Converging Paths

[Shawn]

(Disclaimer: As you may know I’m currently working as a contract consultant for ECC but I did not consult or consider members of ECC opinions before making this post. It’s my own opinion and not reflective of my work at ECC, they will be reading it for the first time here on the forum, same as everyone else)

It’s become evident over the past few weeks that Zcashs current governance structure is under strain. With multiple user threads posting polls, ECC putting out polls, ZF putting out polls, Aquietinvestor putting out polls, ZURE putting out questionnaires, ZECHUB polls, Twitter polls, etc…

Not to mention the confusion of polling of “separate groups” with ZAC, ZCAP, ZECHUB all competing to be the communities voice when in fact many of the same members are in two or more of those groups. It’s easy to see why users are confused, frustrated and feel that thier voice is not being heard.

How did we get here?

Last time the discussion for a dev fund came up ZF and ECC worked together to gather community sentiment on multiple ZIP proposals via various methods:

1. Account age limited forum poll (to prevent ballot stuffing)
2. Miner signaling via OP codes
3. ZCAP polls
4. Tokenholder poll (from ECC)

This process pointed us in the direction that led to ZIP 1012 and eventually ZIP 1014.

In my opinion, as it sits right now, this older process has broken down, badly. Not due to some conspiracy or intentional sabotage by some group but by the fact that we now have so many people that want to be a part of the Zcash governance conversation and they are all trying to help with the same problem. In some ways this is a good problem to have because it shows that Zcash is important and relevant to many different groups across the world. But unfortunately it’s also leading to duplication of efforts, confusion of which poll or survey “matters” to whom, users going over ambiguous questions with a fine tooth comb and calling foul play, users missing polls, etc… This multipath approach is not helping to bring us closer together, it is leading to even deeper divergence between teams and causing an unnecessary undercurrent of mistrust.

So how can we get all these separate efforts onto one path?

---

Step 1. Stop dividing the community into different groups and polling them separately.

ZCAP, ZAC, ZECHUB, and other groups all need to be accounted for in one large group of users. ZF, ECC, and whoever else is in charge of organizing these separate polls needs to share notes on the users they have and work together to make a new group (Super ZAC? Super ZCAP?) that has all of each other’s missing participants but not duplicates. This new combined larger group of Zcash users is where we can poll to gauge a broad community sentiment.

Step 2. Assign a moderator/mediator to design the questions for polling

Another underlying problem with the current approach is people on both sides are not comfortable with the way questions are being presented by the other side. It’s getting very nuanced where even simple words like “Dev Fund” can have multiple meanings. So my suggestion is to hire a professional neutral pollster to come up with a comprehensive list of polling questions based on the many proposals that have been put forward.

Then for ZF and ECC leadership to get together with the moderator to discuss/debate the structure and wording of those questions. This way both ECC and ZF can collaborate to design effective questions and hopefully eliminate any potential (intentional or unintentional) bias in questions.

Next those questions will be posted on the forum for public feedback, to make any tweaks.

Then finally we can present a Helios poll (or two) to the super group with those questions.

Step 3. Utilize some of the methods that worked last time, eliminate the noise

1. Forum polls

Forum polls are inherently poor and untrustworthy because anyone can make multiple accounts to push the vote however they want. To mitigate this risk last time we (Josh C, Andrew M., and myself) set a forum account age limit on which votes on a certain forum poll would be considered valid. The way to do this is after the poll closes you directly query this forums SQL database for those voters that cast a vote in a poll and prune any votes that didn’t meet the set age criteria. I would be happy to help do this again if we feel it could be useful. I would suggest the questions in such a “semi Sybil resistant” forum poll be the same or similar to the super groups Helios poll questions. Any other random forum polls before, after, or during the “official” forum voting time should be ignored.

2. Miner signaling

I know @aquietinvestor has already volunteered to contact mining pools and trying to get a better turnout than we had last time for OP code polling. I would request that someone from ECC or ZF help Jason make a easy how-to guide that pools/miners can reference. Then it’s a matter of choosing a start/stop block height for miners to do thier signals.

3. Token holder voting

Last time this was an experiment that ECC held that ended similar to the miner signals, ie: not very much participation. Zcash is a long way off from having a proper on-chain voting structure and I don’t think we have time to design something that Sybil resistant (ie: one whale dividing into many accounts to appear like more, or multiple people combining into one large account, etc…) so unless someone has a super easy and fast way to set this up I would suggest leaving it out this time around.

4. Twitter/Discord/Reddit/wherever polls

These should not be considered in decision making since they are so far away from being Sybil resistant.

