Jump to content

NEWS
  • WELCOME TO CRYPTOCURRENCYTALK!
  • We've upgraded the website with a lot of new features!
  • 3 NEW THEMES! Click on the + Themes button above!
  • Notice about SOCIAL LINKS: UPDATE YOUR SOCIAL LINKS
  • New video section, add your videos: VIDEO SECTION
  • Advertising will be available shortly, hold tight.
  • If you have any issues with the new site, please submit a support ticket: SUPPORT

Should we Re-Base Gridcoin using this proposal, or should we stay with current codebase and fix it using the Roadmap 4.0 proposals?  

36 members have voted

  1. 1. Should we Re-Base Gridcoin using this proposal, or should we stay with current codebase and fix it using the Roadmap 4.0 proposals?

    • Re-Base Gridcoin using this proposal (a variation of Dash-Evolution/Bitcoin Core Nov 2017 with a Stakeminer)
      24
    • Stay with Current Code Base and fix it Piecemeal using Roadmap 4.0 Proposals
      9
    • Abstain (I do not wish to be involved)
      3


Recommended Posts

 

Quote

I have an affinity for masternodes for a primary reason other than increases in price, let me insist that 90% of the affinity is for feature abstraction and not based on price.

 

Convince me that this is true.  Frankly, I do not believe you.

 

Quote

I watch the external node connections, the peers.dat, the CPID beacon count, and we have over 2500 distinct GRC users.

 

You know the intent of the argument (which in all honesty was pretty clear).  Do not mislead.  We are all -- all 2000-2500 of us -- statistics nerds with varying degrees of experience in programing, development, marketing, funding, research, education, blockchain tech, and more, and we each bring our individual networks -- each as valuable as the next -- to the table.  In other words, don't build yourself up too tall = ).

 

Quote

In my opinion here are the major problems, and the reason I say the solution is fitting:  The latest bitcoin security, script and feature commits, gitian build, mac and windows installer, scalability for airdropping the NN payments to 7000 researchers(dashs budgeting and superblock system), elimination of dynamic witness requirement with dashs gobject voting and superblock modulus, instantsend for small blockchain sizes, Rewards for proof-of-service (solves the online participation issue), Rewards for replacement of interest bearing accounts (POSE), the ability to rank neural nodes (the ranking algorithm can order neural nodes by rank) for cool features for "chosen nodes" (IE the concept that a chosen node acts on an instantsend signal alone), secure voting, neural node assets meaning we have more ability to run custom code in the future if we ever need to add something to only the neural network side.

 

These are not the major problems seen by most users of Gridcoin.  However, they are problems.  But: there are solutions to many of these which are far more beneficial than masternodes, as has been discussed among the community for over 6 months.

 

Quote

If we have 576 blocks per day, and promise to pay out 50,000 GRC per day for research activity, that is 86 GRC would be allocated per block for research.  That means that we would set the block payment so that the block pays 150grc per block (like we did when we started), 90 is held back for research, 45 is held back for Neural Nodes, and 15 grc is paid to the one who mines the block in a green fashion.

 

I think that if our goal is encourage people to connect their IPP to BOINC, this is a bad idea and, again, this is not a Proof of Research protocol, or really any protocol at all.  You seem to be confusing blockchain protocol with incentive structure.

 

Quote

Well, the investor who wants to be a pure investor would lease a neural node.  They would spend $2000 on GRC, lease a neural node from VULTR, set up the software, and then receive about $100 a month in ROI (totally a guess, just for an example here) and leave the $2000 of GRC locked up.  So the terminology might change but correct there would be no investor CPID, no proof-of-stake, no interest, etc.

 

What would happen is anyone who is able right now would set up their masternode.  Future masternodes would require at least several tens of thousands of dollars, or more likely several hundreds of thousands of dollars.  It currently costs ~$585,000 USD to set up a DASH masternode and $30,000 USD to set up a PIVX node.  I would like to see a report regarding economic and political ramifications of masternodes.  Additionally, I would like to see a breakdown of how many masternodes are desirable, how we would target that number, and why.

 

-----------------------------------

 

Regarding the community developed proposals which we have been discussing, exploring, and developing as a community for the past 6 months, I would direct interested parties to the following link:

 

https://steemit.com/gridcoin/@jringo/regarding-the-new-technology-rebase-poll-and-grc-4-0-2018-roadmap-proposals

 

---------------------------------

 

Now.  Who wants to chug some gravy?

 

Edited by jringo

Share this post


Link to post
Share on other sites

@NATEONTHENET I would like to thank you for taking the time to respond to each concern I had. I realize the post comes off as negative, but I do want to clarify that I'm skeptical - not outright objecting the proposal. I would like to understand how this benefits BOINCers and doesn't further consolidate the coins and voting rights in the hands of the few. I'm here because I'm a BOINCer - a proud BOINCer for 15 years (back when there wasn't BOINC and there was just SETI) and I support this coin because it encourages BOINCing.

 

In terms of the voting polls - there actually used to be an option (I don't know about under this new voting system) to vote by CPID. 

