Testnet Committee Charter and Q2 Funding Proposal

so what are the other 15 hours for? was that for devving the tooling instead of outsourcing it and the documentation efforts? we are happy to have you fill the other 15 hours for docs if you are interested.

1 Like

Most of the time was going to be for development and documentation, yes. Let’s talk about the documentation thing, I’ve sent you a dm on discord

My concern is that as far as I am aware all the things mentioned in the $35k Dreamscape proposal have not been completed. So you are lead dev that still needs to finish that proposal, and lead dev on a grant that has yet to finish phase 1. Now you also want $3k/wk to add another 20 hrs/wk to your plate. My initial reaction is that if you have the bandwidth to meet the delivery of all three concurrent projects, than either your projected workload is actually less than what you are charging for, or you probably won’t be able to delivery all 3 projects in a timely fashion, and likely the projects that pay upfront will just not get the hours they should be getting while you devote your hours to the project that still has to hit milestones to get further payment.

Since people are already saying just focus on the testnet node specific stuff, if you are still interested, you could do just that, and have a smaller workload that is easier to believe can be accomplished. Of course, I would also say that would mean asking for less per hour because what you were using to justify the rate that many people believe is too high was the dev work, which is what you would be cutting out from your proposal.

Obviously, if you feel that your time is not worth anything less than $150/hr, I would understand that, but would then say that this job is probably not worth your time. And as a community we should ask if anyone is willing to run the testnet cluster(s) for a rate that we agree is appropriate for the task. There are a number of us that had been putting a decent amount of time into pulsar that past 6 months completely pro-bono (although I do like the idea of some amount of back payment being included in a proposal for the many hours put in). So if that many of us were willing to do it for that long without ever thinking we would get any pay, I’m sure there are community members that would do it for a rate that everyone can agree on

3 Likes

I would say that we do need to have two chains minimum (At least a certain times). One that is a carbon copy of mainnet, and one that has latest network upgrades ahead of a mainnet upgrade. We need to get contract devs working on their stuff ahead of a mainnet upgrade on the newest code, but we need to make sure that doesn’t stop work on the regular tesnet. We could do a fork ahead of upgrades I guess. I would add that to this scope.

From a tooling POV, the most important thing for me is a way to track participation and compensate node runners. We need more than a few people running validators and node clusters out of the goodness of their heart. Does this type of tooling exists? What validators were up for what blocks over a month? I thought it would have been cool if that data was referenced by a contract that allowed validators to claim, but maybe that is overkill

The other tooling I liked was having unified SNIP-20 tokens with faucets for consistency. Again trying to keep it in line with mainnet and make developement of apps that use established SNIP-20s much easier.

But any tooling could be outsourced I suppose. I don’t think the lead needs to do the dev work necessarily. He/she could, but it should be funded in the same way, IE above and beyond the core hours required.

I am coming to a conclusion about how this system may be a bit broken. And my hope is that we could use this committee to shine a path forward. The way I see it, the discussion here is highlighting the need for a separation of concerns in the committee structure. That is:

  1. There is a job that all committees need which is called “lead”. This is by and large a community organization / project management role. To do this role one needs to have some level of subject matter expertise.

  2. There is sometimes work that needs to be done that requires technically proficient people doing various tasks, building things. What those things are, and who those people are, should be determined by the committee with the lead making the final call and then looking to the community for funding.

To me this separation is really important in general, but especially for technical committees. The problem is that the scope of technical work is part of what you need a committee to determine. I am reading a lot of comments on “what it takes to run a testnet” and I tend to think that people are underestimating it, both in terms of the work needed to do the things they think are necessary, and in terms of not imagining what is possible; what could make things better than the minimum. Now, I could be wrong about this. But, the last place I want to solve that is in a single forum thread before the committee even exists. I want the ability to discuss things in real time and express my vision in the nuanced way that would take too much time to write in a way that doesn’t open itself up to criticism.

So, what i suggest is that we try to use this committee to chart a new path forward. We pass the proposal to setup the committee and pay a small introductory amount to Danny M for the work he will do to lead it for the first 3 weeks say. We then will get a new proposal from the committee that will take into account the consensus within the committee on what should be funded, who should do it, what rates should be paid, etc etc.

I want to be clear that I am not criticizing Danny M, in fact I really appreciate the work he did, and I think that his approach fits the approach of all the other committees. A lot of the stuff he suggested came from discussions we had, but I am realizing that they are not coming across well enough to sufficiently persuade people I really trust and respect. So I want to look for a better path instead. I think this may be one of the more important committees we can have. For that reason I want to get this right.

3 Likes

Thats less important to me than people actually using the testnet. Im fine with the incentives but it shouldn’t just mainly focus on paying those people and have no correlation to how much traction the testnet gets. Starting off the focus should be merely covering the monthly expenses but I dont have a lot of strong feelings about the topic.

Regarding your other comments

Lets not overstate the difficulty here in any of this. Most of what is important to build is already built (look to the tooling that is on mainnet), most of the time spent by node runners on testnet is just setting up in the first place, and thing like the load balancer are incredibly easy. Ask around to the people who have a lot of experience with that (the subject matter experts like me, dan, @ronin_secret ), I could run load balancer for any of the testnets for free in 30 seconds and likely put very little rolling effort into it (i actually already do this) support would be the most time but thats only if something breaks. Testnet doesn’t need to be made overcomplicated or convolutwd, that makes the prop seem more like creating a fulltime job just because vs creating a thoughtful testnet.

1 Like

I think best move forward is to put a 1 time proposal on chain to fund creating the testnet and doing the validator upfront/backpay.

Then after that Danny or whoever can join the dev committee for 5 hours a week to be the central point of knowledge/attention of the testnet. meaning they are in charge of keeping it running, paying the correct people and keeping up with documentation. The dev committee can also budget the testnet nodes.

This way we can start the testnet ASAP but dont have to spin up a committee with expectations of regular funding for just a few hours a week.

I would say that this suggests that my concern is valid and the bandwidth is not available for significant additional work on a new proposal, and if accepted will likely delay the dreamscape work even more. Although a smaller proposal for just a few hours a week would likely be feasible. But with the importance of the testnet work, there would need to be some assurance of it being prioritized and not having its focus shifted away to other projects at times that there might be urgent testnet work

A lot has been said already, expecting an updated charter/proposal will follow, keeping it short for that reason.

  • Missing limits, or desired infrastructure. Testnet should not be seen as a potential source of revenue. It should be fairly compensated. The way this proposal is currently written you would welcome 50 validators? I don’t think that is necessary for testnet, I would prefer to see the required/desired infrastructure and a limit on the number of slots and budget accordingly.

  • Lead role, hours and pay/h seems like this is an attempt to loot the community pool. Would suggest to bring this in line with other committees.

  • Share the concern of @baedrik, I would suggest finding another community member to put in the lead role.

  • Scope seems to be wide. Fact that the development of tools is suggested that already exist does not speak well for this proposal.

  • Providing back pay to the Pulsar team is a nice gesture. Asking 36k/3month for yourself and offering 7k/6 months to the previous team seems like an insult(?)

Not impressed with this proposal, not in the slightest. Definitely a ‘no’ in current form.

1 Like

I support this proposal in some form (adjusted based on feedback from the subject matter experts).

On the concerns raised with Secret Dreamscape, I don’t think any further resources (including time) should go toward that project. It didn’t end up getting much traction, so we shouldn’t try to throw good money after bad, so to speak.