Removing Restriction to Access to Multiplier and Allowing Unrestricted Minting of Datacap (FIP-0078)

[anorth]

Edit: FIP draft https://github.com/filecoin-project/FIPs/blob/master/FIPS/fip-0078.md

This is a new discussion topic for the FIP currently in draft at #803 by @The-Wayvy, @Fatman13, @ArthurWang1255, @flyworker, @stuberman, @Eliovp, and @dcasem. This forks the discussion from "Motivation for a Scheduled Reduction in the Quality Multiplier of Fil+ Deals and Example Timelines" #774 which shares many motivations but started off with a proposal for a very different mechanism. I and other FIP editors hope this new channel will enable clearer discussion and debate of this proposal separated from competing/alternate proposals as it finishes the governance process.

Creating a new discussion topic should not be taken as rewinding any discussion or process. The changes in scope here have already been discussed at some length in #774 and anyone interested in the history should start there. Some specific comments and threads leading to this proposal are highlighted here:

• Initial suggestion by @nicelove666 subsequent thread
• CryptoEconLab analysis of scenarios from @kkarrancsu, specifically "Option 3"
• Similar suggestion by @herony-fil
• Clarifying design sketch by @anorth, and subsequent discussion

Current FIP abstract:

The Fil+ program intended to compensate miners for storing non-random data on Filecoin. It was hoped that these datasets would increase the utility of the network and thereby attract paying users. Fil+ did succeed in accelerating the development of data-onboarding tools. It also attracted users to experiment with storing their data on Filecoin, and it is possible that even after subsidies run out some of these users will pay miners enough to have their data continue to be stored on Filecoin. However, the notary system used by 3/4 of Fil+ programs is inefficient, not objective, and riddled with perverse incentives. This proposal removes the permissioned, off-chain pathways to the 10x multiplier, and makes it available to all actors who stake the required collateral.

[f8-ptrk]

can the authors please acknowledge again that they stand behind this proposal. this has gone a long way from what was originally discussed and PR'ed.

Fatman13 (@Fatman13), flyworker (@flyworker), stuberman (@stuberman), Eliovp (@Eliovp), dcasem (@dcasem), and The-Wayvy (@The-Wayvy)

[The-Wayvy]

I support this proposal.

[Fatman13]

I generally support all variants of the proposal that is helping the network transitioning back to a permissionless one, which allows the network to stick to its core value of decentralization defined here. That being said, I have my own preferred variant when it comes down to a multiple choice. As the proposal goes through a "decentralized" governance process, i think it would be great that network stakeholders can have variants to choose from (including maintaining status quo).

• The network is not dependent on any one party, e.g., not dependent on a company, foundation, or other organization to continue operating.
• No centralized parties can control, stop, or censor the network, its operation, or its participants.
• Users and providers are not controlled or governed by powerful intermediaries (unlike markets like Uber or Airbnb).
• Refers to the decentralization movement of 2015+, including IPFS, Bitcoin, Ethereum, and more.

From the mission statements above, it is very intuitive to me that powerful intermediaries like f+ controlling / censoring the network needs to go (or stop meddling with the consensus at the very least). It is very unfortunate that Filecoin network in its current structure is nothing like decentralization movement of 2015+ of Bitcoin and Ethereum.

Again from Filecoin mission...

• The network is not dependent on any one party, e.g., not dependent on a company, foundation, or other organization to continue operating.

One of the most important things is that network stakeholders could actually collectively making big decisions for the network in a decentralized way.

[f8-ptrk]

Fatman13 (@Fatman13), flyworker (@flyworker), stuberman (@stuberman), Eliovp (@Eliovp), dcasem (@dcasem), and The-Wayvy (@The-Wayvy)

#788 stands in direct opposition to #803. please make a decision what proposal you are standing behind and close the other one to, for example, avoid unnecessary governance overhead - what proposal do you want the community to consider for chain inclusion right now?

if both proposals should be considered, be it in the long run, please clarify how this should work out.

