Why do you assume the spam proposal creates a big pile of ICP? It does NOT.
It actually decreases the amount of maturity created.
In visual form we are moving from the scenario on the left to the scenario on the right, especially on weekends. So 2/7 of the days of the year it was going to be really easy for spammers to take 100% of the voting reward away from more passive voters. Currently we see less spam because you have to post a bunch of spam to chip away at that 1.4 weighing. When the weight was at 20 spammers had to produce 20x less spam to achieve the same effect.
You can keep saying there was ‘no reason’ for the spam prevention proposal, but you are just wrong.
There was a reason for it and it reduces the incentive of a serious upcoming problem. It did not completely remove the incentive but it did shift if from spam directly increasing your maturity to much harder-to-divine reduction in inflation.
I was hoping that the second spam prevention proposal that created a 1% accept threshold would even reduce this incentive further by making it much harder to one’s spam proposal to actually count, but it was voted down. I think after the SNS stuff comes out we’ll see some more robust proposals for two-stage proposals that accomplish this more completely.
1 Like
I haven’t seen a spam proposal in weeks? months? We’ve seen some malicious proposals for named neurons which this still wouldn’t stop.
It seems odd that we are trying to solve a problem that we aren’t even sure still exists and to what degree.
What is abandoned ICP and undistributed ICP? Will Dfinity agree to not track something that doesn’t exist?
Because if it exists and is measurable it is either 1) voters rewards or 2) non-voters rewards, right?
Diego, let’s also remember it has been suggested several several several times that this ICP that does not exist could be used to fund the treasury! I’m not that smart but I’m not sure how you can fund a treasury with something that doesn’t exist…
My primary issue is not the spam although I’d don’t see any data ie how many spam proposals since weighting was reverted.
My primary issue is the fact that you @wpb and @dfisher have suggested funding a treasury with something that was created by the spam proposal (ie abandoned/undistributed ICP) which @diegop has now said doesn’t exist…
So are guys trying to fund a treasury with something that doesn’t exist or does it exist?
1 Like
Did you read @bjoernek’s post? The weekend spam incentive is about to 4x. I mean…there is math, statistics, and everything. Sure we could have just waited until we had a bunch of spam, but why not head it off at the pass?
You have plenty of standing to express your concerns that inflation is not the right way to fund a treasury…I’m not sure why you’re clinging so hard to this. It is ok to be wrong about there not being a reason for the spam prevention proposal. You didn’t see DFINITY’s explanation until now. It doesn’t subvert your opinion and desire to not use inflation for a treasury.
I was wrong to to pick the words ‘abandoned ICP’ I could have picked a much more appropriate word…I should have just called it unminted ICP. See super easy to admit when you are wrong about something.
2 Likes
But there is currently no spam…so we are speculating but ok…seems like we created another problem that we need to solve even if it doesn’t exist yet.
You (conveniently) skipped over one important part though…the abandoned/undistributed/unmited ICP that definitely doesn’t exist per @diegop … how can you fund a treasury with something that doesn’t exist?!
You can see in the code here : ic/reward.rs at master · dfinity/ic · GitHub
// Explication for the implementation of
// `rewards_pool_to_distribute_in_supply_fraction`
//
// The relevant extract from the spec is:
//
// -------------------------------------------------------------------------
// We derive the nominal maximum quantity of ICP that can be
// minted and distributed as rewards from the current ICP supply and
// the days since Genesis. To begin with, this is equal to 10% of the
// ICP supply divided by the number of days in the year (365 normally,
// 366 in a leap year). Over 8 years, this falls to 5%. Note that since
// the supply of ICP might grow (or even in theory fall) during this time,
// voting rewards may not halve in practice.