 

And in terms of there being 14M coins on the exchanges. That's less than 5% of the supply, and I think you would find many medium size players (like myself) keep coins there just in case the price spikes so we can buy more gridcoins by selling and buying back in. I keep like 15-20K out of my 80K or so in exchange accounts. I'm relatively old in this community as I joined when we were at $0.006, but there are many others that were in before that and watched the price climb to 8 or 10 cents only to watch it fall back to this 3-5 cent trading range.

Share this post


Link to post
Share on other sites
2 hours ago, Frank0051 said:

And in terms of there being 14M coins on the exchanges. That's less than 5% of the supply, and I think you would find many medium size players (like myself) keep coins there just in case the price spikes so we can buy more gridcoins by selling and buying back in. I keep like 15-20K out of my 80K or so in exchange accounts. I'm relatively old in this community as I joined when we were at $0.006, but there are many others that were in before that and watched the price climb to 8 or 10 cents only to watch it fall back to this 3-5 cent trading range.

 

Just a minor clarification -- there are almost 15M coins on Poloniex, and another 14M on Bittrex, plus other exchanges. I'm guessing the total is between 30-40M, which is about 10%, as I said.

Share this post


Link to post
Share on other sites

I, above all else, am a pragmatist and realise that we have a inherent need for the independent Non CPID Investor in Gridcoin (Never BOINCed and hodl) as well as the CPID Investors (Did BOINC and maybe was a founder user).

 

Yes I get 'pissed' that most of these Non CPID Investors and some CPID Investors do not run their wallets to secure the network as often as they should.

 

Yes I am a small Researcher CPID Investor (BOINC and hodl) who has given many hours of time to the betterment of the Gridcoin Community in my own way using any relevant talents I have.

 

I've given this time willingly, without prejudice, as I have always believed that the primary ethos of Gridcoin was driven by its connection to BOINC - Crunching for science/humanity and being rewarded for my efforts by Gridcoin to defray my expenses and/or save my GRC in the hope that some day it might 'bear fruit' as an investment.

 

This is my post on steemit on the 'first month anniversary of joining Gridcoin' - https://steemit.com/gridcoin/@m3rcos1ty/the-old-man-and-the-gridcoin-apologies-to-ernest-hemmingway

 

To me this post still applies and my ideals are still the same!!

 

"Gridcoin is an open source cryptocurrency (Ticker: GRC) which securely rewards volunteer computing performed upon the BOINC platform in a decentralized manner on top of proof of stake.

BOINC is an open-source volunteer computing grid which combines the processing power of all individual users for the purposes of scientific research."

 

(Quotation taken directly from the Gridcoin website)

 

These words in bold and italicised are the reason that I joined Gridcoin and the community.

 

How close are we to these 'ideals' at this juncture?

 

Have we lost our way?

 

I have abstained in the above vote as I cannot support a proposal voted on in this manner ..

 

 

Edited by Mercosity

Share this post


Link to post
Share on other sites

I like this proposal in general. That is not too surprising as I also proposed to choose PIVX/Dash for a rebase some time ago on slack.

The only thing that bothers me is the consensus algorithm, the green cpu miner.

 

@Rob Halförd - (Gridcoin), as I understand you correct a cpid can be registered by burning a small amount of GRC if the cpid holds some RAC. So there is nothing that prevents me from creating 100 cpids, rent 100 virtual servers, run BOINC and registering these 100 cpids to mine GRC. This means I have just increased the probability to mine a block a hundered times compared to mining with a single cpid.

 

It also means that it is more profitable to mine grc on a separate cpid for every machine I run BOINC on. That will not only spam the BOINC databases but also introduce a different kind of competition/attack vector to GRC.

I don't see the benefit of the green cpuminer over a PoS system with a fixed block reward (the reward structure you proposed for example) like PIVX uses. We wouldn't have to invent a new consensus algorithm but could use a already tested and trusted one.

 

17 hours ago, Frank0051 said:

So again, what happens to the little guys? It's not entirely cleanr to me how this plan benefits BOINCers rather than big fish.

 

First of all, if the price per GRC rises it does benefit the BOINCers as well, you simply will earn more money from your BOINC work. This will most certainly increase the competition, but more people running BOINC is a huge part of what Gridcoin is all about, right?

But I still think we need to think about more mechanisms to hold the GRC price up in the future. I envision a system where people and companies can burn GRC on the blockchain to donate or advertise. Every burned GRC supports scientific computation. So if a company burned GRC on our blockchain it could advertise that it supports scientific calculations. I would be happy to use an online shop that uses part of its profit to burn GRC.
I guess that is another topic though.


It is true that Masternodes introduce a certain amount of centralisation, as a switch to graphene would. But there are some benefits attached to it.
1. we would solve the BOINC server DDOS
2. the masternodes would provide resources for advanced features like the NN, light and mobile wallets and act as full nodes to ensure network stability
3.Running full nodes comes with a certain amount of costs and work. Bitcoin has been criticised for pumping alot amount of money in mining operations but not into its network infrastructure. Running a Bitcoin full node does not get any reward and gets more expensive as the blockchain grows. Rewarding Masternodes solves this problem.