[The-Wayvy]

I support both 788 and 803.

I also support a variant on 803 where we give everyone access to datacap and also cut the multiplier linearly or by halving.

IMO:

All three options accomplish our goal without any unacceptable trade-offs.

The FIP-editors should just merge one of them and get it over with.

[anorth]

We can and should merge the PRs, but this is just a step toward each of the proposals going into the final phases of governance. Nothing technically prevents both from passing Last Call, but clearly if they did then Core Devs would not have a clear mandate of what to implement and so would probably appeal back to a governance process to resolve it. I think this is a pretty unlikely outcome though, so no need to solve it prematurely.

[The-Wayvy]

Hey Alex @anorth

What needs to happen for FIP-0078 to enter last call?

[EVS2024]

Just to confirm, does this proposal support the SPs who use random data can get（mint） unrestricted datacap?

[The-Wayvy]

Yes

[flyworker]

I think it is better to keep one FIP instead of keep multiple, so it there is no huge difference. This one sounds good to me.

[MegaFil]

This proposal would cause filecoin network to fall back to the beginning, completely failing to address the junk data while killing any remaining faith and confidence in the market.

This FIP looks smart but is essentially stupid, and that will change the vision of filecoin. If the proposal can't say no to junk data, we think this discussion is worthless.

[The-Wayvy]

胆小鬼寄生虫

[The-Wayvy]

你是Slack上的 Filecoin Protector

作风一摸一样

[Eliovp]

@MegaFil
When evaluating decisions, it's important to look at the collective feedback.
If the majority (clearly), have concerns about a particular idea and you stand alone in supporting it, it's advisable to engage in self-reflection.

Sometimes, support for an idea may be driven by personal gain rather than its true value..
@cryptowhizzard (this one is for you as well ;-))

Some individuals may struggle to accept contrasting opinions, making dialogue with them somewhat challenging (as is clearly reflected in all of these github/slack discussions.

Essentially, fruitless under the banner of dictatorship ;-)

[MegaFil]

@Eliovp You may need to figure out who is in the minority? Otherwise you'll make a joke out of it.

[Eliovp]

If I would make a list of all the leechers, greeders(not a real word I know), the ones here just for personal gain, well, then yeah, we would definitely be in the minorty here ;-)
I'll make a joke out of that :)

[Chris00618]

This suggestion has been discussed before in the Fil+ Governance proposal and it didn't seem to be well-supported.
Regardless, we'll take the idea that voting is the best way to go, and none of us like being represented by anyone.

filecoin-project/notary-governance#974 (comment)

[The-Wayvy]

@anorth @jsoares

I think people already said what they wanted to say in 774 and on Slack.

We haven’t had a new comment in a week.

Can we get this is into Last Call?

[The-Wayvy]

@kaitlin-beegle @anorth @jsoares

IMO

Soft consensus has been reached on the need to change Fil+.

We should start quantifying support for this specific proposal among different constituencies.

[jsoares]

There is no particular tooling, and I don't know that there is consensus on either question, but I suppose that could be determined.

@kaitlin-beegle Would it be appropriate to include this in the agenda for next week's core devs call?

[The-Wayvy]

I’d love to see ranked voting on the different options that have been presented.

Including option 2 from 774 and some others

[kaitlin-beegle]

@jsoares yep! Already on the agenda :)

[The-Wayvy]

@jsoares @kaitlin-beegle @anorth

Can you tell us what you talked about in the core-devs meeting?

[cryptowhizzard]

Well, again to reiterate from my side, I don't support this proposal. I think giving datacap to someone storing junk is just plain stupid and it not according to the mission of Fil+. Datacap should be granted to data that brings value to the network, i.e. data from respected organizations to show them that decentralized storage is a valuable option for them so the Filecoin network can benefit of that in the near future from a commercial standpoint.

[The-Wayvy]

I think Fil+ is plain stupid.

