New BSIP or BSIP24 discussion: stake lock-up mechanism, count only real "locked" stake as voting stake

[abitmore]

Update: @grctest wrote BSIP-0024 - Locking Bitshares away as 'Bitshares Influence' for voting privileges on the BTS DEX back in 29/08/2017 which is similar idea.

-- my opinion wrote here --

IMHO, for a healthy ecosystem, decision making on important things should be serious and relatively stable, so governance voting should only be allowed for long-term self-owned stake only, and interests should be aligned.

• stakes in balances can be liquidized / transferred / sold at any time, allow voting with balance means instability, may negatively impact decision making, because everyone can obtain (buy) some stake and participate in and influence decision making (voting) at any time immediately, then sold them at any time immediately when no longer want to participate, with no consequence;
• utilized stakes should be considered as "used" but not "being owned", thus stakes in collateral or market orders should not be counted in;
  • for stakes in collateral, it means debts exist and the positions can be liquidized in various ways, thus the stakes are not 100% owned;
  • for stakes in market orders, it means they're for sale, which showed unwillingness to hold;

To reduce influence caused by random stake movements, it's a good strategy to adopt some kind of stake lock-up mechanism.

• long-term stake holders might be willing to lock up their stakes for a period;
• people who have influenced decision making (voting) should take responsibility, especially if negative consequences have been made, they should not be able to sell their stake quickly enough when the price started to drop;
• one controversial topic is, should people be allowed to suddenly start influencing decision making with a big stake, especially if said decisions will invalidate earlier efforts made by other stake holders. IMHO it's better to minimize influence of this behavior, e.g. requires a holding period for voting.

The vesting mechanism used in Steem, namely VESTS, has some behaviors described above. However, VESTS in Steem is being heavily utilized, thus IMHO is not a good base for governance voting.

My proposal:

• stake holders can lock up a part of their stakes, to gain corresponding voting power
  • only locked-up stakes can be used to vote, don't count stakes in balances/orders/collateral
• voting power accumulates by time, generally, stakes locked longer have more voting power
  • cap voting power in some way to avoid endless accumulating
• stake holders can start unlocking their locked-up stakes partially or fully at any time
  • unlocked stake will go to balance after a relatively long period, say, 3 months;
  • immediately clear accumulated voting power when started unlocking

More things to think about:

• If really want to cash-out, people can sell their accounts with locked-up stakes as a whole.
• Legal issues?

Thoughts?

[abitmore]

Related to #81.

[TheTaconator]

My main concern with the proposition is for the stake in collateral. It seems unfair to those holders to limit their influence on voting especially due to the value that their collateral brings to the ecosystem.

Maybe a compromise is to only count collateral stake if it is older than some age. But that might be difficult to track especially as the holdings of an account change with time.

[svk31]

I like the idea a lot.

Like in Steem having parts of your stake locked up offers additional security since there's a time delay between being hacked/losing the private key to the thief being able to actually transfer funds out. I do think 3 months is excessive and is too big of a disincentive to people actually locking their stake. A shorter period like 1-3 weeks seems more appropriate to me.

I'm not sure accumulating coin days is really necessary as it complicates the system unecessarily imo, why not keep it simple? If you want to prevent immediate voting with newly locked stake, simply make it a linear accumulation until the max value of your staked coins over the inverse period of the release delay, so 1-3 weeks in my example above.

Another advantage of this would be to remove voting power from exchanges.

I agree with taconator that collateral should be counted, but I'm fine with stake in open orders not being counted.

[abitmore]

EOS has a 3-day lockup mechanism and many exchanges are voting.

[startailcoon]

I really think this could be a good addition to the chain. It would make a good incentive for users to keep and lock their assets.

BTS locked in collateral could be counted as locked assets. Even if it could be liquidised its also a risky business while making good volume for the bitAsset as well. It could be considered a good thing that the asset is used, and thus should be counted.

Open orders aren't locked, since they are for sale, and shouldn't be counted.

The time involved to unlock assets should be long enough to make it something that should be thought about a second time before doing. If its to easy to unlock, like a mear few days, large exchanges could be able to plan around this anyway. I feel it should be at least 4 weeks to unlock all of the assets locked, or the value you choose.

Another possibility is that all BTS has voting, like today, but that locked BTS has 10x the power, or similar.

[grctest]

I wrote BSIP-0024 - Locking Bitshares away as 'Bitshares Influence' for voting privileges on the BTS DEX back in 29/08/2017 - similar idea?

[abitmore]

@grctest thanks. OP and title updated.

[xeroc]

Great discussion. I'd like to add a few remarks with respect to what I have learned dealing with the opinion letter.

The opinion letter makes heavily clear that BTS is a utility token that is used for governance purposes but can be used for other (smart contracts) as well (read: collateral).

If we were to require a powerup to enable governance features I see the following major issues:
This changes the previous deal which has potentially been used by investors to decide to buy in. Given that BTS is much more decentralized than STEEM (there is no 'steemit inc.') this might open up the possibility for class action suites against a) the proxies who approved that change, and b) the witnesses who applied the change. Given that I am not a lawyer, I cannot say how to evaluate that risk, but we should at least be aware that this change may cause investors to be unhappy. Fortunately, a lockup (I assume) will only be for short period of time and not 6 months. Also, if this proposal makes it thru, we need to educate and announce this change way ahead of its implementation on chain to give sufficient time for investors to get the hell out of BitShares :D

Also, for this change to execute, we need a soft-migration on-chain To ensure that we don't end up in the situation where only little stake votes.

Can we make a list/table about where BTS are 'stored' and how to best deal with them, please?

• Balance: Not voting
• Powerup: Voting
• Collateral: I would argue this should be voting because it is not quickly liquidated in case the BTS valuation goes down. Also, people have been asked to go short if they really want to support the system and we would take away their voting right now. Doesn't feel right.
• Open Orders: not voting
• Vesting: ?

[xeroc]

Another possibility is that all BTS has voting, like today, but that locked BTS has 10x the power, or similar.

I like this idea .. Kind of removes the legal trouble above.

[clockworkgr]

Worth adding to the discussion as reference:

http://bytemaster.github.io/article/2016/01/04/The-Benefits-of-Proof-of-Work/