 

What I like about Masternodes over Graphenes DPoS:
1. With masternodes you are sure that the nodes are invested in the coin. As I understand it, with graphene a witness is not required to hold any considerable amount of coins as long as it earns the trust of enough voters.
2. There are some privacy concerns regarding graphene. How do you trust an anonymous entity?
3. In Graphene only the witnesses produce blocks, with the proposed masternode system all the nodes can create blocks. Therefore graphene is less decentralised in my opinion.

Edited by huppdiwupp

Share this post


Link to post
Share on other sites
1 hour ago, huppdiwupp said:

I like this proposal in general. That is not too surprising as I also proposed to choose PIVX/Dash for a rebase some time ago on slack.

The only thing that bothers me is the consensus algorithm, the green cpu miner.

 

@Rob Halförd - (Gridcoin), as I understand you correct a cpid can be registered by burning a small amount of GRC if the cpid holds some RAC. So there is nothing that prevents me from creating 100 cpids, rent 100 virtual servers, run BOINC and registering these 100 cpids to mine GRC. This means I have just increased the probability to mine a block a hundered times compared to mining with a single cpid.

 

It also means that it is more profitable to mine grc on a separate cpid for every machine I run BOINC on. That will not only spam the BOINC databases but also introduce a different kind of competition/attack vector to GRC.

I don't see the benefit of the green cpuminer over a PoS system with a fixed block reward (the reward structure you proposed for example) like PIVX uses. We wouldn't have to invent a new consensus algorithm but could use a already tested and trusted one.

 

 

First of all, if the price per GRC rises it does benefit the BOINCers as well, you simply will earn more money from your BOINC work. This will most certainly increase the competition, but more people running BOINC is a huge part of what Gridcoin is all about, right?

But I still think we need to think about more mechanisms to hold the GRC price up in the future. I envision a system where people and companies can burn GRC on the blockchain to donate or advertise. Every burned GRC supports scientific computation. So if a company burned GRC on our blockchain it could advertise that it supports scientific calculations. I would be happy to use an online shop that uses part of its profit to burn GRC.
I guess that is another topic though.


It is true that Masternodes introduce a certain amount of centralisation, as a switch to graphene would. But there are some benefits attached to it.
1. we would solve the BOINC server DDOS
2. the masternodes would provide resources for advanced features like the NN, light and mobile wallets and act as full nodes to ensure network stability
3.Running full nodes comes with a certain amount of costs and work. Bitcoin has been criticised for pumping alot amount of money in mining operations but not into its network infrastructure. Running a Bitcoin full node does not get any reward and gets more expensive as the blockchain grows. Rewarding Masternodes solves this problem.

 

What I like about Masternodes over Graphenes DPoS:
1. With masternodes you are sure that the nodes are invested in the coin. As I understand it, with graphene a witness is not required to hold any considerable amount of coins as long as it earns the trust of enough voters.
2. There are some privacy concerns regarding graphene. How do you trust an anonymous entity?
3. In Graphene only the witnesses produce blocks, with the proposed masternode system all the nodes can create blocks. Therefore graphene is less decentralised in my opinion.

Hi Hupp, I agree with your assessment about masternodes, its a realistic positive outcome compared to the negative points the naysayers bring up.

 

Lets delve into the green cpu miner.  I started writing the miner proof of concept last night.

 

So,  I changed proof-of-work to proof-of-research-age.  Its primarily a proof-of-age-of-last-cpid-mined-block algorithm.  (As the word beacon is to be changed as iFoggz has some good names to replace beacon).

 

So in this new client, I replaced the Groestl hash algorithm (which is one of the X11) and replaced it with a Gridcoin hash algorithm - that breaks the algo to only work on CPU (for now).  (This is not a custom made algorithm, its a port of Groestl with new seeds).  Although over the long term that wont matter much because of our proof-of-age feature, its just to get started on level playing field with everyone mining, and to have our own hash algorithm in the wallet.  

 

So here is what stops the attack above:

To Register a CPID, your CPID is only regarded Once in the chain (new registrations replace old).

If you register more than one CPID, you would also need More GRC addresses, as GRCaddresses are only represented once in the chain.

You have to have > 1 magnitude to receive the benefit of the beacon-age for the CPID.

If you decide to maintain many addresses, you could mine more blocks, but remember the payments are going to others anyway and all you would gain is 15GRC for maintaining an extra address, CPID, and mag > 1 (hardly worth it).  

 

For those who think masternodes is a consolidaton of power, the issue is the benefits of the masternode system far outweigh the negatives, and I would recommend voting for a lower escrow requirement since leasing a neural node is open to everyone.  I believe the broad cross section of gridcoin, leasing nodes on the neural network will represent Gridcoins voting consensus.  Let me say that Im not against adding the existing Voting UI into the new client and somehow adding the ability for non-neural network holders to vote with magnitude, if IT comes up with a future solution to consolidate the votes sum(magnitude) into a masternode-cpid-vote.