//
// * We want the rate at genesis to be 10% per year
// * We want the rate at genesis + 8 years to be 5% per year, and to be flat
// thereafter
// * We want the rate to be a quadratic function of time
// * We want the rate to be differentiable wrt time at the point where it
// becomes flat
// -------------------------------------------------------------------------
//
// Calling R0 the initial rate at genesis time G, Rf the final rate, and T the
// time at which the rate becomes flat, the unique solution is:
//
// R(t) = Rf + (R0-Rf) [ (t-T) / (G-T) ]^2
//
// Note that:
// R(G) = Rf + (R0-Rf) [ (G- T) / (G-T) ] ^ 2 = Rf + (R0-Rf) = R0
// R(T) = Rf + (R0-Rf) [ (T- T) / (G-T) ] ^ 2 = Rf
// R'(t) = 2 (R0-Rf) (t-T) / ( G-T )^2
// R'(T) = 0
//
There is clear intention here to establish an amount to grow the supply based on the total supply each day. This formula exists. It produces a real number of minted maturity each day. Now I 100% get in retrospect that a bunch of people feel that if we don’t give it as a reward we shouldn’t mint it at all, but given this extensive mathematical calculation, that one can certainly make an assumption that DFINITY thought through it clearly and had a good reason to mint at least that much Maturity a day.
If it is well thought though, and there was a reason, and now we’re not going to hit that number now, it just doesn’t have to take much imagination to think that some people would try to find solutions for that number to keep being hit in a productive way. You are wholly within your rights to vehemently disagree with the methodology chosen and propose a different one(including that it wasn’t thought through and it is ultimately much more advantageous to let the number decline with reduced inflation). But to express the amount of disbelief that continues to be thrown around just doesn’t make sense to me.
This was not an inappropriate thing to propose given the assumption whatsoever. The continued tone that it was is just not helpful. There are plenty of ways to justify different, better approaches that are much more constructive. Prove the assumptions are wrong or misguided and stop poking the people who made them. Nobody brings all the right assumptions to the table every time. That is why we want diversification of thought and approach. We can all make each other better if we work together.
5 Likes
Well said @skilesare. Thank you.
I think it may take days to get people out of their heads that the “abandoned ICP” does not exist (and will not exist) because it will never be generated… I don’t understand why it is so difficult to understand…
What I’m trying to get people to understand is that these guys are trying to create a treasury but in order to do it they needed to create abandoned/undistributed/unminted ICP.
So let’s be clear about abandoned ICP…it was inflation rewards that went to voters now that this is reduced it creates the gap between planned inflation that they will/are trying to use to create a treasury. These are their words not mine.
2 Likes
Man…I don’t know. I get that you really want these guys to be super smart and execute some Ocean’s 11 type crypto-conspiracy…but it just doesn’t add up.
" Hanlon’s razor is an adage or rule of thumb that states “never attribute to malice that which is adequately explained by stupidity.” Known in several other forms, it is a philosophical razor that suggests a way of eliminating unlikely explanations for human behavior.
I’m not implying that they or I am stupid, but we certainly have ignorance. I don’t remember if exchange rate proposals were weighted at Genesis at .01 or not, but that simple addition to tokenomics created spam incentive. I’m like 98% convinced that you can either have equal weighting across all topics or a voter based system, but if you don’t pick one of those two you are letting spam incentives in.
Do I think @wpb was laying the groundwork for an inflation-based treasury by promoting the increase of rewards to weight 20? Come on. Did I found a non-profit that requires a stupid amount of IRS paperwork so I could not pay myself a salary in September '21 and then build a voter-based reward system into Origyn in November '21 so I could suggest it for the IC in May '22 and then conspire with DFINITY to bring it back up in September '22 in an effort to extract value out of the IC in October of '22? Yeah…sure…I pulled all that together…and we almost had it… if only it hadn’t been for you meddling kids. https://www.youtube.com/watch?v=hXUqwuzcGeU /s (Sorry for the attempt at levity…but I’ve got to laugh at something these days.)
Yep…you nailed it. And almost universally, in every conversation I’ve had where this was proposed, cooler heads have said…“hey…that seems like it could be misused…it could be useful…but we really need to think about it first”. If you were looking to raise an alarm about an idea you thought was bad, you have 100% succeeded…like I just don’t get why you’re still intent on the accusation that this was all a big conspiracy. A simple, no-oversight, value-extracting treasury funded by inflation is pretty much off the table(I don’t think it ever was being proposed at the extremes that seem to be implied…but if it was…you all have done a great job keeping it from moving forward like that).