That’s all, hopefully it makes sense and I hope the community can recognize the need for something to change to avoid a potential quagmire of indecision, even if it’s not following my suggestions exactly.

(Note: That I edited my post to remove a suggestion of who specifically the polling question moderator should be. ECC and ZF will need to agree on whoever is to be a question moderator)

[zerodartz]

i will just say one thing.
different small groups is interesting to see if the opinions are similar or very different to big groups or not.

in the end it doesnt matter too much, but if different would be nice to know imo.

agree the questions going out into ZCG poll should be discussed.
and they actually were, but not enough and werent posted on forum before the first voting and much feedback wasnt taken in and changed the questions.

maybe next round we can do better?

[Dodger]

I fully support all of this.

[gordonesTV]

I fully agree with this proposal.

[peacemonger]

Hey Shawn,
Thank you for your post.

Things are, indeed, getting a little messy. There is some confusion, frustration, poll fatigue, etc… But that’s what change looks like. It’s not a smooth process. Let’s hang in there anyway.

Collaboration is happening. Not in all of the threads between all of the parties. It simply doesn’t work between all of the parties: 233 comments (and counting) got us absolutely nowhere. That’s ok. The community is stepping up and trying other ways. That’s decentralization. That’s change. Gatekeepers are becoming old news. As a community, we don’t have to follow the paths they insist on. We can figure out something else, so we are. And there’s plenty of respectful, productive collaboration happening between community members. The fruits of that will come. I am convinced that we are moving in the right direction, even though it doesn’t always feel like it. That’s exciting.

Let’s zoom out for a second and remember why many of us are here: we want to change the broken financial system. That’s a bold mission. How are we going to do that, if a hint of turbulence makes us uncomfortable? If duplicate surveys are bumming us out, I’ve got bad news. This current problem we are solving is nothing compared to the challenges ahead. The ride isn’t going to get smoother. That’s what it takes. This is a good exercise for us as a community. With training wheels still on, we are slowly moving ahead despite the distractions. I say we buckle up and get through it. It’ll be worth it.

[peacemonger]

I think it’s quite orderly

[peacemonger]

You can think that. I wont take it personally.

[noamchom]

I like the concept here, but in practice we can already all see what the existing polling is accomplishing.

Existing polls let us know that ZAC is ~60 people, and ZCAP is ~200 people. (Most user ran polls here, on Twitter, or in Discord seem to get 50-80 participants at the max) One person can spend an hour right now, to determine if a thing such as a superset even exists. This forum has been open and available for anyone on Earth to join/ participate in for years!

Polling is demonstrating that a majority want the direct Dev Fund model eliminated, they want ZCG separate from ZF, they want more orgs developing for Zcash, they want future funds to distribute via a DAO-like system, they want the trademark agreement eliminated, et al.

How many times do we need to ask the same couple hundred people, the same variations of the same questions? (This isn’t polling on the order of magnitude/ complexity like that of the entire American electorate)

The non-English polling from @aquietinvestor is great, if we need more of anything, that is what I’d suggest we double down on.

Do we really need to allocate more resources to try and solicit/ massage the opinions of the roughly ~200-300 global Zcash ecosystem/ social media participants, who have already been given more than enough venues to share their ideas/ preferences?

[ZcashDeFi]

Let’s be blunt, majority wants ZF to be gone.

[Shawn]

@noamchom

all see what the existing polling is accomplishing.

That’s exactly my point, we can’t clearly see what all these polls have accomplished because it’s not clear. Too many cooks in the kitchen asking variations of the same question, just have a read through the “Moving the dev discussion forward” thread and you can see we are not aligned yet.

My idea is to help distill all these separate polls/questions/ideas into actionable items the ECC and ZF can implement as real world code for audits, pull requests, and software. Ideas of DAOs, splitting off ZCG, non-direct funding models are ideas that we can see some people “support” but they are still nebulous concepts at this point.

What precisely are we doing in November? Or sooner yet, to get the code, audits, and ecosystem ready for a November rollout?

[noamchom]

True, I meant that in two ways… part comic but part serious.

The comic part is obvious, but the serious part is where I’m suggesting that consensus seems pretty obvious. Remove direct protocol funding in November (aka end the direct model), while picking up a specific roadmap to get DAO funding system things done along the lines of what Josh has in his proposal.

