search

ethereum / pm

Project Management: Meeting notes and agenda items
Other
1.58k stars 323 forks source link

Ethereum Core Devs Meeting 82 Agenda #155

Closed Souptacular closed 4 years ago

Souptacular commented 4 years ago

Ethereum Core Devs Meeting 82 Agenda

Agenda

The current plan (which can be changed) is to have 1 hour devoted to the Eth 2.0 BLS signature/EIP-1962 and a second hour devoted to ProgPoW. I am collecting 1-2 people from the pro and anti ProgPoW sides to join and have civil discussion.

The BLS curve discussion is to discuss the following EIPs:

The ProgPoW discussion is to discuss, primarily:

Pro-ProgPoW Representatives:

Anti-ProgPoW Representatives:

Representing His Compromise Idea:

Next call: March 20, 2020 14:00 UTC

vorot93 commented 4 years ago

I'd be happy to give a talk about RandomX as an interesting alternative to ProgPow.

timbeiko commented 4 years ago

@vorot93 IMO that would be best suited for a follow up call if the conclusion of this one isn't something along the lines of "we don't want to change the PoW algo until Eth2". "

We can add it after ProgPow to the agenda, but I doubt we'll realistically have time to get to it.

carver commented 4 years ago

Some folks summarized their arguments against ProgPow: https://github.com/MidnightOnMars/EIPs/blob/master/EIPS/eip-2538.md

I don't want to start a discussion here about ProgPow inclusion. I just thought it was worth providing some background for folks who will join the call.

OhGodAGirl commented 4 years ago

@vorot93 I explained why RandomX is a bad idea here:

To summarize: Ethereum is a dominant coin, and that is a power to be wielded wisely. By choosing RandomX, you are fighting with another coin over a hardware base that they have worked very hard to build for and acquire. I don’t think that’s a good faith move.

You never want to share hardware bases with a coin. Ethereum has the hardware base because GPU’s were the ASIC for the chain (and PP by extension).

I think the community discussion needs to be: do we want to keep the principle upheld in the white paper of GPU-only?

gcolvin commented 4 years ago

I strongly object to EIP-1057: ProgPoW being on the list for EFI Review. How many more times can we second-guess our own decisions and retain any credibility?

DECISIONS 77.4: EIP 1057 - Accepted and final. DECISIONS 81.5: Berlin would happen when the BLS pre-compile is ready, ProgPOW would happen the next third Wednesday after that.

CryptoBlockchainTechnologies commented 4 years ago

I move to have Hudsons write up of ProgPow removed from circulation. It is full of his opinions and makes false claims that the "community" is against ProgPow. What community are you refering to? ProgPow only affects miners and this is the only community that matters for EIP-1057. The last miner vote was over 95% in favor of ProgPow.

Finally, as a large scalel GPU miner I have listened to EVERY core dev meeting live. Go back and review all the calls where and Ethash change was proposed (ProgPow was not the first) and Hudson always moved past the discussion as quick as possible obviously hoping it would die. If I did not know better I would say Hudson is on the take by ASIC manufactures. The paper he wrote is garbage and full of his unfounded opinions and seeks to stop ProgPow from happening.

The GPU mining community has waited long enough for Ethereum to live up to the yellow paper being ASIC resistance. We were told to wait over two years ago that there were other more important EIPs. This is the only chance we will ever have to stand up against ASICs. In the end history will show how the devs responded or did not respond to ASICs on the network,

CryptoBlockchainTechnologies commented 4 years ago

I strongly object to EIP-1057: ProgPoW being on the list for EFI Review. How many more times can we second-guess our own decisions and retain any credibility?

DECISIONS 77.4: EIP 1057 - Accepted and final. DECISIONS 81.5: Berlin would happen when the BLS pre-compile is ready, ProgPOW would happen the next third Wednesday after that.

Corrected my post as I better understand your statement above. I agree why does this need to be reviewed when it was blessed and then audited?