 

Let me say, this is a rebase of gridcoin to simplify as much as possible first, then after its running properly, new changes/tickets and the floodgates to improve the client through the standard opensource channel would be open again.  However I really do not want to "pollute" the new client with unecessary features as simplicity and reliability are paramount.

 

 

 

 

 

Share this post


Link to post
Share on other sites
1 hour ago, huppdiwupp said:

What I like about Masternodes over Graphenes DPoS:
1. With masternodes you are sure that the nodes are invested in the coin. As I understand it, with graphene a witness is not required to hold any considerable amount of coins as long as it earns the trust of enough voters.
2. There are some privacy concerns regarding graphene. How do you trust an anonymous entity?
3. In Graphene only the witnesses produce blocks, with the proposed masternode system all the nodes can create blocks. Therefore graphene is less decentralised in my opinion.

 

This is a very fair point - particularly needing to buy in.

 

1 hour ago, huppdiwupp said:

First of all, if the price per GRC rises it does benefit the BOINCers as well, you simply will earn more money from your BOINC work. This will most certainly increase the competition, but more people running BOINC is a huge part of what Gridcoin is all about, right?

But I still think we need to think about more mechanisms to hold the GRC price up in the future. I envision a system where people and companies can burn GRC on the blockchain to donate or advertise. Every burned GRC supports scientific computation. So if a company burned GRC on our blockchain it could advertise that it supports scientific calculations. I would be happy to use an online shop that uses part of its profit to burn GRC.
I guess that is another topic though.

If big whales are able to dominate the pay-outs - which is happening now - small little 1/2/3 computer operators won't be able to get payouts without joining the pool (and the pool sucks if you want to be on a team or otherwise get the credit in your name). That's the entire problem with the new PoS that we're using now. We want little researchers to benefit - and that means provide a system that  pays out routinely or at least allows little guys to get thier due. I have a mag 4-7 depending on the day but I get paid regularly because I bought in when GRC was cheap in order to get pay-outs regularly. It's too high of a barrier to entry to do that now.

Share this post


Link to post
Share on other sites
10 minutes ago, Frank0051 said:

 

This is a very fair point - particularly needing to buy in.

 

If big whales are able to dominate the pay-outs - which is happening now - small little 1/2/3 computer operators won't be able to get payouts without joining the pool (and the pool sucks if you want to be on a team or otherwise get the credit in your name). That's the entire problem with the new PoS that we're using now. We want little researchers to benefit - and that means provide a system that  pays out routinely or at least allows little guys to get thier due. I have a mag 4-7 depending on the day but I get paid regularly because I bought in when GRC was cheap in order to get pay-outs regularly. It's too high of a barrier to entry to do that now.

 

Frank, if we do go with this masternode plan, let me lay out what life would be like with 7 magnitude:

 

If you have 7 mag and do not lease a seat and assuming 7 magnitude pays 7 GRC per day:

You would receive an automatic payment of 7 GRC every day in your wallet, and Occasionally (once a week or so), If you are running your wallet, you would receive  an additional 15 grc for mining a block (when your cpid-age accrues).  That might be 15 grc a week to be fair, but the primary offline payments would be 7grc per day, even if you are offline.

 

 

If you Do lease a seat:

You would receive an additional 30%(per-block) (divided by) (The NN-Node-Count) in revenue once per day.  Realistically, this is an ROI of about 24 months, regardless of the buy in price of the seat.  (Its not fanciful or a guess, I wrote a simulation that proves the ROI is usually 24 months).  If the cost of the seat is $8,000 then the monthly GRC revenue would be about $250 per month - so the daily airdropped NN lease payment would be 8.5 GRC in this case.  When I said the lower buy-in earlier I was using the lowest escrow tranche example.  Higher buy-ins result in the same ROI, but I agree limits the system to whale leases, and that is not necessarily good hence the vote.

 

 

To clarify: No pool in the new system.  The NN signs the magnitude-payment owed, votes on it, injects it into the core, and one block pays all researchers once per day automatically.

 

 

 

Share this post


Link to post
Share on other sites

Dear Rob,

thanks for putting this together and thanks for making this a solid proposal! We are probably at the brink of pushing Gridcoin into its final direction, i.e. its final(?) codebase. I think it is necessary not to rush anything here. Literally I think this could be a decision whether Gridcoin grows beyond a 1 bil USD marketcap or stays an underdog forever (especially the marketing momentum of the decision is important!).

 

General Questions

  • First: What do we want to be? Gridcoin is supposed to be a next generation blockchain that incentivizes international crowd computation, erases the energy-wasting properties of POW and the Matthew effect of pure POS, I think we all somehow agree on that.
  • How can we do that? We are already doing that, but we lack scalability, usability, userfriendliness, public attention and the incentivation to hold GRC. While some of the issues may be solved by sticking to the current codebase and filling holes, most of the issues could be solved by changing the codebase so I am in favor of looking into that possibility and I am open for any solution presented.
  • What do we have in relation to others? Additionally to the above mentioned eco-friendly blockchain security we have on-chain voting for everyone, a feature that has definitely not been advertised as it should have been and I agree with others that we should keep it at (m)any costs.