The other points of consensus aren’t protocol related, so they can happen whenever makes the most sense (I’m talking about ending the trademark agreement, liberating ZCG, on-boarding as many new developer orgs as possible Qedit, Shielded Labs, etc)

The polling to create a consensus around a 1 year extension seems backward right now, because a majority seem to 1. acknowledge that the regulatory risk isn’t worth the small reward, 2. writing direct wallet addresses into the protocol is antiquated, 3. to lean-in on a DAO like funding model, we need a strong first step in that direction, 4. the track record of direct subsidies hasn’t reliably put value into ZEC, 5. do we really want to have to conduct this same exercise again next year (after the 1-year has come up to expire)

@Shawn

What precisely are we doing in November? Or sooner yet, to get the code, audits, and ecosystem ready for a November rollout?

Ideally, we do nothing. Let the current subsidy model expire, let the direct wallet addresses fall out of Zcash, remove the need for audits/ tests/ new software writing. And in the meantime, lets get to work on the DAO funding system, and not to mention deprecating zcashd.

Yes, the above is contrary to the proposal that I put forward a few months ago. I would still like to see my proposal taken to fruition, but I also admit that it isn’t lined up with ecosystem sentiment… so as a rational person, I’m adapting my thinking about what is the next best future.

[i2pZ7812HTZV69]

Vote Vote I want to Test some Mega Polls

[i2pZ7812HTZV69]

How can polls not be great when its anon these days we saw the UN-expected swing from one direction to another lol

And now we are discussing all about it

[skyl]

I rather like the chaos. More is more in some ways. Arguably, there should have been many more polls much earlier, rather than trying to force a binary at the 11th hour in a 3-question poll. Trying to stop the polling and make one official, executive-approved, poll could be viewed as a way to control the debate and to manipulate the result.

There will be polls in many different venues regardless of whether or not we want there to be.

One of the most crucial parts of survey methodology is the first part: deciding the population you want to sample from and coming up with a good way to get a representative sample.

So, the population is “the zcash community” and the sampling methodology is … a mailing list controlled by ZF & ECC? What are the possible sampling biases? What types of people who might be rightfully considered part of “the zcash community” could be excluded from that list? People who don’t want to register their identities? People who have been vocal about ending direct funding for the organizations who control the list? Is being on ZCAP or other lists proof of using shielded ZEC or holding ZEC? What are the prequisites to being part of the population in question? Would a super ZCAP be a good sample of the population in question? What if a different data point with a different sampling method returns significantly different results? Ultimately this is science. You can’t throw out messy results, messy experiments that don’t show the results you want. Ideally, you have a ton of experiments that converge towards some truth in some meta-analysis. Meta-analysis is how consensus can be achieved in this case - not by throwing out the unworthy experiments and making One Single Official Experiment.

---

In related news, I’m hoping to get around to implementing polling again on free2z - this time with SQL instead of viewing keys. We will implement some cool things like account age, proof of shielded ZEC, and perhaps make voting optionally contingent on other “badges.” I like in particular the idea that you have to send eg 0.001 shielded ZEC to vote (in polls configured that way). Votes will cost one 2Z (~0.01$). 2Zs are only bought or donated… The same questions could be asked with different limitations to understand different subpopulations…

If anyone wants to contribute to the free2z codebase, let me know. The code is healthy and easy to work on and we still have most of the ZCG ZEC for incentives (no USD tho, sorry).

---

Anyways, tldr, in short, I disagree that science is better done with fewer experiments. More experiments, more understanding of the truth. I do agree with your point about a professional poll. I think your overall idea to collect a much better data point from (super) ZCAP is good. Sorry to say I would have to fail these first 2 ZCAP polls in my hypothetical Survey Methodology 101 class.

[Dodger]

Here’s a breakdown of the distribution of voters across the five groups who were polled using Helios:

This is a crude* Venn diagram showing the degree of overlap between ZCAP and the other groups. The area of the circles is proportional to the number of voters:

﹡ Crude because it’s not possible to accurately represent the overlap amongst the smaller groups in a two-dimensional diagram.

40 people voted in two of the polls, 8 people voted in three polls, and 3 people voted in four polls.

There is a small margin of error with these numbers. I’ve made some assumptions about the identity of some people (e.g. I assumed that the Josh, Kris and Paul who voted in the ZAC poll are Josh Swihart, Kris Nuttycombe and Paul Brigner who voted in the ZCAP poll). Conversely, if someone who has used an entirely different nym for two groups, there’s no way to identify that they are the same person.

[noamchom]

In runoff polling, are there any protections against doube/ triple/ 4x voting?

[aquietinvestor]

Thanks for putting this data together. I’m happy to see that neither Zcash Español nor Zcash Brazil have significant overlap with the other groups. We worked with these communities to translate the polls into Spanish and Portuguese to allow more community members to participate. It looks like it worked. I hope to see these groups grow larger and to add more non-English translations in the future, including Russian, Korean, Japanese, Arabic, and Chinese.

The Zcash community is much bigger than we realize.