CryptoBlockchainTechnologies commented 4 years ago

Some folks summarized their arguments against ProgPow: https://github.com/MidnightOnMars/EIPs/blob/master/EIPS/eip-2538.md

I don't want to start a discussion here about ProgPow inclusion. I just thought it was worth providing some background for folks who will join the call.

It is amazing to see how much money, time and effort is being put into to stopping ProgPow. Those with the most money always have the loudest voices. You notice there is not a paper like this one in favor of ProgPow? It is simple, the GPU mining community is now the minority and the power of ASIC manufactures have long reaches to those of influence.

Bottom line, please provide the community in addition to the names that signed your paid for paper listing which one of them are GPU ONLY miners. GPU miners are the only community that will be affected by ProgPow. ASICs that were made for an anti ASIC coin accepted the risks when they developed them. This is why it was moved to a seperate fork. It is quite apparent those that are against ProgPow are the ones backed by ASIC manufactures or are ASIC miners themselves.

CryptoBlockchainTechnologies commented 4 years ago

I'd be happy to give a talk about RandomX as an interesting alternative to ProgPow.

Let's not muddy the water. No matter what Hudson says there was not all of the sudden a large GPU mining community shift against ProgPow. While Hudson is entitled to his opinion his is definately not a representative of the mining community and does not have a clue what we want. The GPU communtiy FIRMLY stands behind implementing ProgPow after Berlin.

Hudson I dare you, prove me wrong, take another miner poll, AFTER the 4GB date (April 9th) when the majority of ASICs (that are public) will be off the network.

I will bet you 100 ETH that the poll will be over 90% in favor of ProgPow.

OhGodAGirl commented 4 years ago

@CryptoBlockchainTechnologies I understand the passion and frustration, but this kind of personal attack is not helpful, nor should it take place in the meeting notes. Accusing Hudson of being part of the "ASIC cabal" is just as silly as accusing me of being part of the "ASIC cabal" (which, I would note, happens a lot).

Let's try to keep things civil. No personal attacks. We set a precedent for how our respective communities act - in both #YesProgPow and #NoProgPoW - and I do not want to set the precedent that it is okay to attack people, speculate on their affiliations, or punish people for stating their personal beilefs.

We have a responsibility to guide people to why we think ProgPoW is valuable for the security of a chain - not on how it 'saves GPU mining', but why that is best for our chain security and long-term viability. And in doing so, we need to be excellent to each other.

CryptoBlockchainTechnologies commented 4 years ago

Finally Ehtereum has the ASIC manufactures in a position of high vulnerability. Do you really think they were not aware of the 4GB deadline on ~4/9? In fact I bet they were ready to release to the public a new 6-8GB ASIC that will be viable for the next 3-4 years. This is why you saw people coming out of the woodwork saying WTF, why is this all of the sudden coming up now? This discussion threw a marketing wrench into them announcing a new ASIC. Now they are just waiting for the Ethereum devs to not move to ProgPow. They have paid all the bashers, paper writers, paper signers to ensure ProgPow does not happen, they have the loudest voices. If they are succesful, within one month of a thumbs down decision by the devs for ProgPow they will announce new Ethash ASICs with more memory to get past the 4GB DAG.

Are we really this foolish to not understand what their next move is? Are we really this cowardly not to take on the ASICs this coin vowed to be against when it was created? Stand up ETH devs, stand up to the ASICs now or forever be accused of bowing down to them when they were the most vulnerable.

If you move forward now to ProgPow I gurantee you will deal the most serious blow you could ever have done to the ASIC manufacturers. Think of all the money they have spent developing the next gen Ethash ASIC, with higher speed and more memory. If you are truly anti-ASIC you will hit them now before they are able to recoup ANY money from that wasted R&D.

CryptoBlockchainTechnologies commented 4 years ago

@CryptoBlockchainTechnologies I understand the passion and frustration, but this kind of personal attack is not helpful, nor should it take place in the meeting notes. Accusing Hudson of being part of the "ASIC cabal" is just as silly as accusing me of being part of the "ASIC cabal" (which, I would note, happens a lot).

