<u​nkn8wn69:matrix.org> Hey guys I've created a Python CLI tool called PSST(Polyseed secret sharing tool) that you can use to store your seed more securely! It uses Shamir's Secret Sharing Scheme and is very easy to use.

<u​nkn8wn69:matrix.org> With this setup you won't have to worry about boating accidents anymore.

<u​nkn8wn69:matrix.org> I coded this today so contribution's are very much welcome)

<u​nkn8wn69:matrix.org> github.com/Unkn8wn69/psst

dumb question maybe but why does polyseed need its own sss app

<u​nkn8wn69:matrix.org> This uses the polyseed index to compress the resulting hex code. If you just take your seed in hex to do it it'll be 80+ words per share. All other slip39 tools are designed for BIP-39

<u​nkn8wn69:matrix.org> At least what I could find

<u​nkn8wn69:matrix.org> Its also not a full redo. This app basically only serves as a better tool for setting up the groups and to have a smaller final word seed. It then uses Trezors python-shamir to do the actual sss

i'm studying the monero p2p algo. it all makes sense bar one point- in a completely new network how do seed nodes bootstrap the peerlists? is the initial peerlist hardcoded? i'm talking 2014 era monero bootstrap, imagining there are no nodes that have a copy of the peerlist to send around.

There is a hardcoded list of seed nodes

I don't know how it happened in 2014, but I suspect people added peers manually to join the network

ah yes manual peer connections makes sense. thanks

How did BTC do it?

Inge, sorta the same way. I remember fluffypony also mentioning that it used to connect to the #bitcoin IRC channel and get IP addresses that way

from reading github.com/monero-project/monero/blob/master/src/p2p/net_node.inl i see that we add our own address to the peerlist if --hide-my-port is false, so with this functionality, along with manual connections, i can see how you can bootstrap a network

Bootstrapping a p2p network is a whole area of research, actually. Many papers on that.

Common approaches are "well-known peers" (aka seed nodes), manual peer adding, and saved peerlists. There's also a DHT table but DHT is a p2p network itself, so it's kind of chicken and egg problem.

And there's also whole ipv4 space scanning if p2p network is already big (works with DHT)

it's a bit disgusting that IPv4 scanning is actually a thing, and that it works.

The world should have moved onto IPv6 at least a decade ago

Then we'd just use multicast groups and not worry about it. routers would find our peers for us transparently.

Multicast is only for LAN though, and what I found on IPv6 Internet multicast is still highly experimental

I suppose. but it's been available for ages now. if it was getting heavy use, it would rapidly mature out of experimental status