This is a very good initiative. Great work to VDao.
Please consider redelegating the bootstrapped One to new validator(s) when the elected bootstrapped validator is able to independently be elected in the next epoch without the bootstrapped One (even prior to 100 epochs.)
there are multiple places where new validator register and interact, so it is possible to identify the genuine ones, at least to some degree.
Yup! Weâll be undelegating from them once they hit 1.2xEMS back down to 1xEMS and using the freed up funds to bootstrap others 
I love this idea! This should definitely help us reach our goals faster in regards to getting validators online. Iâm all for it 
We will get there together 
Great point validity. And it was discussed at length in the VDAO meetings. We came to a consensus that it was better to assume the best rather than assume the worst.
As @RhythmValidator stated below, the VDAO will be keeping close tabs on the prospective recipients of bootstrapping funds.
Congrats to the VDAO, to our fellow validators and to Harmony. This will benefit all of us. I have a few minor concerns/doubts, though:
1.- 10k as the minimum required to receive the VDAO delegation seems too low and could attract unscrupulous people to benefit from this. If I had 50k ONEs I could set up 5 validators and I could easily fool you (no offense). Instead, if the minimum required was a little higher (self-stake + delegations), you could make sure that the validator is making an effort to attract delegators and engage with the community, and you could probably keep most bad actors at bay because of the âincreased effortâ.
2.- Allowing 10% fees is an even greater incentive to make an effort to game the system. Also, if your idea is for people to able to get delegators, a 10% will make it more difficult. I suggest 5%.
3.- Assuming that you have already considered the previous points, could you share some more details on what your screening process will look like?
4.- From what I understood, youâll be taking validators from 0 to heroes in ONE go. This is great, but from my experience, people get most of their delegators when they have a higher than average ER. Itâs just a thought.
5.- For each new validator that gets the delegation, either one smaller validator will have to go unelected or a âbig guyâ will have to give up one of their keys. For this bootstrap initiative to be effective the idea is that bigger validators give up their keys rather than already elected validator getting pushed out of election. Otherwise, itâll become a zero-sum game. I know this is not in your control, but hey, itâs just another thought.
6.- Given that this proposal is ready and youâve already posted it here, in case youâre open for feedback, which is the best medium to make suggestions? Here, Telegram, Discord?
Again, congrats, guys! Iâm just playing devilâs advocate here. A well-intentioned devil. Also, Iâm not considering HIPs that havenât been voted yet.
Really exciting to see an initiative like this going live!
Hey @HarmonyValidatorDAO great proposal! Should we require the validators also have a presence on the Testnet to qualify? Wondering instead of raising the bar of the 10K ONEs, we can leave that there, and just add a training-wheels pre-qualification to require applicants to also have a Testnet validator launched. Thoughts?
I love the initiative! This will really help to get more validators elected in the long run!
But we should have an eye on @Rutilant_Hub 's point 5 (Pushing out small validators).
@Jacksteroo from a financial perspective, if a validator is far from election as I am, a training wheel would be no problem. As it wouldnât hurt to modify their current infrastructure to not run mainnet for some days. Setting up a separate node would add even more financial pressure. I personally ran mainnet for a week, but it doesnât represent mainnet very well, as the requirements are much lower for the node. I think first election is kinda a shot into the dark, relying on the feedback on fellow validators.
If a validator has been elected once, I donât think itâs necessary at all (unless it was a shipwreck, which should show in the ER)
We discussed including some requirements for testnet, but there were differing opinions on how helpful it would be. I was originally in support of a testnet requirement, but here were some of the concerns that were raised by the group:
Again - those arenât my personal views or concerns, just a summary of some of the concerns the governors discussed. I see merit on both sides of it, but our approach for now was to keep the program simpler and potentially flesh out those details in the future if we could agree on something. If itâs something the core team really wants to see Iâm sure we could come to an agreement on the requirements.
This is a nice initiative, and I think itâs great the VDAO is acknowledging there are limitations in the current process. I appreciate that you all have agreed to assist the community in this manner
That said, this initiative does nothing to address the root cause of why the VDAO has to do this in the first place: Access to BLS keys for aspiring validators is restricted. They are not easily attained for new/smaller validators, and the protocol does little to assist in making it more accessible
So while I think this bootstrap initiative is a positive gesture/step forward, I hope we donât think this is going to solve the issues that are currently afflicting Harmonyâs BLS key process. I would prefer we build a bigger boat in order to incorporate more/all validators, instead of tossing only a few life preservers overboard and asking the other validators to tread water while waiting their turn
My suggestion is to bypass all of this. Automatically assign all eligible validators one key. After which point you can either continue the automated key-assigning for all additional keys, or allow the larger validators to continue utilizing their current bidding strategies. If you want to have more elected validators - which is explicitly mentioned in the summary section of this initiative - then auto-assigning keys to all eligible validators is a much more effective method
===
@Rutilant_Hub made a few points that I agree with and want to add to (using their numbering system):
#2. Allowing 10% fees seems a bit high when the bootstrap validator will be largely subsidized by the VDAO
I agree with Rutilant that if you allow the fee to be set too high, it will discourage delegators from staking with them, inhibiting the benefit of the VDAOâs support in the first place
#4. I also agree with Rutilant that it will be hard to grow the Bootstrap validators beyond the level that the VDAO is investing in them. For starters, the high APY of ânewâ validators is definitely an attraction that helps them grow their delegation/delegator count
And secondly, a lot of validators simply hit a âceilingâ that is very hard to break through. You suggest ending the bootstrapping of a validator if they reach 1.2x EMS (6.27M) before the end of the 100 epochs. But out of the 161 elected validators listed on staking.harmony, 87 (54%) of them are below that 1.2x (6.27M) threshold! So I think it will be incredibly rare that you would ever need to end the bootstrapping prior to the conclusion of 100 epochs. It goes to show how difficult it is to grow your delegations as a new/smaller validator
Also, if Iâm reading it correctly, if a bootstrap validator doesnât reach a large enough delegation to remain elected without VDAO support, and there are other validators waiting to be brought into the bootstrap program, that the currently elected validator will lose 100% of their VDAO support after 100 epochs? I think thereâs a high likelihood that many of these validators (perhaps 100%) wonât achieve the necessary level of delegation required to remain elected once the VDAO pulls its support. Theyâd be starting over at square one. These validators will just keep cycling through the bootstrap program every few hundred epochs, imo
Are we ok with resigning ourselves to that reality? Is there a better way to get validators elected and to keep them elected? Shouldnât we be looking at those options? Again, this initiative is better than the status quo. But I think auto-assigning a key to all eligible validators would be a much more effective means of achieving Harmonyâs goal of 1,000 validators
Iâm unable to edit the original post from the VDAO account for some reason 
Great discussions here.