Let's try to keep things civil. No personal attacks. We set a precedent for how our respective communities act - in both #YesProgPow and #NoProgPoW - and I do not want to set the precedent that it is okay to attack people, speculate on their affiliations, or punish people for stating their personal beilefs.

We have a responsibility to guide people to why we think ProgPoW is valuable for the security of a chain - not on how it 'saves GPU mining', but why that is best for our chain security and long-term viability. And in doing so, we need to be excellent to each other.

While I agree with you that I am being quite harsh, this is not the first time Hudson has sidelined the removal of ASICs from Ethereum. My decision to criticize him was well considered over the confluence of the last 3 years. I call a spade a spade and I am a straight shooter. I have listened intently to every call over the last 3 years. I know there are large money interests here and they have the ability to influence those that can influence at the right moment in time. Nothing else can explain his assertions in the paper that the community does not support ProgPow....nothing.

I stand behind my 100 ETH bet with Hudson to poll the miners again after April 9th and the removal of ASICs that the miners will vote over 90% in favor of ProgPow.

holiman commented 4 years ago

I strongly object to EIP-1057: ProgPoW being on the list for EFI Review. How many more times can we second-guess our own decisions and retain any credibility?

DECISIONS 77.4: EIP 1057 - Accepted and final. DECISIONS 81.5: Berlin would happen when the BLS pre-compile is ready, ProgPOW would happen the next third Wednesday after that.

Where were you when the devs decided to bless EIP-1057. You are doing exactly what you stated regarding second guessing. Either show up to meetings when important issues that you "strongly object" to or accept the decisions by those that cared more than you did to participate in the discussion. You cannot show up a year later and hand grenade that decision.

Just to be clear, @gcolvin objected to putting it up once again for discussion about inclusion, on the basis that it's already decided to be included.

forshtat commented 4 years ago

Hello. Is this the right place to ask for some attention for https://github.com/ethereum/EIPs/pull/2542? This proposes adding 3 new opcodes: TXGASLIMIT, CALLGASLIMIT, TXGASREFUND. Without these opcodes, meta-transactions are facing some major limitations.

CryptoBlockchainTechnologies commented 4 years ago

@holiman thanks for the clarification, I corrected my post.

tvanepps commented 4 years ago

I suppose this should be discussed: https://github.com/kik/progpow-exploit

is PP even ASIC resistant?

CryptoBlockchainTechnologies commented 4 years ago

@tvanepps Yet another well timed release of anti ProgPow propoganda. The anti ProgPow printing presses and bots are running overtime trying to stop this release.

ProgPow underwent a thorough audit. Something as simple as this would never have been overlooked. That being said if there any issues they can be checked and fixed during the testing phase. This has no bearing on approval as it has already been approved.

tvanepps commented 4 years ago

@CryptoBlockchainTechnologies Audits miss things all the time. you seem a bit consumed by conspiratorial mood, the author has nothing to do with community members opposed to PP.

maybe we should let people who do this sort of thing review and respond. the PP gitter seems to think it is legitimate

CryptoBlockchainTechnologies commented 4 years ago

@tvanepps I agree with you and the devs will decide not bought and paid for Github bots and spokespeople.

gcolvin commented 4 years ago

To be even more clear @holiman, I objected to seeing #EIP-1057 moved back to Draft. That is, to no longer even Eligible for Inclusion. At this point I think it would be reasonable to drop the target date, but not to change the status of the EIP. Like any EIP it will not be deployed until we are reasonably confident it is free of defects. This why we test. This is why our source is open for the world to inspect.

