Dr. Quinten Stokkink

> Goodness on interaction partner is based on ??? Capacity? Balance? Trust? Reputation? Depending on what we choose the Sybil attack will change too Perhaps we can generalize this as...

> latency diversity @grimadas Do you need this now? It's currently not part of IPv8's wild codebase (as the related research is not accepted anywhere yet). I can hook it...

Alright, I'll leave that out then. It would probably also only serve to hinder you anyway: the DAS-5 running 1000 instances per node would be detected as a Sybil attack...

If you need inspiration, there have been some academic works that look at the tag reinforcement of user-generated tags in the Steam Tags system (e.g., http://dx.doi.org/10.1145/3377290.3377300).

For reference, this also overlaps: #3484.

A bit of a memory dump; this is the old way of checking for duplicates: - Store only correctly signed messages by [`mid`](https://github.com/Tribler/py-ipv8/blob/master/ipv8/peer.py#L45) and [`global_time`](https://github.com/Tribler/py-ipv8/blob/master/ipv8/deprecated/payload_headers.py#L20). - Drop incoming messages with...

@devos50 my best idea for solving this at the time was to make the swarm private using [BEP44](http://www.bittorrent.org/beps/bep_0044.html).

@ichorid If I understand correctly, long story short: anonymous payouts don't/won't work?

Do note that IPv8 has been created (and functions right now) as a free-wire-format low-community-count overlay toolkit/framework. Your mid-level use-case (as well as TrustChain witnesses and the hidden services' PEX...

Historically, our biggest problem with large databases (~1GB) is startup time (e.g., #255, #4898). Practically, this leads to users straight-up removing their entire `.Tribler` dir, as Tribler is pulling in...