4 Likes
To be helpful, I will first quote what I said above and then dive deeper @mechaquan :
in any protocol, there are many attacks or issues have that decrease or have NOT happened, but should be measured and addressed.
As I said above, the tokenomics incentive for spam still remains. That hole is still there. That is why we voted to adopt a design that would address the incentive. That is not odd at all. There are many proposals and implementations that solve or address issues that have yet to happen or (seemingl) stopped happening.
Using a hypothetical: if the IC were DDOS’d on due to a security issue, it would be very reasonable to spend time to fix the issue even if the DDOS happened only once.
We at DFINITY see it the same way: the incentive for spam is STILL there, irregardless of how many actors use it. Some argue, the removal of the exchange rate proposal, only helps further incentivize spam.
For context, the incentive for spam was ALSO there before the actors realized it and started spamming. I think it’s reasonable to also deprioritize spam before the issue shows up. Once spam showed up, I think the calculus changed.
Does this make sense @mechaquan ?
2 Likes
Well, we have multiple things going on:
- “abandoned ICP” or “Unallocated ICP” or whatever name they call it was meant to be a metric to record in the NNS the difference between actual and potential rewards. Notably, this only helps to show how much ICP was NOT created.
Also notably, it would be very possible (thought not as clean) to calculate the amount of ICP not created on any given day (A spreadsheet can do that). So people will still be able to figure out how much ICP was NOT created due to the spam proposal.
Will Dfinity agree to not track something that doesn’t exist?
If people feel really strongly about it, I can talk to people at DFINITY about withdrawing support for it, but it’s important to note what it is and what is not. It is NOT some dark pool or slush fund of ICP. People on socials said this, but it is not.
Because if it exists and is measurable it is either 1) voters rewards or 2) non-voters rewards, right?
But it does NOT exist, so rest does not matter.
Diego, let’s also remember it has been suggested several several several times that this ICP that does not exist could be used to fund the treasury! I’m not that smart but I’m not sure how you can fund a treasury with something that doesn’t exist…
This is also a misunderstanding.
The unallocated ICP does NOT exist.
What some people in the forums are arguing is this (in simplistic terms):
-
Before spam proposal, there will be (example fake number) 100K ICP rewards given per day.
-
Once spam proposal is adopted, there may be (example fake number) only 75k ICP rewards given per day.
-
The IC should further decentralize all its activities and use tokenomics to fulfill activities, so IC is reliant on more entities. Same way node providers are paid for their work. Could Developers, marketers, etc… be paid for their work? This is not a unique idea in Web3.
-
Since there Spam proposal reduces the amount of rewards, some argue that creating ICP for decentralizing work can happen WITHOUT adding any ICP inflation to the known schedule (which goes from 10% per year to 5% per year).
Now, you may…
a. Agree with the idea of decentralizing work, but think that the NNS and voting is too immature for this. Maybe there some problems people see that need to be addressed first.
b. Agree with the idea of decentralizing work, but think that the inflation costs are too high. Maybe there is some inflation cost that is more reasonable.
c. Agree with the idea of decentralizing work, but not like the funding sources (MANY funding sources have been proposed from donations to minting to giving locked 8 year neurons).
d. Disagree with the idea of decentralizing work in the first place because you think it will not work as intended.
e. Disagree with the idea of decentralizing work in the first place because you disagree on more philosophical reasons that no practical change can really address.
f. Be Open to the idea if decentralized work, but think it is too risky or doubt the certainty of the arguments.
All these are very valid ideas. I think they should be open discussed and evaluated in the marketplace of ideas.
6 Likes
Ok I think maybe I wasn’t direct enough so let me try again…
- does this actually stop spam? The financial incentive to spam, maybe, but you can still market via a motion proposal to the IC voters for a mere 10 ICP!
- I would also argue that the spam incentive is still there to pump a “treasury” fund with rewards from non-voters…otherwise what value does a weighting of 20 for gov proposals have if my maximum rewards are capped if I vote on everything?!
- is spam really equivalent to a security risk?! Again, this doesn’t stop someone creating a proposal with morally unacceptable material, right?!
