[Discussion] Activating FLOT

[Nagalim]

FLOT responds to a change in demand and puts nubits into or takes them out of cirulation to match this change. Nu pays FLOT a flat reward per unit time without any regard for performance. If we begin to consider performance incentives we need to be very cautious because we may influence the signer’s opinion of the best thing to do in a given situation by personally incentivizing one option over the other.

The current form of flat rates allows Nu to give raises or pay cuts via motion, so there is still oversight. However, it is slow and can easily turn into a popularity contest unless signers are simply paid uniformly.

Adjusting for some kind of automatic criteria, however, has a lot of problems. For example, if we pay for a successful pool balancing, even if we are 100% sure this was a good and useful txn, we have influenced signers by not incentivizing their more controversial decisions.

Because that’s the core of the issue. If we alter the egalitarian way we pay our signers we will disincentivize controversy amongst signers by necessity. And controversy is the heartbeat of a healthy system.

So, paying per unit time seems to be the way to go. One last option I can think of to incentivize FLOT activity is to require another ‘per unit time’ activity that is actually the source of their pay. For example, posting an ‘I’m awake’ post once a day and not being paid for days the signer skips. Of course, shareholders still have oversight to fire a signer, however this will give a way for shareholders to measure how much signers are actively reading posts and making decisions, or at least that they’re visiting the forum.

[masterOfDisaster]

Do you currently see need for improvement or is this rather to make FLOT activities more SMART criteria compliant in the future?

[jooize]

One last option I can think of to incentivize FLOT activity is to require another ‘per unit time’ activity that is actually the source of their pay. For example, posting an ‘I’m awake’ post once a day and not being paid for days the signer skips.

Discourse tracks presence, but I don’t know if it can be used for this purpose. I don’t really like the idea of having to make a post every day.

https://discuss.nubits.com/users

[masterOfDisaster]

I don’t really like the idea of having to make a post every day.

I don’t like it as well and it makes no sense.
If a tx needs urgent action (and not the up to 36 hours window) FLOT members can be pinged and receive an email notification.
There’s no need to visit the forum daily or even make a daily post
But I still like the idea of making the FLOT more SMART criteria compliant.

[Nagalim]

Im coming at it from a theoretical point of view. A large part of the OP is the statement that we cannot use signer incentives for SMART criteria because it disincentivizes controversial decision making. That left me with the statement that the only thing we can base signer incentives on is some measure of activity per unit time. The issue of course is how to avoid sybil attacks of signers feigning activity using bots and scripts.

I think there is a distinct difference between a signer that actively reads threads and participates and seeks to create new beneficial transactions and a signer that only responds when called upon. I’m trying to find a way to incentivize the former over the latter.

[masterOfDisaster]

What about requiring to have all possible signers sign transactions - even if the minimum amount of signers has already signed a tx?
That’s at least close to the terms, which require the FLOT members to sign - or discuss why they won’t do that - within a certain amount of time.
That creates unnecessary (with regard to the ability to broadcast a tx) efforts for the FLOT, but helps finding out who stayed within the time frame.

[Nagalim]

require the FLOT members to sign - or discuss why they won’t do that - within a certain amount of time

I think we should keep thinking about the cost of that extra effort, but I definitely like that direction.

[dysconnect]

In many teams there are roughly 3 types of people that can be kept as long as their output can be afforded, from the lowest weight:

1. Those that choose to coast along but maintain the minimum level of responsibility, and don’t want to work harder even offered a rise.
2. Those that would perform at a level of their pay, and can do more for a pay rise.
3. Those that over-compensate for their pay.

Among flot, I’d say moD fits squarely in type 3. I would think I’m 3 on a good day, but overall not too far from being type 2. I refrain from commenting too much on others at this point, but they lie on a spectrum between 1 and 3. Even 1 is still a perfectly reasonable place to be because very few of us can commit a lot, and at least I don’t think FLOT members are anywhere beneath that.

It’s easy to deal with those that consistently over-compensate; you just give them more and more, well-paced with a carrot on the stick, and then they’ll catch up to the point that they can’t afford to visibly over-compensate anymore. Perfect slave to capitalism.

Type 1 people aren’t too hard to please either. We can ask people to be up front with it and then determine the minimum amount of work justifiable by a token amount of compensation.

Type 2 is the trickiest and I haven’t made up my mind with a low-friction way to deal with them.

[masterOfDisaster]

Those 3 types describe indeed a lot of people’s behaviour, and performance!
As long as the NSR holders find all FLOT members in range 1 to 3 everything is fine, right?
You can think of ways to improve the performance, but that would only be necessary if the current level of performance would be insufficient.
Unless you require the improved performance you waste money if you buy superfluous extra performance.

You should consider different types of motivation as well. Some people can be motivated with money (a type of extrinsical motivation).
Others might have other motivations (maybe intrinsical ones).
Most might have a mixture of extrinsical and intrinsical motivations.
Obviously you can influence only extrinsical motivations.

[Dhume]

It’s hard to say, when we started I felt the fee we’re charging might be a bit steep. However seeing how active FLOT had to be due to essentially having no functioning T3 layer combined with the BTC volatility I think it was justified. I feel T3 is improving and if this continues the demand placed upon FLOT should decrease, justifying a decrease in FLOT “salary”.

Along with that I would like to see the NBT group extended to increase all 8 members. Currently there’s not much to do for us NSR signers and I don’t see a scenario where this changes. In contrast this places a higher strain on the ones in the NBT group compared to the ones in the NSR group. I see no good reason why we couldn’t and shouldn’t change that.

