Funding proposal 1: Parallel Initial Block Download

[ardocrat]

Done: TxHashsetDownload is starting in parallel with TxHashsetPibd · Issue #3754 · mimblewimble/grin · GitHub.
This time it happened at beginning of PIBD synchronization, on getting 3 / 17856752 leaves.

[l33d4n]

it sounds good but I think we should wait with the funding to check how it goes with the rust first. The beta just released 2 days ago so in a few weeks/months after enough users try it and report bugs/improvements we will have a more stable version and enough information to discuss it.

My main concern is to avoid another round of funding for fixing bugs, optimization, etc.. I think that funding implementation now while it is still in the beta version is a safe way for additional expenses.

@Anynomous

1. To have access to upfront payment there are two conditions a) all the deliverables for the previous funding requests must be finished, b) upfront payment for a milestone/deliverable is only accessible to developers and community members with a proven track-record such as @davidtavarez and @Yeastplume .

I am against any funding model that is not defined as Code > Release > Payment (or “Milestone > Payment” in this case)

Upfront payment for milestones It’s just like the old funding request model, the only difference is the frequency of the payments and doesn’t solve any of the issues I mentioned here:

@l33d4n

the funding requests method is much more complicated, especially when we can’t enforce it, make sure what the end result will be or what will go wrong along the way as happened several times. It worked, until it didn’t.

@l33d4n

The question that needs to be asked is, why put community funds at risk in the first place?

Bounties reduces the risk to zero and simplifies the whole process with a clear definition for the end result, so we can avoid unnecessary arguments and dramas

I can expand if you want but we all know that the last funding requests did not go as we expected so if we want to avoid drama and keep it simple, that sounds fair:

@ardocrat

I think it will be more fair for all people who is going to be paid by CC Fund to follow same rules and be equal. We already had problems with track-record of last 3 months, so we learned from this.

[Anynomous]

All the above mentioned issues raised are all covered for by:

@Anynomoud topic:10584

3. To have access to upfront payment there are two conditions a) all the deliverables for the previous funding requests must be finished, b) upfront payment for a milestone/deliverable is only accessible to developers and community members with a proven track-record such as @davidtavarez and @Yeastplume

But your opinion is noted, you basically want pure pay after bounties.

It is fair since everyone can build a reputation and track record to get access to upfront payment. Since it is on deliverables level, the risk to funds is much smaller than in the current 4 month upfront paid Funding Requests and the inconvenience is much less for developers who already have to wait since going from approval to payments can take up to two weeks.
Besides, we also pay @Cobragrin upfront, otherwise that would be to unfair compared to long term devs.

Regarding funding PIBD. Implementation will take months, so we can wait but do not have to since it will take time to implement and test in Grin++ before getting activated.

[ardocrat]

@Anynomous

bounties

Yeah, this is the most fairest way to sponsor CC projects with equal rights for everyone. As we learned from mistakes many times. Some people dropped work/gone after getting payment, even @igno.peverell left after 3 months of funding. I think he was the first who pointed this problem for community, it can happen any time if we will pay on front. Let’s take example with Freelance Work. Workers are getting payment as guarantee that customer will not go away with results of work or just will not find another worker, here developers trust to Fund and Project itself. Trust to this side is 100% strong.

@Anynomous

It is fair since everyone can build a reputation and track record to get access to upfront payment.

Its also easy to lose reputation for what we saw for last months sadly. We also learned from story with @satoshocrat and others before him… We can see how fast CC Fund is declining, my opinion if we are learning from mistakes, we should avoid to make them again to spend energy and Fund money with discussing same problems again if such problems will come. This is life. Everything can happen.

@Anynomous

Besides, we also pay @Cobragrin upfront

I see he is making all tasks and even more I think his role can be extended as full-time manager of CC projects. I don’t think we need full-time developers at CC, cause CC is not about single Core development like in Grin Rust, @Yeastplume, @tromp, @phyro are doing good work in Rust, Cryptography and Science. Here we are using different languages and different technologies. We just need better management for projects sponsored by CC Fund.

@Anynomous

Implementation will take months

All code/logic is already written in Rust, from my point of view it can take 1 month+ to migrate, but I understand @davidtavarez is more experienced in Python than in C++, this is why I suggested to bring @david as C++ expert in cooperation to help him to reduce future bugs and take some leadership on architecture with his experience. Maybe somebody else will come to help if this bounty will be interested and open for everyone, writing tests, as example, is also work.

[Cobragrin]

@ardocrat

Some people dropped work/gone after getting payment, even IGNO left after 3 months of funding. I think he was the first who pointed this problem for community, it can happen any time if we will pay on front.

Core team but frankly does it matter ? CC or OC funding/spending guidelines should apply both. Checking deliverables is same. At this point we trust @Tromp, @phyro, @quentinlesceller about checking deliverables and about @davidtavarez we trust @mcm-mike @Anynomous @hendi @Mokhtar @Neo as well.

I see no point of discussing future funding proposal for CC or OC before the deliverables met. I am sure davidT will complete Api+Cli+ Doc.

@igno.peverell drew the line, but some people forgot it obviously.

[l33d4n]

@Anynomous

But your opinion is noted, you basically want pure pay after bounties.

In short, yes.

@Anynomous

It is fair since everyone can build a reputation and track record to get access to upfront payment.

It depends on how we define the last funding period and everything around it, is it “proven track-record”? I tried to avoid personal pointing in my reply above because of his contribution over the years but If you want to go into details about the last period, the facts is:

The last funding request was for a 4 months period from January to April 2023. Now, one and a half months later, none of the funded tasks have been completed and our only way to get updates is by following his Github activity or asking you as a CC member, as I did few times.