You oppose this proposal because easy access to DC has been your competitive advantage.

[cryptowhizzard]

Your accusation is incorrect and has no factual basis.

[joshua-ne]

I am wondering what is going on with this FIP? Will it be included in the next upgrade, or are we expecting some community voting or anything?

[Fatman13]

To summarize...

• Despite FIP-0078 being the most popular FIP ever, PL don't think consensus has been reached among community.
• The nucleated fil+ governance team and other former PL tooling teams need fil+ to live to monetize (ex, 1, 2, 3).
• PL is determined to push the "actual storage being used" narrative to investors even though the market is clearly not buying it.

On a side note, community funded FEVM based voting tool is being actively iterated and developed.

[fhmd4k]

Very disappointed with PL, even casting a vote is very difficult.

[joshua-ne]

• PL don't think consensus has been reached among community

I'm wondering why it is not in the voting process yet. We really should have a vote on this, which is not technically difficult to do.

[Fatman13]

To summarize...

A totalitarian voting design is being proposed where a guild of 7 people can overrule the voting result of the network stakeholder, which is in complete violation of Filecoin mission. I.E. if network stakeholders voted yes, the aforementioned guild of 7 people can just flip the voting result and say no.

Besides, the proposed voting composition is also questionable, which includes non-stakeholders like clients and core-devs...

Again, on a side note, community funded FEVM based voting tool is being actively iterated and developed.

[anorth]

Actually a voting process is techically somewhat difficult, when you include clearly defining it and allowing appropriate time and space for disussion of the voting constituencies and rules and ways to measure them objectively. We all wish we had it sooner, but we don't, and it's coming, and part of the reason it's slow is to leave room for discussion like @Fatman13 's above (which is in the wrong place for having any effect). We'll get to vote on this when we have a mechanism for doing so that's reasonably likely to be taken seriously by everyone involved.

[Fatman13]

We all wish we had it sooner, but we don't, and it's coming

Maybe it will come out when the network has already reached high percentage of DC rendering the "incentive scheme" effectively void... missing the 天时 (heavenly timing) and critical momentum 😂

[The-Wayvy]

If Juan wanted to poll miners and holders he could have gotten an app built very quickly.

We all know that he doesn't like this FIP

[flyworker]

2024 is coming, we need to conclude it, #774 was stated on [Aug 3, 2023], it has been 6 months since then.

Next step or close, do we want to keep it open for 12 months?

[jaychoucc]

@anorth

[jaychoucc]

support this proposal.

[jennijuju]

@kaitlin-beegle @luckyparadise I am trying to create a poll for this FIP on https://metropolis.vote/ (developed and suggested by FF). I can create a Poll but it requires me to add 5 example responses, however, I am not able to add a FIP to get a sentiment check like someone did for FIP0001v2, sealer ID and so on. Could you please suggest how we can open one?

[jennijuju]

@anorth thanks for the links, seems like the call was focused on the solution proposed in #774 instead of this one (run out of time) and i dont think most of disagreements raised for 774 necessarily apply to this as well. Anyhow - like you said opinions could have changed since then & I have brought this thread&proposal to core devs attention today in the channel.

But in truth authors have been driving it throughout the YEAR and core dev just keep ignoring. 😓

@luckyparadise governance team probably should troubleshoot on why an FIP author feels this way and identify areas for improvement. I want to do some quick validation on the claim that's made by Fatman around "core devs keep ignoring":
- this seems to be the last mention of the FIP and it didn't get to editors/core devs particularly. Governance team suggested to one of the authors they could request the FIP to progress to last call.

• this was the last time fip editors were tagged on this FIP, and Kaitlin responded.

In both threads core devs were not involved so i dont think its fair to say they are ignoring it.

but it could be a very helpful (to the authors) step to demonstrate that there may be much more consensus than previously understood. Take the opportunity.