Questions and remarks about the technology decision

  • Why Dash? You have already stated this in the paper and in the thread that it solves major problems for us, but we intend to be one of the biggest currencies out there, so eventually we should inherit smart contracts, right? Why don't we fork Ethereum, increase the POS block creation part to 100% (past Casper) and build our own superblock-, POR payment- and voting-logic on it (again) that is build as modular as possible, maybe as smart contracts*, to be able to inherit all features Ethereum may roll out in the future. It would be a massive marketing possibility! It is just a gut feeling of course, but I guess adding smart contracts to Dash will be more difficult than adding our features to ETH. Am I wrong here? Personally I would love to promote GRC as a blockchain that is capable of having smart contracts on an energy efficient, non-Matthew-effect basis that even promotes international science. (*If we can't use the original ETH tokens for POR production securely, we could issue POR tokens on our own blockchain and run automated bots on DEXes that always keep 1 GRC = 1 POR, but I guess that won't be necessary.)
  • Have you looked into different blockchains or has Dash been the most obvious one that could serve for our features? I have quite a strong tendency towards ETH being one of the most sophisticated blockchains but maybe others are worth taking a look, of course. CM is repeating the name graphene over and over, but I just realized it is used by many different tokens I think maybe CM has a point here.
  • I need to ask you (and @jringo and @C.M and anyone who would like to add an opinion) from a marketing perspective now: do you think choosing Dash could potentially have the same marketing momentum than saying that Gridcoin forked ETH to make it eco-friendly and sustainable? I believe with Dash or any other smaller blockchain technology we would not have such impact. Seriously, I can see a factor in GRC price of *10 from the day we write a roadmap to do so and another factor *10 when we incorporate the foundation funds in a legal way to pay ETH / Solidity devs to a) keep us on the newest updates of ETH and b) maintain the blockchain and c) add more business cases for us.
  • There is another upside rebasing to ETH. There is a big developer community focusing on Solidity and we as blocklink AG (I am CEO) could employ said devs to offer services and maintainance for the smart contracts as well as the blockchain itself, seen as give-and-take relationship. This can only happen if a popular blockchain is chosen.

 

A small remark on the ROI of Robs proposal

  • If 30% of the block reward go to masternodes and 60% to researchers equaling 50k GRC/d, then it is 25k GRC/d for masternodes. Then lets assume 200 Mil GRC are in escrow for masternodes, we will have a 4,6% interest rate p.a. (this means an ROI of ~20+ years).

Share this post


Link to post
Share on other sites
35 minutes ago, Quez said:

Dear Rob,

thanks for putting this together and thanks for making this a solid proposal! We are probably at the brink of pushing Gridcoin into its final direction, i.e. its final(?) codebase. I think it is necessary not to rush anything here. Literally I think this could be a decision whether Gridcoin grows beyond a 1 bil USD marketcap or stays an underdog forever (especially the marketing momentum of the decision is important!).

 

General Questions

  • First: What do we want to be? Gridcoin is supposed to be a next generation blockchain that incentivizes international crowd computation, erases the energy-wasting properties of POW and the Matthew effect of pure POS, I think we all somehow agree on that.
  • How can we do that? We are already doing that, but we lack scalability, usability, userfriendliness, public attention and the incentivation to hold GRC. While some of the issues may be solved by sticking to the current codebase and filling holes, most of the issues could be solved by changing the codebase so I am in favor of looking into that possibility and I am open for any solution presented.
  • What do we have in relation to others? Additionally to the above mentioned eco-friendly blockchain security we have on-chain voting for everyone, a feature that has definitely not been advertised as it should have been and I agree with others that we should keep it at (m)any costs.