Over the past few months any attempt by any community member or our GK (@ardocrat, @Trinitron, @phraudsta, @leriseh291) to ask for updates about the funded tasks has been met with angry or insulting replies:

@davidtavarez

My guy! I’m not dating you! I’m not gay, do you, but I’m not dating you.

@davidtavarez

Thanks for the clarification. My advice: get a life, you need one, and remember that envy corrodes the soul.

@davidtavarez

Is this a topic about me now? again?! what is the obsession with me? seriously, get a life, guys.

@davidtavarez

I will not engage with this account created 5 minutes after writing thing …Stop wasting my time. Get a life, get some help.

Another issue was the progress tracking for the funded tasks:

@l33d4n

I brought the follow-up progress issue to the CC meeting (2023-4-25), the original idea was to follow-up the GK tasks, but in the end I suggested adding to the GK duties follow-up for each funded task.

@davidtavarez and @Anynomous as CC members supported it, and @Cobragrin created the progress tracking file as agreed but when he asked @davidtavarez one week later (2023-5-2) to update the status of his ongoing funded tasks he resisted:

@davidtavarez

No, and I never agreed on participating on any “Progress Tracking Scheme”…

Now just to make it clear, the previous tasks are in progress and not abandoned but the status is still unclear because of lack of communication, ego games and drama all around.

With all due respect to his previous contribution I don’t want to go through this again and again, the funding model I suggested above (not my invention, it exists and works everywhere else) prevent us from getting into situations like this without making it personal for specific dev or definition of “proven track-record” and reputation.

[davidtavarez]

@l33d4n

Now just to make it clear, the previous tasks are in progress and not abandoned but the status is still unclear because of lack of communication, ego games and drama all around.

You are correct, nothing has been abandoned. Here you can follow the progress of each task individually: Grin++ v1.2.9 · GitHub more details on each repo (PR#221 & PR#7). Anyone can try it for themselves and feel free to contact me with any questions.

I have lost interest in participating in conversations that never end. There are better things to spend time on, for example: The recent NBA postseason was epic. I would have wanted the Heat to win, but maybe it was finally Denver’s time. I wonder what Boston will do for next season, any ideas?

[AceKaplin]

Bounties bounties bounties. We have countless flops to show that paying in advance doesn’t work. Why are we trying the same thing?

1. bounties works well for gitcoin
2. bounties works well for Monero fund
3. Grin tried pre-funding, but that didn’t work well

Given the above data points, why would Grin CC try a pre-funding strategy again? Why isn’t grin trying bounties now? Trying the same thing again and expecting different results is the definition of insanity.

[Anynomous]

@AceKaplin

Bounties bounties bounties.

We have bounties, we have had Funding Requests. So far bounties did not work out so well, only coin Swap and the Python cffi secp256k1-zkp wrapper worked out with bounties.
Switching to bounties only would be more limiting than our current system. I see no logical reason why having only bounties instead of bounties + RFC or conditional pay upfront bounties would be better. Dirrent funding types attract different types of developers and we need all devs we can find.

[Trinitron]

I would argue that milestone bounties have not worked well, but total bounties have. Total bounties have not paid out for unfinished work afaik. If milestones are used again I think they need due dates and specific progress requirements.

[AceKaplin]

Where are these bounties? I had no idea they existed, or I would be bounty hunting.

[Trinitron]

One that exists only on the forum is the bounty from Tromp, 1 BTC for an open source MacOS grin miner meeting a certain performance requirement.

[ardocrat]

Check Projects → Wish List at GrinCC website.
Seems like outdated a bit.

[Cobragrin]

Some are here published on newsletter.

[AceKaplin]

None of these are bounties. A bounty should include two things:

1. the deliverable
2. the reward

Tromps bounty is the only clear bounty I can find anywhere and it isn’t even published officially:

1. deliver: a miner that can achieve 0.5 GPS on fancy Mac proprietary software machine
2. reward: 1 BTC

Until we publish bounties like this, we cannot say “we tried bounties”

[Edit] fixed detail of tromps bounty

[AceKaplin]

Also, if we want to attract devs, bounties should be published somewhere very easy to find. E.g. only one click away from grin.mw home page.

[tromp]

@AceKaplin

• deliver: a miner that can achieve 1GPS on fancy Mac proprietary software machine

My bounty offer [1] required only 0.5 GpS on an M1 Ultra.

[1] Bounty - 1 BTC + 100 Grin for a MacOS M1 C32 Open Source Miner with ≥0.5 Gps - #2 by tromp

[ardocrat]

@AceKaplin

None of these are bounties.

Can be a reason why

@Anynomous

So far bounties did not work out so well

[AceKaplin]

Exactly. We can’t say bounties didn’t work well. We haven’t posted any bounties.

[Anynomous]

I think we should move this discussion back to PIBD.
Anyone feel free to start a seperate discussion on the three funding models:

1. Funding Request
2. Bounties with conditional upfront payment of deliverables
3. Bounties

Regarding all three of these funding Requests,ANYONE CAN MAKE A REQUEST FOR FUNDING OR BOUNTIES.

Sorry for capitalizing but this so important to stress that everyone can do so and as such propose or find tasks to work on!The only task of the CC and OC is to give a final verdict and make the transaction since we manage the funds.

@AceKaplin feel free to propose your own projects a bounty. Probably funding has to weight till BTC is above 30k though, but you can already make plans and discuss it.