👍 ❤️ ➕

@Fatman13 - i found some core devs are more likely to prioritize engaging in FIP with some codes + clear spec, and find some core dev/implementers that agree with the FIP direction, get alignment from them on the impact + priority and collaberate with them to drive the progress of FIPs is what I found most effective for getting FIP moving forward. You are managing one of the implementation teams (Venus, part of core dev) - so you have some privilege here comparing to some other community members.

[luckyparadise]

Currently, Metropolis is live and available for crowdsourcing sentiments from the community on this FIP.

To clarify, Metropolis is not a formal part of the FIP process but a tool introduced as part of our new governance improvement mechanisms. It is not a voting tool and is not specified on-chain. This means the results from Metropolis cannot determine if a FIP is accepted or rejected. However, the results can be invaluable for FIP Authors and FIP Editors in assessing community interest and support for FIPs. Metropolis also allows for adding comments and providing qualitative data to support sentiments.

Regarding @jennijuju's previous suggestion, Metropolis is designed for sentiment checks on early-stage ideas and improvements when they are opened as PRs against the repository. Since this is no longer the case for this FIP, I urge FIP Authors or FIP Editors to explore another function of this tool by manually opening a poll for this FIP. We (FF Governance) can assist in disseminating this poll and mobilizing community members to signal their sentiments. I am also happy to set this poll up if it is easier.

[Fatman13]

Thank you for the all the clarifications on Metropolis!

it could be a very helpful (to the authors) step to demonstrate that there may be much more consensus than previously understood. Take the opportunity.

Could any Github account vote on Metropolis? If so then the result is really not as meaningful because it could be manipulated. Wouldn't it be the same to use thumbs up and thumbs down emoji on Github to achieve similar functional results? Why going with extra step?

btw this proposal got a lot momentum in terms of Github account counts, but stuck in a similar fashion.

like you said opinions could have changed since then & I have brought this thread&proposal to core devs attention today in the channel.

I am not sure much has changed besides people have vested interests in f+ have even entrenched further. The change i personally see is that it is much harder to reach consensus now than a YEAR ago.

get alignment from them on the impact + priority and collaberate with them to drive the progress of FIPs is what I found most effective for getting FIP moving forward.

i think this might be a bad way to move FIPs forward as many core-dev don't necessarily have a coherent strategic view of the project.

You are managing one of the implementation teams (Venus, part of core dev) - so you have some privilege here comparing to some other community members.

I am afraid that I am no longer a community contributor of Venus anymore and will start our own org soon.

We (FF Governance) can assist in disseminating this poll and mobilizing community members to signal their sentiments.

What kind of support can we get from FF in terms of mobilizing the community?

[luckyparadise]

By this I mean we could disseminate the poll widely across social media, and or Weekly Governance Updates. I am happy to discuss other areas we could support

What kind of support can we get from FF in terms of mobilizing the community?

Any Github account can signal sentiments on Metropolis. The idea is to slowly and consistently utilize Metropolis as a sentiment check tool. Again, Metropolis is not built to gate how FIPs move along in the process, surely not a mandatory step to take as well. However, we welcome feedback to help us make this tool meaningful and useful.

As has been expressed severally in this thread, soft consensus is a high bar and possibly cannot help unblock FIPs like this.

Could any Github account vote on Metropolis? If so then the result is not as meaningful because it could be manipulated. Wouldn't it be the same to use thumbs up and thumbs down emoji on Github to achieve similar functional results? Why going with extra step?

[Fatman13]

By this I mean we could disseminate the poll widely across social media,

Got it. Thank you!

soft consensus is a high bar and possibly cannot help unblock FIPs like this.

FIP0003 is basically forced upon the community with no soft consensus to begin with. By that token, the repeal shouldn't require a high bar either. The current way of distributing DC is anti-decentralization which is obviously against the very core principle of Web3.

[steven004]

I would like to simply express my support for this proposal as it is being brought up again.

