The Asset Priority Poll, August 12 at 4pm UTC

We have an asset priority poll coming up on August 12 at 4pm UTC which is scheduled to run for 7 days.

This poll meant to give the community, governance, and the interim risk team a heads up on which assets to focus on first. It will also allow us to collect some information about general levels of interest for some tokens. Unlike most polls in the system, there are no requirements to justify, prove, validate or defend the poll, or the results it produces. This is simply establishing the order of work to be done leading up to MCD launch

The tricky part here is that there is no facility in the portal to allow us to drag and drop a list of items into the order a voter would like to see.

We can reverse engineer some solutions though.

Option #1: One poll for all the assets

In this model we’d have a single poll which includes seven options to choose from:

Which asset do you think the community should evaluate first?

  • zxy
  • abb

Because we can only pick one choice in a poll, users can only select their #1 choice. At the end of the poll we would tally up the stake weight for each asset types and order them that way.

Option #2: Yes/No poll for each asset

We have seven concurrent polls, one for each asset. Each poll contains the question:

“Do you wish to prioritize this asset?”

  • Yes
  • No

This allows voters to express their interest in each one. The total MKR staked for NO would cancel out MKR staked for YES for each token. If drunk whale comes along and votes No to all of them we would order by least negative votes.

Option #3: Weighting poll for each assets

This is a bit of an outlier option. We have seven concurrent polls, each poll contains the question:

“How would you rate the priority of this asset?”

  • High
  • Medium
  • Low

In this set up we would allocate a weight to each response. High == 4, Low == 2. This allows us collect more detailed signals as each asset has a numerical modifier attached to the stake weight. Gets a bit complicated though.


At the risk of getting meta, here is a poll to determine what kind of poll we should put in our polling system:

EDIT: Options two and three previously contained a selection “Don’t evaluate.” It has been removed as per the conversation in the thread.

  • Option 1: One poll for all the assets
  • Option 2: Yes/No poll for each asset
  • Option 3: Weighting poll for each asset
  • None of the above, I have an idea I’ll explain below.

0 voters

This poll will close on August 5th.

Some thoughts and questions:


In terms of the ‘Don’t evaluate’ option, what sort of response would be required for the Foundation to entirely drop (from the first wave, not forever) one of the suggested assets? If we go with option #3 the poll outputs a score. How low does this score need to be for a collateral to be dropped?


In option #2, how do we translate the following situation into an action:

Yes + No > Don’t Evaluate
Don’t Evaluate > No
Don’t Evaluate > Yes


How many assets are we trying to launch with? Would it help if we had a target? If we are looking for Eth + 1 other, then we could easily take poll Option #1. If we are looking for as many as the community wants to see in the first wave, then the poll should be structured as below (but then we need a success criteria)

  1. This asset should be evaluated for the initial set.
  2. This asset should not be evaluated for the initial set.

If there is an internal target for the launch of MCD, how many assets can be evaluated and greenlit by governance in the allotted time? Can time and effort be saved by culling assets from the process early? In this case would it be preferable to state upfront the foundation expects to have time to evaluate ‘x’ assets, and that we would evaluate the top ‘x’ in the poll?


Seven tokens would include Eth, is this necessary? Is it safe to assume that everyone is happy with Eth? Is there any conceivable future where MCD launches without Eth as a collateral type? If not, it shouldn’t be part of the poll.


It’s great to see this happen, I’m very excited to see how the poll turns out, and also to see dashboard v1.5. Lots of these points come off as negative, but in general, I think these are all sensible options for a poll. I just think it’s valuable to point out edge-cases and potential problems before we run into them.

That’s a lot of questions. Here’s a first crack.

No idea. I’m extremely reluctant to try to come up with thresholding for a system we are in the middle of testing. I would guess that we get the numbers back from the poll then talk about what they mean.

Many popular and unpopular assets are going to be added to the system, and they need to be evaluated by the Risk ecosystem to determine if there are quantitative or qualitative reasons to exclude them. My guess is the short list of exclusionary reasons are related to the introduction of existential threats into the maker ecosystem and/or illegalities in the collateral being considered.

I don’t think it would be appropriate to block anything from consideration at the ‘popularity’ polling stage. The Don’t Evaluate option was added as a way to capture a complete lack of interest and/or general community sentiment. If it creates an expectation that it represents a ‘veto’ option it should probably be rephrased or removed.

Tough question. I personally would like to see three or more simply for the sake of having enough asset types in the system to fully understand the changes that MCD brings. It will also make for a better narrative than “MakerDAO launches MCD with ETH”.

We have no idea. There are two tracks; the launch date is dependent on the community and the risk teams getting assets into the system while the Foundation powers through the roadmap. If something disastrous happens with the roadmap, we will have to delay. If it turns out that we get bogged down in endless debates and can agree on nothing, then we will have to push off launch until we figure out how to optimize the process.

If governance can diligently work through a minimum amount of assets in a reasonable amount of time, and the Foundation and it’s partners complete their milestones, then we will launch SoonTM.

It can if the low interest assets represent a greater amount of risk evaluation.

Never say never, but… no, I can’t imagine a situation where that would be possible.

3 Likes

Thanks for the responses Rich, I agree that ‘Don’t Evaluate’ should be rephrased or even just removed altogether, I think it does imply a ‘veto’ option.

It also sounds like Eth should not be a polling option if it will be added regardless.

In general I’ve been thinking about the requirements and structure around on-chain polls. One thing that I think is very important is to make it extremely clear before the poll takes places what actions the different possible outcomes will lead to. Having these clearly defined beforehand should reduce disappointment and misunderstanding around the eventual outcome.

It will also make for a better narrative than “MakerDAO launches MCD with ETH”.

We need to think about how much we care about this, while I agree this is a better narrative, it also makes the transition more complicated. It would be great to get some more discussion on https://forum.makerdao.com/t/navigating-the-waters-between-now-scd-and-mcd-with-the-dsr/ about the topic.

I’ll remove it, the signal isn’t terribly valuable. We can probably assume that voter turnout for each asset type is a reflection of interest.

It raises an interesting question about whether rephrasing a poll in motion invalidates the previous votes… We should probably come up with a rule about that.