- bottomline my point regarding the lack of spam and if this is really a problem that needs to be solved is more about how it doesn’t actually stop spam and seems more like it’s to create a funding vehicle for a treasury by creating abandoned/undistributed/unminted ICP.
If that’s the intent let’s just be transparent about that!
I don’t think it is…it’s been said several times as a way to fund a treasury…even that it’s a preferred option. Did you read @dfisher and @wpb co-authored forum post?
I have two fundamental issues: 1) we made a very centralized decision to reduce voters rewards without being transparent about that but rather under the guise of spam prevention (which as mentioned in my previous post…I don’t think it actually achieves) and 2) I don’t think the NNS should be funding projects or foundations and especially not something as subjective as marketing.
I’d like to see a very detailed list and roadmap of what these projects are planning to do…very similar to what Dfinity had to do to get their funding…that seems fair to ask for before any treasury/funding conversations, right?
1 Like
Very fair questions @mechaquan , I can answer those I can.
- This motion in question reduces the incentive to spam. The proposal of raising to 10 ICP just makes spam costlier. Both are valid tactics. Both may be ultimately done, maybe we need to both reduce incentive for spam and also make spam costlier to ultimately address the problem. That is very possible.
That being said, I will answer why your solution will NOT solve spam by itself:
- Assume motion proposal costs go to 10 ICP.
- If the tokenomic incentives remain, there are many people who would make more than 10 ICP per day from proposals in rewards. To them, the higher cost means nothing. Also, we have seen cases where people “crowdfund” the motion proposal cost. How many people can pay 10 ICP and still get rewards? Many. But maybe it is more insidious, if 20 people can receive 0.5 ICP per day each from a spam proposal, they can all pool it together… again, something they would only do if there was an incentive. And we have seen this happen before.
I disagree the spam proposal was created to “pump a treasury” and I am not honestly not sure what more evidence I can provide for this in this thread. So let me reverse it: what data or evidence can i point out to convince you the spam proposal design is there to solve spam? Maybe the fact that it was actually proposed a while back and it was re-introduced? Not sure what evidence you are looking for. If you tell me what, I can be honest what evidence I know of and do not know of.
I agree. No it is not in same category as a security risk, my answer above only used a security risk as a metaphor to illustrate the design process: just because bad event X has not happened in a while, it does not mean it is fixed if the underlying cause still remains.
Being direct: I do not know what evidence to the contrary you are looking for. For example, I often ask myself, “I would change my mind if X was true” when crafting my own thinking. There are elements my beliefs pivot on. What is the X that would change your mind?
- I have shown in this thread there have been many, many spam proposal2
- I have shown the spam proposal tokenomic incentive remains (and likely to even grow with echange rate proposals change)
- I haves shown that even if we accept the spam proposal and implement it, it does not necessitate or require any sort of NNS treasury so one does not follow the other.
- In this thread I have linked to multiple open threads and discussions so nothing is secret, all pretty much in the open
What would convince you there is no secret ulterior motive between proposals? Or that one does not require the other?
2 Likes
This is a misunderstanding of the motion proposal in question: Spam proposal design.
The spam proposal deliberately calls out that it does not touch this issue (if it had, DFINITY would have voted against it).
DFINITY voted on the proposal in question, not their forum post.
1 Like
@mechaquan upon further reflection, I do not think I properly gave you the credit I do think:
If you believe spam could trivially be solved by “raising to 10 ICP” and you see a more complex proposal that involves tokenomics incentives, I think it is reasonable for your spider sense to think, "If the complex proposal just an smokescreen for something else… like a city council approving an unnecessary bridge just to funnel money to a company?" I would think the same thing.
If I were you, I would also investigate further.
Our differences of opinions really hinge on a few things:
- I think the spam cannot be solved as trivially as you think
- I think spam proposal for changing incentives is not that complex and it reduces ICP created
- I think one can implement spam proposal and not be forced into anything else
Does that make sense, what do you think?
1 Like
Yes, agreed the treasury isn’t solutioned in the spam proposal but Austin does mention his preferred handling of the newly created “abandoned ICP” is to use it to fund a treasury so I do think it’s linked.