Where communities thrive


  • Join over 1.5M+ people
  • Join over 100K+ communities
  • Free without limits
  • Create your own community
People
Activity
  • Jun 20 2016 02:11
    @alexvandesande banned @algotrader2013
  • Jun 05 2016 10:31
    @alexvandesande banned @adamskee
Sonia-Chen
@Sonia-Chen
oh we will definitely qualify customers, we learned that the hard way. long story. once the Ethereum Foundation sells VDF they will do the same.
Andrew
@vtbradle_twitter
The rationale does not compute, why commit to ethash for a chain with questionable economic incentives like ETC, while knowing ETH has a PoS roadmap. If ETC was truly your target, then ETH’s switch shouldn’t be a factor for you.
Kristy-Leigh Minehan
@OhGodAGirl
Let's address this one by one:
Sonia-Chen
@Sonia-Chen
it is not. @vtbradle_twitter not interested in volunteer input?
Kristy-Leigh Minehan
@OhGodAGirl
  1. Core Scientific and ProgPoW have nothing to do with each other. Timelines do not add up. Core Scientific hired me as their CTO after their CEO joined. ProgPoW was done well before then.
Sonia-Chen
@Sonia-Chen
progpow is very interesting to us,we are learning a lot. when we started we were hoping that the ETH devs could give us rules! little did we know that they are being overrun by nvidia.
Kristy-Leigh Minehan
@OhGodAGirl
  1. NVIDIA and AMD have both had equal input into ProgPoW, and numerous individuals in the EF were a part of that journey.
  1. nChain has absolutely nothing to do with Ethereum, and I am not a professional consultant or advisor to them. Though my work in the blockchain does ensure I touch every single coin.
Shall we talk about how you are really upset, because you did not receive funding?
Sonia-Chen
@Sonia-Chen
@vtbradle_twitter we have started an open progpow asic design (yesterday). The initial feedback was great and we will - slowly - continue.
Kristy-Leigh Minehan
@OhGodAGirl
And your design was already debunked.
You are off - again - by a factor of 64.
You deliberately sow misinformation.
Sonia-Chen
@Sonia-Chen
2000 Kristy.
our next steps: continue with open progpow design, process feedback, make it better. find out more about power consumption. I don't think we are interested in this asic, because of what I said before, and because bitmain and inno are better setup for that.
Kristy-Leigh Minehan
@OhGodAGirl
Wonderful. And I will continue to refute you.
Sonia-Chen
@Sonia-Chen
I'm setting myself another self-imposed 48hr silence in this group, starting now, to improve ETH governance. Good luck everyone! Everything will be good.
Kristy-Leigh Minehan
@OhGodAGirl
Let the best technocracy win.
Danno Ferrin
@shemnon
I feel discussion about the particulars of ProgPow should not be on Governance. There are gitter channels for progpow.
Ghost
@ghost~55c3ed250fc9f982beac84b3
I actually think that unless there are other pressing matters to discuss here, this channel is just fine for discussing ProgPOW
Andrew
@vtbradle_twitter
At least keeps it out of AllCoreDevs
Ghost
@ghost~55c3ed250fc9f982beac84b3
sometimes it is kind of annoying that once a really interesting discussion starts, someone tries to "shoo" it away :)
Ghost
@ghost~55c3ed250fc9f982beac84b3
I don't really mind where the discussions are happening, as long as they are happening :)
Andrew
@vtbradle_twitter
With ya there
Danno Ferrin
@shemnon
I think governance about ProgPow is on topic, details of Asics and miner economics are better served in ProgPoW-review.
James Hancock
@madeof_tin_twitter
I am trying to find the right room to discuss the difficultly adjustment algorithm. Any tips on where is best?
I’ve been doing some research :)
consciousEntity
@consciousEntity
I would suggest gitter AllCoreDevs
Nick Savers
@nicksavers
@madeof_tin_twitter if it's just a question you could ask in any of the client implementations rooms (Discord for Geth, Riot for Parity) otherwise if it's a complicated discussion a thread on the Ethereum Magicians might be more helpful.
The AllCoreDevs Gitter chat is more for clients devs among each other for coordination.
James Hancock
@madeof_tin_twitter
Geth has a discord 🤔 didn’t know.
I think it is probably more appropriate for one of the rooms. Just a few questions
Greg Colvin
@gcolvin
@sneg55 "The committee would vote on EIPs to be implemented.” So I suppose this commmittee is going to pay the client developers to implement them? Or are open source contributors supposed to obey The Committee for some other reason?
Nick Sawinyh
@sneg55
I have nothing to do with this coup proposal, just found it on twitter.
Greg Colvin
@gcolvin
Twitter—where the bird shit falleth from the sky...
Brent Allsop
@BrentAllsop