We must ask ourselves here, if we donât require the Bootstrap graduates to run Testnets, are we going to be lowering the bar here? Also, having the graduates mess around with Testnet is what Testnet is for. They get to see how itâs being run and what it looks like, how each epoch transition happens, and many more behavioral incentives.
We are not in a rush to move to 200+ validators. Our network is secure enough with 150+ validators. Graduating more well versed validators will also help reduce the VDAO and communityâs burden of answering newbie questions, where these could be flushed out in the Testnet environment, hence strengthening the newcomers. This is important since next year, thereâll be upgrades that will be automatically assigning shards, 1-second finality (devops kung-fu to tackle bandwidth and p2p), zk-rollup, possibly state-expiry pruning, etc.
We will also benefit from having additional Testnet validators this way
IMO we shouldnât set such a tight restriction. In fact, we shouldnât require any max fee, or set a higher limit like 20% max for them to take in some rewards (see ONEs actually show up in their wallets, cherish those moments, let it sink in) and recommend going towards 5% before the end of the 100 epoch delegations from VDAO. Letâs see how human behavior kicks in.
Will the new validator keep it at 20% in all 100 epochs and hurt their chances of gaining new delegators early on then fall off being elected? Or will they move to 0% quickly to gain more delegators? Weâll let the delegators vote with their wallets. Thatâs how delegated proof-of-stake models work. Harmony CTO @rongjian crafted the chain decentralization and security model this way, with EPOS, as he loves to see how people will react and interact with one another in these game theoretical scenarios.
Hope it makes sense
Great remarks @Jacksteroo
We had debates on the minimum fees and the testnet validator requirement.
For the minimum fees, I personally was in favor of 10% to be less restrictive. My point was if the validators need to raise their fees to cover the server costs, they should have the freedom to do it.
But on the other hand, the arguments in favor of the 5% were to prevent the validators to âabuseâ the system by raising the fees too high. Also a new validator, raising his fees to 20% right away, will probably have more difficulties to attract new delegations and be independent.
There are definitively pros and cons, 10% might be a good compromise here?
For the testnet validator requirement, we were also a bit divided on the question, like @RhythmValidator already said.
First of all, the testnet is helpful for new people to get the basics, how to get elected etc. My question is what if a validator is already running on mainnet ? Should they go back to the testnet to qualify?
In the interview, we definitively want to ask questions about the experience on testnet.
Thank you for all your remarks!
1). I get your point but if we make the minimum 50k and you have 250k, you can start 5 validators so the problem is not really solved.
There is a thread (Discussion: Lowering Entry Requirements (Currently 10,000 ONE) - #40 by WhitePeach) open with discussion to lower the minimum to start a validator so in the meantime we cannot ask a validator to have at least 100k or more to apply here I think.
5). Yes this is a great point you made there, thatâs why other proposals made recently (such as in Pre-HIP) are important imho.
Hey @ONE4All, thanks for replying. I agree that a self-stake of 10k should be the minimum. What I meant was that on top of the minimum self-staked amount, the candidate should already have some delegations (50k-100k). This would show the candidateâs commitment and that they made an effort to get there. This requirement would probably also reduce the amount of people willing to put in the time to game the system. Does this totally solve the problem? No. Does this work as a filter? Yes.
Hey @Jacksteroo, thatâs an interesting point, although I disagree. IMO the argument of âletâs see how human behavior kicks inâ would certainly be valid if a validator was acting independently. It would also be valid if there was an unlimited amount of slots. However, given that thereâs a limited amount of keys, for every candidate that is funded by the VDAO, other validators will get affected. Thus, I donât think that a lax approach to the commission fees is the right one.
If a candidate fails after 100 epochs because of his commission choices, not only itâll be detrimental for him, but also for the rest of validators that were affected during 100 epochs only to see someone fail because of some poor choices that couldâve been avoided from the start. Once the validator is no longer funded by the VDAO, then they can take all the risks they want.
In conjunction with this initiative, Iâd like to highlight a group of Harmonauts that are passionate about helping small validators to get elected. I ran across them from a recent tweet and this group is putting their mission in action one new validator at a time. I would like to bring this group to the attention of a larger community as well as some really big fish in the harmony sea to aid them in their mission. It appears that they are working with the validators they sponsor with a support structure that will help them sustain their elected status once elected. All which fall in line with this initiative, in my opinion, but it doesnât require a long process and funding approval from the Harmony Team. Itâs just every day delegators trying to get to the goal of decentralization faster, One validator at a time. More if the group were to grow more quickly. Reddit Group Name: Harmoforce | Twitter Handle: HarmoforceONE . Thanks for reading.