Questions and remarks about the technology decision

  • Why Dash? You have already stated this in the paper and in the thread that it solves major problems for us, but we intend to be one of the biggest currencies out there, so eventually we should inherit smart contracts, right? Why don't we fork Ethereum, increase the POS block creation part to 100% (past Casper) and build our own superblock-, POR payment- and voting-logic on it (again) that is build as modular as possible, maybe as smart contracts*, to be able to inherit all features Ethereum may roll out in the future. It would be a massive marketing possibility! It is just a gut feeling of course, but I guess adding smart contracts to Dash will be more difficult than adding our features to ETH. Am I wrong here? Personally I would love to promote GRC as a blockchain that is capable of having smart contracts on an energy efficient, non-Matthew-effect basis that even promotes international science. (*If we can't use the original ETH tokens for POR production securely, we could issue POR tokens on our own blockchain and run automated bots on DEXes that always keep 1 GRC = 1 POR, but I guess that won't be necessary.)
  • Have you looked into different blockchains or has Dash been the most obvious one that could serve for our features? I have quite a strong tendency towards ETH being one of the most sophisticated blockchains but maybe others are worth taking a look, of course. CM is repeating the name graphene over and over, but I just realized it is used by many different tokens I think maybe CM has a point here.
  • I need to ask you (and @jringo and @C.M and anyone who would like to add an opinion) from a marketing perspective now: do you think choosing Dash could potentially have the same marketing momentum than saying that Gridcoin forked ETH to make it eco-friendly and sustainable? I believe with Dash or any other smaller blockchain technology we would not have such impact. Seriously, I can see a factor in GRC price of *10 from the day we write a roadmap to do so and another factor *10 when we incorporate the foundation funds in a legal way to pay ETH / Solidity devs to a) keep us on the newest updates of ETH and b) maintain the blockchain and c) add more business cases for us.
  • There is another upside rebasing to ETH. There is a big developer community focusing on Solidity and we as blocklink AG (I am CEO) could employ said devs to offer services and maintainance for the smart contracts as well as the blockchain itself, seen as give-and-take relationship. This can only happen if a popular blockchain is chosen.

 

A small remark on the ROI of Robs proposal

  • If 30% of the block reward go to masternodes and 60% to researchers equaling 50k GRC/d, then it is 25k GRC/d for masternodes. Then lets assume 200 Mil GRC are in escrow for masternodes, we will have a 4,6% interest rate p.a. (this means an ROI of ~20+ years).

Hi Quez,

 

Welcome aboard and thank you for the positive tone you bring to the table!

 

Ill pick at this and edit it today after I fix this web site problem.

 

Let me address the last point first on the MN reward ROI:

The ROI figure of 24 months that I referenced earlier is your Return-On-Investment duration on the cost of tying up your escrow.  If the escrow requirement is $10,000 (250K GRC), the ROI is 24 months (that is the projected amount of time to earn back the original locked investment).  The original investment is still yours at the end.  I agree that technically, the money paid to the NN is an expansion of the money supply, (although keep in mind, POS interest bearing accounts would be removed at the same time), and these NN payment components are similar to interest, but it also locks up the initial requirement at the same time, (IE a different characteristic than we have today). The masternode investment results in an increasing money supply and I agree, it is close to a 3.5% expansion in the money supply for the effect of the masternodes being in the network.  (Based on 500 seats, tying up 25MM of our supply, with resulting NN lease payments expanding the money supply by 3.5%) but resulting in a 24 month ROI for the lessee.

 

EDIT:

Regarding rebase opportunities, I went through the top 20 wallets on coinmarketcap.  Im going to save Ethereum for a little later in the day, and give you the benefit of the doubt and talk about that!  Let me skip ethereum for a minute.

I did evaluate PIVX.  I was not impressed, not bashing them, but the community went through the same thing we did with Gridcoin II.  The codebase ended up hugely complex to just implement a better stakeminer.  Dash otoh has a very clean interface and seems one magnitude ahead regarding our requirements.  I did get stuck on Stratis!  Very strongly.  (I even built it and synced it against bitcoin).  The problem with it is me being c# biased, and .NET biased, I killed the idea as its not cross platform.  If we were starting over, I almost would have suggested a stratis port with a new c# UI, but that would have limited us to Microsoft Windows-   (although I still love what they are doing) and I moved past Stratis.  I eliminated others based on complexity, and abstraction problems and public complaints.  The reason Dash fit so well is :  compatibility with Bitcoin, wallet key compatibility for our existing balances, secure voting, the gitian builder, and most importantly the phases the existing budget and superblocks go through.  I still believe in the decision, as long as we are sticking with a bitcoin base.  I went on to compile, modify, and run a dash port for the last 3 months.  

 

Ill be back a little later to talk about Ethereum.  

 

 

 

Share this post


Link to post
Share on other sites
1 hour ago, Rob Halförd - (Gridcoin) said:
1 hour ago, Frank0051 said:

 

 

 

 

To clarify: No pool in the new system.  The NN signs the magnitude-payment owed, votes on it, injects it into the core, and one block pays all researchers once per day automatically.

 

 

So we're suggesting that there will be a super-duper block that pays out daily - which is one of the suggestions we had during the PoSv3 conversion fight. Are the other developers on board with this because it seemed like a pretty divisive idea when some of us [myself included] started posting wanting to see more information on it.

Share this post


Link to post
Share on other sites
4 minutes ago, Frank0051 said:

 

So we're suggesting that there will be a super-duper block that pays out daily - which is one of the suggestions we had during the PoSv3 conversion fight. Are the other developers on board with this because it seemed like a pretty divisive idea when some of us [myself included] started posting wanting to see more information on it.

Well mostly, the ones that are honest.  The ones who have not Hacked Production.  I contacted the lead devs, Marco, iFoggz (and CCed Quez) and had conversations to verify we had enough support to execute this proposal if it is voted.  Some of the devs in Brods camp are obviously biased towards his proposals (Dynamic witness, secure voting enhancements, adding gitian to gridcoin, etc) -that is the piecemeal solution.  Im not entirely against that, I just have an affinity to doing it this way in one swoop.  So far in my tech environment , 95% of the custom code is eliminated and we broke the system up into 3 modular pieces for support by "key" individuals.  It would be much more pleasurable to maintain and add our voting UI into this, imo.  