Greg Colvin made the following comment on the Gitter Ethereum/governance chanal:

Greg Colvin @gcolvin Mar 29 17:15
“I personally abhor Twitter and Reddit, but I shouldn't let that color my judgement so much. Yes, they are means of communication. What I said is that they are not communities. What I would like to see is more organization, more attempts to be legitimate voices for the community and its subcommunities. Right now it's like a shouting mob that I have to tune out to get any work done.”

Canonizer.com is for exactly for this kind of organization and consensus building . It stops all the yelling since everyone can just reference what camp they are in. If anyone wants something, they can support an existing community or camp already working on that or create a new consensus building topic, or new competing camp in an existing topic for others to join and help.

The only hard part, is finding enough people that want the same thing you do. Once you achieve that, people will find a way to make it happen.

We’ve started the Ethereum Consensus Project (see: https://canonizer.com/topic/210-Ethereum-Consensus-Project/1) for exactly what Greg is asking for. We’ve seeded it with the following 3 consensus building survey topics:

ProgPoW: https://canonizer.com/topic/211-Programmatic-Proof-of-Work/1
State Fees: https://canonizer.com/topic/212-Ether-State-Fees/1
Ethereum Consensus Algorithm: https://canonizer.com/topic/213-Ethereum-Consensus-Algorithm/1

These are just 3 seed topics, to get things started. Anyone can make any improvements, as long as they get approved by existing supporters. And anyone can start any additional topic to build a consensus community on anything.

If there is anything we can do to help with this process, reach out to me, or use support@canonizer.com . Our team at Canonizer.com is fully dedicated to the Etherium community. And please be aware that this is still a crude prototype, still with lots of issues and things that need to be fixed. So any an all help with this is greatly appreciated. This is an open source system (https://github.com/the-canonizer/canonizer.2.0), being developed for free, by volunteers.

Jean Cyr
@jean-m-cyr
@BrentAllsop Am not overly optimistic algorithmic consensus will help. My main concern with any approach, be it gitter, magicians, canonizer, or otherwise, is the allowance of anonymous registration and voting.
Brent Allsop
@BrentAllsop

It’s not algorithmic consensus building. It just measures the consensus of the wiki camps people build and support any way you want. (i.e. expert consensus vs popular consensus… of existing wiki camps representing growing communities.)

We have a phone verification system, where people can be verified by providing their phone number. This works in the US now, but we have plans to implement is worldwide as soon as possible. So someone will need to pay for 2 phone numbers, to have 2 votes. And we are working closely with the Self Soring Identity community for a much more capable KYC verification system in the near future. So canonizer algorithms will be able to easily filter out anonymous supporters, if you wish.

Nick Sawinyh
@sneg55
anonymous supporters could use disposable numbers for sms verification
Brent Allsop
@BrentAllsop
For now, yes, but soon we will have a much more robust KYC self sovrign identity verification system, which will filter them out. So it would be a waste of time, knowing any support would soon be filtered out, once we have KYC self sovereign identity proofs.
And we will be able to do do constant SMS verification, not just one time, to mark someone as not valid, if they do not maintain their phone.
How easy is it to get disposable numbers?
Nick Sawinyh
@sneg55
pretty easy, there is a bunch of services for that, i.e. https://mobilesms.io/#pricing
Brent Allsop
@BrentAllsop
Those are all one use numbers. You can't get the same number, every month, forever, right?
Nick Sawinyh
@sneg55
It’s called numbers with revalidation and it will be more expensive for anonymous supporters, but it’s possible.
Jon Stevens
@lookfirst
Twillio should be easy to use to blow that method out of the water.
Brent Allsop
@BrentAllsop
As I said, it is a temporary not perfect solution, till we can get full KYC user proofs through the up and coming self-sovereign identity systems. We are a small team working as fast as we can. We’ll get there, eventually.
None of that will get in the way of real users, creating real high quality state of the art camp structures, representing concisely and quantitatively what someone wants and why. It will stop all the yelling.