[MEIYAN666]

I'm against the proposal.
We can stop or continue the incentive programme, but we shouldn't openly allow the storage of spam data, which would destroy the vision of filecoin.

[Chris00618]

Also think this proposal is very risky, if it is supported, filecoin will be reduced to the CHIA network where tokens can be pledged. Surely no one will want to store valuable data into filecoin.
Reducing the incentive multiplier for junk data would be better than allowing unrestricted minting of DC . Incentives that don't differentiate between data quality despite free storage are not accetpted.

[ArthurWang1255]

How much have you paid for the DC, or earned from the DC?

[Chris00618]

IMO, datacap allocation plan should be stopped in a gradual manner, not given out unrestrictedly.

[f8-ptrk]

i am happy to present you the quality code we use to generate our quality data

x == 0
while true
    x =+ randomInt(0,1)
    if x > 17'ish GiB 
       break
done

StoreOnFIL(x)

[f8-ptrk]

i am aware that you have to only run this once and the subsequent runs just add a 0 or a 1 or both at the end. as the first part of the "files" is already quality (here is the ultimate mental magic trick: notaries sign this off - so this is per definition quality data - might be humanities most important!!! - we never had any of this data taken off the network - so yeah, its notarized high quality data!) we conclude that the new "file" must be quality too. logic. so we go get datacap and store as many copies as we possibly can.

[ArthurWang1255]

fil+ is the North Korea of ​​the web3 world. It wants to replace the power of the market with a planned economy, but ignores the complexity of the market and the limitations of the team's own knowledge and computing power. If fil+ is necessary, then the cost of this existence is that the entire filecoin ecosystem lags far behind other economies in the web3 world.
Although I still think we should return to a network where all sectors are 1x. Because filecoin's Bretton Woods system should not end for the locked circulation supply indicator or for the interests of a few people. But if fil+ can become a history, then an unlimited supply of multiplier and DC may be a big but acceptable flaw.

[The-Wayvy]

100% agree

[cryptowhizzard]

Can you elaborate on what you mean with the comparison of the planned economy?

What limitation of knowledge do you see?

Since FIL+ went to allocators it seems there have been many opportunities to do things the right way. What gaps do you see in the current allocator system?

[flyworker]

It is 2024 now, real user should pay for the data stored on filecoin network.
If it not, it means the data is not important.

[The-Wayvy]

Can we vote on these two FIPs now that the voting tool is live?

[The-Wayvy]

@anorth @jsoares

Guys, what do you think?

[jsoares]

I remain in favour of moving FIPs out of eternal limbo but, as far as I understand, not much has changed with respect to the how. Per Kaitlin's recent message, the plan is to start beta testing the on-chain voting tool in September, with no date yet for making it binding.

One could still do a Metropolis poll on this to take the temperature on community sentiment in advance of the old process, as @jennijuju proposed above. The community seems to have mixed feelings on Metropolis itself, so not sure how much participation there will be, but my opinion hasn't changed since that thread, i.e. I still think it's the best path currently available.

[luckyparadise]

Just to clarify that the on-chain voting tool is not 'live' yet. What is live is the temperature check tool - Metropolis.

[alberto-gonzales]

HI.

@anorth are there any plans to move this forward in the process or do you aim at leaving this in discussion stage?

[anorth]

I am not personally intending to do this, but would welcome someone else taking it on. I would probably recommend starting fresh with a new FIP draft (based on this one), and opening a new discussion alongside it. This history from this discussion makes it a little hard to get a clear understanding, and it's been a while so the situation may have changed somewhat.

With a little more effort, a cleaner version of this could also remove datacap entirely, and just allow SPs to specify the pledge/power they want per sector (within limits).

I can offer advise and early review if you decide to take this up.

[alberto-gonzales]

@anorth

i am not terribly concerned with the content of the FIP. it is more about procedure and how to unstuck fip proposals that linger around and pose danger to void future fip proposals if picked up again and being implemented.