 

Quez, I still promise to fully comment on Ethereum.

 

 

Share this post


Link to post
Share on other sites

@huppdiwupp @Frank0051

Quote

First of all, if the price per GRC rises it does benefit the BOINCers as well, you simply will earn more money from your BOINC work.

 

Nothing is that simple = ).

 

Quote

But I still think we need to think about more mechanisms to hold the GRC price up in the future.

 

I disagree.  What we need is more use-cases for GRC.  This is sort of what you go on to explain, but in a way that seems to benefit financially well-off entities and no one else.  This is a subject for another day, however.  I would recommend researching SideStaking by PINK, as one possible example.  Its potential is incredible.

 

Quote

It is true that Masternodes introduce a certain amount of centralisation, as a switch to graphene would. But there are some benefits attached to it.
1. we would solve the BOINC server DDOS
2. the masternodes would provide resources for advanced features like the NN, light and mobile wallets and act as full nodes to ensure network stability
3.Running full nodes comes with a certain amount of costs and work. Bitcoin has been criticised for pumping alot amount of money in mining operations but not into its network infrastructure. Running a Bitcoin full node does not get any reward and gets more expensive as the blockchain grows. Rewarding Masternodes solves this problem.

 

DPoS is a representative democracy while Masternodes are oligarchic in structure.

Graphene is not the only DPoS protocol.  Let us not forget EOS and Ouroboros.

 

1. There are other ways!  Possibly better ways!  DWP is one.

2. Good point, but so would DPoS if I understand correctly. 

3. In this scenario, Masternodes = DPoS Delegates -- I do not see the difference.

 

DPoS nodes are voted in and can follow an algorithmic constitution written and ratified by everyone.  If the network does not like what they are doing, or if they break any law in the constitution, they are removed from power.  With masternodes, once you buy in, you're in.  One is there to create blocks based on an agreed upon set of morals, ethics, and their corresponding laws, while the other is there to be controlled by those who have the most money.

 

DPoS centralization is entirely different from masternode centralization.  The former is representative democracy and the latter seems to be oligarchy.  Please correct me if I'm wrong.

 

Disclaimer: I do not support a switch to graphene at this time, however DPoS exists on several chains and not just graphene.  I also do not support a switch to DPoS at this time as it is still new and has yet to be truly put through its paces.

 

Quote

What I like about Masternodes over Graphenes DPoS:
1. With masternodes you are sure that the nodes are invested in the coin. As I understand it, with graphene a witness is not required to hold any considerable amount of coins as long as it earns the trust of enough voters.
2. There are some privacy concerns regarding graphene. How do you trust an anonymous entity?
3. In Graphene only the witnesses produce blocks, with the proposed masternode system all the nodes can create blocks. Therefore graphene is less decentralised in my opinion.

 

The main points I want to make in response to this are the following:

 

1. Graphene is not the only DPoS protocol.  Let us not forget EOS and Ouroboros.

2. DPoS can be based on an open source algorithmically defined constitution which, depending on implementation, could or could not solve all of these issues.  For example, what if the constitution demands the DPoS delegate hold 100,000 of the coin in an escrow account in order to be eligible?  If at any point these coins are moved or spent, power is taken from the delegate.  What if DPoS delegates are required to verify their identity through some process in order to be eligible?  If at any point their IP changes (should the constitution call for static IP, for example) they are removed from power.  So power is taken from delegates automatically if they break the constitution.  Power is also taken from delegates if they lose the trust of people delegating power to them, or if another delegate offers better improvements to the blockchain thereby earning more delegated power.

 

Your third point is far more complex.  When you look at things pragmatically, PoS is already DPoS -- those with the most Stake make the blocks.  DPoS acknowledges this element of PoS and puts laws in place to govern these entities of power.  It then says to the little guy, "You get to choose which entity with stake gets to make the blocks.  Write a constitution, ratify it, and keep it up to date.  You, little fella, are actually in charge."  One quick benefit to this system is that most people aren't going to want to mine or stake blocks on the block chain -- they will want to contribute to BOINC.  DPoS acknowledges this and creates a system that gives them the power to choose who makes the blocks so that they can set up their software and check it maybe once a month.  Masternodes give power to those with financial support, regardless of their actions and regardless of what those "without" think.

 

@huppdiwupp since you generally support this proposal, could you explain how it will benefit BOINC, research, and the general goals pointed out by Quez?  Or maybe break down exactly what masternodes do and how they function.  Thank you!

 

Disclaimer, I do not support a switch to graphene at this time, however DPoS exists on several chains and not just graphene.  I also do not support a switch to DPoS at this time as it is still new and has yet to be truly put through its paces.

 

@Quez In short, I think masternodes will be destructive to the ethos and long term development of Gridcoin, unless someone can convince me otherwise!
 

