Pre-Dev-Update Thread! Yay! :D

[happybeing]

Not merged yet, but safe_vault is stirring…

https://github.com/maidsafe/safe_vault/pull/942/files

[Mindphreaker]

Nice catch. I wonder if they managed to fix the memory bug with this release. I think it’s the only thing that’s holding back external vault testing but I can’t find a related Github issue. Looking forward to tomorrows update.

[Shane]

It’s so close now, I can smell the commits.

[tfa]

Today safe-api (master) isn’t compatible again with safe_vault (fleming). Maybe my test was too early and repos are not ready yet. Note also that I have updated Rust to version 1.41 (yesterday I didn’t pay attention that I was using 1.40).

[gouda123]

what can we expect today is a good question

[anon40006692]

No fleming that’s certain

[Southside]

We can expect more snidey comments from you and an honest assessment of the hard work that has gone in over the past week from the dev team. Whatever happens, you won’t be happy.
How much MAID do you hold and what will it take to buy you out?
Never mind crowd-funding bug-hunters, I think there may be support for buying you out so we don’t have to endure your malign presence here anymore.

[scottefc86]

@gouda123 just in case you missed it

[upstate]

Yeah @Southside, don’t be rude! Just be passive aggressive like other folks!

[bridge]

How about have a self vacation from maidsafe? Maybe 3~6 month, then you can find something.

[Audity]

It really makes my day to be able to rise above and not get mad about incomplete sentences floating within posts on a forum without any capitalization or punctuation. [Maybe I just haven’t seen “starsmik levels” of dishonesty in awhile, just idle banter. Nastypasty’s remark bothers me more, but that’s all I can/care to say.]

Anyway just enjoy the ride and find new things in your life that help you live it.
You might know the answer but refuse to partake in it!

[Southside]

Are we certain that NastyPasty is not Mr CheesyNumbers alter-ego?
Could it be there’s a wee tag-team circle-jerk going on…?

[anon40006692]

Why it bothers you wanted to trigger gouda123 stay calm

[Southside]

one keyboard, two accounts…

[tfa]

Today safe-api (master) is compatible again with safe_vault (fleming).

Status changes every other day. Last minutes updates before delivery ? (but no rush required from my part).

[Audity]

I am calm.

You, on the other hand…

[gouda123]

New repo appeared in Github- GitHub - maidsafe/bls_dkg: Implementation of a BLS DKG mechanism, requires signing key, encryption key and SocketAddr of participants

What can this mean, new issues/ tasks, dont recon this was mentioned anywhere in the updates, or anyone?

[Southside]

First mentioned back around July last year IIRC, EDIT: actually May 18 by @Nigel MaidSafe Dev Update - May 10, 2018 - #12 by Nigel

DKG Distributed Key Generation is a supporting technology to BLS.
I’m sure someone will find a more accurate/elegant way of putting it but thats my understanding.
EDIT: Did you click on the link to ?

dashpay/dips/blob/master/dip-0006/bls_m-of-n_threshold_scheme_and_dkg.md#distributed-key-generation-dkg-protocol

[Shane]

This has been mentioned a few times, as recently as November in updates.

This is a very low-level implementation which isn’t really relevant in practice to the majority of the forum readers, but the long and short of it is that the network needs to find a way to prove whether or not a node in the network is being honest about its transactions, and a way to generate quorum so that the network can agree as a whole whether it accepts a nodes actions.

What this repository you’ve flagged is going to do is allow the network to agree on cryptographic signing keys, prove after the fact that a given node had the keys it said it did (thereby proving its actions were all honest) and generate the signing keys / quorum decisions without having to have a central trusted body (like certificate authorities in the HTTPS/TLS system of the internet).

Put simply: this is an incremental step towards an autonomous network, rather than a network controlled by Maidsafe exclusively.

[dirvine]

Absolutely. We already have a version of DKG implemented that uses parsec, but currently does not provide a complain/justify path. So this repo will allow that but also be more generic so clients can also use it and freely set m and n parameters and not be limited to thresholds of 2f+1 (f == faulty/byzantine) which is a weaker quorum and not a democratic quorum in a traditional BFT algorithms (sync or async) as they have a >2/3 quorum, so the threshold max in using 2f+1 is 1/3, which is less than 50% and unnatural This crate will simply improve on that and is likely required for client multisig to be meaningful.