Souptacular commented 4 years ago

The current plan (which can be changed) is to have 1 hour devoted to the Eth 2.0 BLS signature/EIP-1962 and a second hour devoted to ProgPoW. I am collecting 1-2 people from the pro and anti ProgPoW sides to join and have civil discussion. This is to discuss, primarily:

CryptoBlockchainTechnologies commented 4 years ago

The current plan (which can be changed) is to have 1 hour devoted to the Eth 2.0 BLS signature/EIP-1962 and a second hour devoted to ProgPoW. I am collecting 1-2 people from the pro and anti ProgPoW sides to join and have civil discussion. This is to discuss, primarily:

  • Community dissent of ProgPoW.
  • Technical updates on ProgPoW.
  • Next steps for ProgPoW.

Please make sure ANY dissent sources are quanitified. Twitter bots or papers with hundreds of signatures of paid ASIC supporters do not qualify as community dissent.

Also why is there not a disscusion of "Community approval of ProgPoW." on your list above? It is apparent just the way you are wording this discussion you are again trying to sway the masses that there is some sort of all of the sudden non ASIC community dissent. The opposite is acutally quite true, that there is community approval (do a youtube search of ProgPow and see all the recent videos from the community GPU miners who have posted how to videos). So why state it the way you have?

An example of quantifying supporters/discenters in a meaningful way is:

Without quantifying the supporters/discenters how can you make an intelligent unbiased decision? Without it, any decision made will look biased to the side that loses; and to be quite honest the only side that has anything to lose are the GPU miners. The others only have a risk of loss if there are bugs (if missed by your rigourous testing phase prior to implementation), GPU miners WILL without a doubt lose if this get shelved again.

Finally, if you cannot quantify the data of discenters, then your data is garbage and should not be included in any intelligent decision making process.

gcolvin commented 4 years ago

We already made our decision. At this point only fundamental flaws in the algorithm should change its status in our process.

Souptacular commented 4 years ago

We already made our decision. At this point only fundamental flaws in the algorithm should change its status in our process.

It will be the decision of the core devs if there is a status change to the ProgPoW EIP. Your perspective on the process and decisions made will definitely get time on the call.

Stoniestfool commented 4 years ago

I'll speak on my experience. I started mining when eth was only 5 usd each. I have informed no less than 1000 people about eth. How it differs from bitcoin. What dapps are. The truly great community that eth supports and that supports eth.I believe in progpow for fundamental reason of decentralization. I believe the majority of the community is for fair use and decentralization. Without that the whole point of cypto is lost.

Souptacular commented 4 years ago

Hello. Is this the right place to ask for some attention for ethereum/EIPs#2542? This proposes adding 3 new opcodes: TXGASLIMIT, CALLGASLIMIT, TXGASREFUND. Without these opcodes, meta-transactions are facing some major limitations.

Hey @forshtat! These meetings are the right place for that. However, for the meeting on 3/6 we are only focusing on 2 subjects. It is best to comment on the agenda for 2 weeks after this meeting once it is up. Reach out if you have any questions.

MadeofTin commented 4 years ago

If there is time after the BLS-precompile discussion. I would like to hear updates from these EIPs.

EIP-1962 and its BLS-Signature working group updates https://github.com/ethereum/pm/issues/152#issuecomment-586653678 EIP-2515 - Difficulty Bomb https://ethereum-magicians.org/t/eip-2515-replace-the-difficulty-bomb-with-a-difficulty-freeze/3995 EIP-2315 Simple Subroutines for the EVM 
 https://ethereum-magicians.org/t/eip-2315-simple-subroutines-for-the-evm/3941 EIP-2456 Time Based Upgrades
 https://ethereum-magicians.org/t/eip-2456-time-based-upgrade-transitions/3902/11

MadeofTin commented 4 years ago

Update on the design of 2515 https://ethereum-magicians.org/t/eip-2515-replace-the-difficulty-bomb-with-a-difficulty-freeze/3995/5

CryptoBlockchainTechnologies commented 4 years ago

Adding for context. This is the community that objected to ProgPow and claimed it would cause a community split. SpankChain is a porn coin. ROFLMAO. image

timbeiko commented 4 years ago

Closed in favor of https://github.com/ethereum/pm/issues/159