Quote

For those who think masternodes is a consolidaton of power, the issue is the benefits of the masternode system far outweigh the negatives, and I would recommend voting for a lower escrow requirement since leasing a neural node is open to everyone. 

 

 

@Rob Halförd - (Gridcoin) I would love to hear your thoughts regarding economic and political ramifications of masternodes.  Additionally, I would like to see a breakdown of how many masternodes are desirable, how we would target that number, and why.  Masternode systems do not arbitrarily choose a lease value.  There is intention, expectation, and control.

 

To those following along, note: PIVX is a fork of DASH.  Essentially same codebase, granted it has some interesting additional features.  Also, why would marketcap define quality of technology, particularly in the rapidly developing blockchain technology industry?

Edited by jringo

Share this post


Link to post
Share on other sites
3 hours ago, Rob Halförd - (Gridcoin) said:

Hi Hupp, I agree with your assessment about masternodes, its a realistic positive outcome compared to the negative points the naysayers bring up.

 

Lets delve into the green cpu miner.  I started writing the miner proof of concept last night.

 

So,  I changed proof-of-work to proof-of-research-age.  Its primarily a proof-of-age-of-last-cpid-mined-block algorithm.  (As the word beacon is to be changed as iFoggz has some good names to replace beacon).

 

So in this new client, I replaced the Groestl hash algorithm (which is one of the X11) and replaced it with a Gridcoin hash algorithm - that breaks the algo to only work on CPU (for now).  (This is not a custom made algorithm, its a port of Groestl with new seeds).  Although over the long term that wont matter much because of our proof-of-age feature, its just to get started on level playing field with everyone mining, and to have our own hash algorithm in the wallet.  

 

So here is what stops the attack above:

To Register a CPID, your CPID is only regarded Once in the chain (new registrations replace old).

If you register more than one CPID, you would also need More GRC addresses, as GRCaddresses are only represented once in the chain.

You have to have > 1 magnitude to receive the benefit of the beacon-age for the CPID.

If you decide to maintain many addresses, you could mine more blocks, but remember the payments are going to others anyway and all you would gain is 15GRC for maintaining an extra address, CPID, and mag > 1 (hardly worth it).  

 

For those who think masternodes is a consolidaton of power, the issue is the benefits of the masternode system far outweigh the negatives, and I would recommend voting for a lower escrow requirement since leasing a neural node is open to everyone.  I believe the broad cross section of gridcoin, leasing nodes on the neural network will represent Gridcoins voting consensus.  Let me say that Im not against adding the existing Voting UI into the new client and somehow adding the ability for non-neural network holders to vote with magnitude, if IT comes up with a future solution to consolidate the votes sum(magnitude) into a masternode-cpid-vote.

 

Let me say, this is a rebase of gridcoin to simplify as much as possible first, then after its running properly, new changes/tickets and the floodgates to improve the client through the standard opensource channel would be open again.  However I really do not want to "pollute" the new client with unecessary features as simplicity and reliability are paramount.

 

 

 

 

 

Hey Rob,

 

thanks for the explanation.  So the requirement of mag > 1 is the guard against a multi cpid attack. I can see that with the removal of the team requirement the costs of gaining 1 magnitude might make it unfeasible to attack the network that way. But what if the NN breaks or all BOINC projects happen to be down? This might be unlikely but what is the fallback mode in that a case?

 

 

Share this post


Link to post
Share on other sites

Okay I can see two main reasons for not being open for the concept if I can condense it to this correctly (not counting in more specific questions from @huppdiwupp etc.):

1. No clear pro and con list of technologies to be able to understand Rob's decision.
2. More centralization than before.

1. I know Rob has his reasons but I am also certain he will provide more info about his deciding-process.
2. Just to make things clear, we need the Magnitude Nodes for the magnitude calculation and for voting, right? We can't have unlimited Magnitude Nodes because we would DDoS BOINC servers but otoh we want as many nodes as possible to vote.

What if we just ignore the voting feature of the current DASH Masternodes and adapt our own voting system in a secure manner? We would not have any more centralization than necessary and take one very important part back into the equation to rate Gridcoin.

Lets brainstorm for a second in case our current voting system is not compatible. A smaller fish chooses his delegate. The people in control of the Magnitude Nodes vote, but smaller fishes can send a contract (small transaction) to change it's vote dependent on the balance the wallet holds. If the Mag Node has 1 Million GRC as says "yes", it needs 1Million+1 GRC to push it to "No". Just a thought.

If it is possible we can scratch that from the discussion now and put it on the roadmap.

Share this post


Link to post
Share on other sites

Create an account or sign in to comment

You need to be a member in order to leave a comment

Create an account

Sign up for a new account in our community. It's easy!

Register a new account

Sign in

Already have an account? Sign in here.

Sign In Now


  • Recently Browsing   0 members

    No registered users viewing this page.

×

Important Information

By using CRYPTOCURRENCYTALK.COM, you agree to our Terms of Use.