo or has been used in many projects without license] … and therefore it can be included in a larger open-source protocol like PARSEC without any legal issues?

Do you know the origin of this current implementation of the gossip code? e.g. repo source link?

cheers

The gossip protocol is quite old (1989) and common in multitude of software design.

http://bitsavers.trailing-edge.com/pdf/xerox/parc/techReports/CSL-89-1_Epidemic_Algorithms_for_Replicated_Database_Maintenance.pdf

4 Likes

Gossip is definitely part of the establihed state of the art, so no worry for patent issues here.
I think the earliest paper describing the idea is from the 80s:

Epidemic Algorithms For Replicated Database Maintenance, by Demers et al.

There have been many more papers written since then, but I think this is the first one. (I may be wrong)

4 Likes

Thank you @digipl and @pierrechevalier83 for that information.

I think that clearly answers the question.

3 Likes

To be clear, there is no real maidsafe version of gossip, we just use gossip and its proofs and build on that. Hashgraph (the algo in question in your comments on your youtube video) also uses gossip, but so do many many projects using network communications amongst groups of nodes. It is very common. The complainer would just as well have said, they use software so does IBM Microsoft et al and IBM Microsoft et al have patents. It makes no sense.

Even in the smaller field of distributed consensus, you have IOTAS NANO Bytball Raft Paxos(all varieties) Bitcoin and more that use gossip. So it is very common and definitely not patentable. So there are no concerns there. In more general terms hashgraph is only patented in the USA for a start, also its patent is not infringed by PARSEC at all. PAtent lawyers can confirm that, but just reading their patent and our white paper clearly shows they are achieving similar (but weaker) results using different processes (algorithms). Again no worries

15 Likes

Thanks for that clarification @dirvine. I looked at the hashgraph white paper and at the patent. I think the person who commented on my video may have just looked at the opening paragraph of the whitepaper where he (LEEMON BAIRD, the paper’s author) indicates that at hashgraph’s core isn’t merely a gossip protocol but a gossip about gossip protocol →

From the hashgraph white paper:

It is based on a gossip protocol, in which the participants don’t just gossip about transactions. They gossip about gossip. They jointly build a hashgraph reflecting all of the gossip events.

When I looked at the claims in the patent however, it wasn’t clear to me that he was making a claim about this particular ‘invention’ (a protocol for “gossip about gossip”). I am no expert though. Still I can see how it might lead someone to this conclusion and thought it would be good to get some clarity on the issue.

1 Like

Yes IIRC the word gossip does not even appear in the patent.

4 Likes