I could even see a new division where we essentially decrease FLOT into a smaller team that handles the changes in demand and holds a smaller portion of the BTC and NBT funds in a 3 out of 5 address and a larger team that holds the majority of the reserves in a 5 out of 8 address (FLRT, First Liquidity Reserves Team). Time and activity requirements for the FLRT would be lower than for the FLOT team and payment should reflect this. Also it would be easier to require more active participation from the FLOT members since as a smaller team that’s their job, the FLRT is more of a backup and only requires the occasional (maybe 2/3 times a month) signing to restock the FLOT address when needed. However this would mean that 3 of the current FLOT members would need to do a step back from the current situation, something I’m not very enthusiastic about since I feel all have fulfilled their motion requirements.

[dysconnect]

Along with that I would like to see the NBT group extended to increase all 8 members. Currently there’s not much to do for us NSR signers and I don’t see a scenario where this changes.

If you do feel that way, perhaps you could have also responded to the recommendations I’ve made that the NSR group can help with non-signing tasks or participate in discussion by the NBT group, even though technical issues bar some members from being in charge of signing.

Perhaps it would have been better for me to ping everyone everytime I said it, but missing all of my several posts related to this is a sign of passiveness. I’m not dissatisifed with how FLOT has operated for most of the time, as I opined about commitment and also considering the size of the fee, but it’s not a good feeling to see it brought up this way.

[Dhume]

I merely meant that I don’t see a good reason why we currently have 5 members of FLOT assigned to NBT which clearly requires much more signing then NSR. I understand that those not in the NBT group can contribute in other ways but I don’t see a reason why we shouldn’t be added to the NBT group as well. Is there a good argument against a more equal sharing of the work (signing)?

[dysconnect]

You did talk about NSR group members sitting around doing less work, so my overall theme wasn’t entirely unwarranted.

As per your suggestion, currently one possibility is to create another 3-of-5 NBT group, and a similar feasible formulation is one single 3-of-8 group. Either is plausible; there may be discretion issues in these arrangements but certainly there is something to be gained. On the other hand, I vaguely remember a similar discussion in the past that never gained traction; it was about factoring the BTC group though, and there were too few participants in the discussion to develop a consensus. Perhaps there were individual concerns left unexpressed @jooize @ttutdxh @cryptog @mhps @woodstockmerkle

[jooize]

Two 3-of-5 groups would fragment some risk, I guess, and shouldn’t be worse (messier) than now if we use different signing threads. If we want all of FLOT to sign (or state their disagreement with) everything, two groups wouldn’t cause more work (perhaps a bit, and there’s some potential for confusion).

Is there any worry in having more potential signers with the same required amount of signers? Disregarding any potential risks, additional signers would make FLOT more agile.

Alternatives

• Single 3-of-5 (current situation)
• Two 3-of-5
• 3-of-8
• 5-of-8 like FLOT BTC
• 4-of-8?

Which of these are technically viable in the current state of Nu?

[Passed] Peercoin FLOT Reserve

It can be done, but it won't work very well until many nodes have upgraded, because older versions will not propagate these transactions (they will accept them within a block though). Another solution is to build a 2.0.x with only this change.

Did I respond with what you asked for?

[Dhume]

I’m not criticizing your interpretation merely refining my point.

My suggestion would be to install a 3 out of 8 address (only 3 signers warrants more speed). If we feel uncomfortable having 150k Nubits on a 3 out of 8 address, we could also keep the current 3 out of 5 address and use it as a “reserve” address. We then transfer say 30k to the new address and use that for day to day interactions with T3 and others. This way the whole of FLOT would be more actively involved with the balancing of both the buy and sell side.

[mhps]

Is there a good argument against a more equal sharing of the work (signing)?

What is equal? equal number of groups to belong to? equal number of signings per month? how is timeliness of signings put into calculation? If someone can come up with a calculator the FLOT can bill shareholders based on actual work done. Oh wait that is not good. Because it incentivises a FLOTmember to make a unneeded txs to get paid more. More thinking is needed.

install a 3 out of 8 address

That is fine with me.

Note: The reason why the NBT group has less signers was that T4 buy side BTC was greater in value than sell side NBT when FLOT was formed. Now there are $127k worth of buy side BTC and $153k of NBT. Either we should make the NBT group 5-of-8 or make the BTC group 3-of-8.

[Nagalim]

So what if we have 2 groups (to simplify @dysconnect’s 3 types of people and remove the questionable type 2). Level 1 FLOT members are paid like they are now and function the way they do now. Level 2 FLOT members must sign every Txn within their window (or give a reason why not) and are paid in a different way. The danger here, again, is the sybil attack. For example, if we simply pay Level 2 more than Level 1 then it would be profitable for a level 1 to design a bot that automatically signs every Txn that hits the forums. However, this would have to pass a Turing test provided by the forum goers and most basic attempts at level 2 farming would fail and end in ejection from FLOT and tarnishing their reputation.

Another thing to think about is giving incentives for forming new FLOT reserves, or tieing incentives to the magnitudes of the reserves. Again, we have to tread very very lightly here to avoid messing with the good faith of the signers and creating conflicts of interest.

[mhps]

So what if we …

What was the problem to be solved can someone remind me?

[Nagalim]

If we begin to consider performance incentives we need to be very cautious because we may influence the signer’s opinion of the best thing to do in a given situation by personally incentivizing one option over the other.

FLOT can bill shareholders based on actual work done. Oh wait that is not good. Because it incentivises a FLOTmember to make a unneeded txs to get paid more. More thinking is needed.

^That

[mhps]

If we begin to consider performance incentives

That started with an “If”. I don’t clear see a performance problem that warrants complicated mechanism to be installed.

I appreciate the enthusiasm. All FLOT members and you are active Nu community members. And that is part of the problem. I can’t help feeling we are all spending time polishing the shining steering wheel of a boat when the boat is having bigger problems in its direction. I’d rather see us talking about nubits adoption and a working profit model.