I've thought a solution like yours before and I don't really like it. One issue is that it doesn't let people capture the value they added to a name. The other value proposition of this DAC is that you can finally *own* a domain, not just have it leased to you by ICANN.


The goal is not that "your" domain is available (not being 'squatted'), just that it is being sold by someone who wants to sell it to you for a fair price.

These incentives just make it more profitable for you not to squat and simply hold nameshares unless you know that you can flip a domain for *higher than what the market currently prices it at*. There is still room for "squatters", but these are more like real estate companies. They want to sell their space and will eat their losses as soon as it becomes obvious they made a bad bet.

There's another discussion to be had about whether to limit the number of concurrent auctions near the start when it is not clear that the .p2p namespace is worth anything at all, and whether to keep it limited to some larger number once it is just to make it so that humans can go through and evaluate all the domains for sale at a given time.

I disagree, I think domain names are basically "the commons" and should be treated as such.  This puts the domain name into the hands of the people who can maximize it's value... I think that's more efficient for everyone.   You can still profit from a domain you added value to by subleasing it.  If you have your own domain that isn't something real generic, no one has a reason to bother you or bid on it and you'll have it cheaper than something from godaddy.

I also think your current plan has big squatter problems.

 

Quote from: Agent86 on June 28, 2014, 06:03:20 pm

Quote from: emski on June 28, 2014, 05:47:32 pm

If the majority (51%) adjusts the rules at will they can set share price at will.

I'm not understanding this

For example in the scenario where 51% while controlling 101 delegates adjust transaction taxes for the remaining 49%.
This should automatically lower the prices of ALL shares.
And enables "the 51%" the ability to increase taxes exclusively for "the 49%".

Ok, so that basically makes everyone's shares instantly worthless, until a new chain/fork is started with the 51% stake removed so trust/integrity is restored.  Why would the 51% do that to themselves?  How do they gain anything?

Yes It will be obvious but there will be no incentive for controlling party to resolve the disagreement.

There is HUGE incentive to resolve disagreement.
If a majority group tries to abuse their power, one of 2 things happen:
-The minority sells their shares
-They fork the network without honoring the unreasonable majority

Both of these actions leave the majority with a GIANT loss of value.  How does this unreasonable colluding majority profit??

Their shares will be worthless and there will be no "new minority" willing to buy shares and get abused.

A DAC is a FREE ASSOCIATION of people who's interests are aligned (growing the value of the shares.)
I have no idea how you expect one party to gain 51% of the network at GREAT expense and then just blow it with bad decisions for no reason.

Quote from: Agent86 on June 28, 2014, 02:46:18 am

You also seem to be advocating that a minority should be allowed to slow down the network for everyone else if they don't like things.  How small a minority should be allowed to do this?  If you pick 33%, why that number?  Sometimes when you've just picked a number that "seems about right" to you, it can help to think about it more.

Yes. The number can be adjusted...
I believe that delegate representation for minority (above certain %) groups should be granted.

Adjusted how?  Adjusted to what?  If you adjust it back to 1% you have the same system we started with and just replaced - (you're back at square one)
If you wish to propose a change that is taken seriously you must articulate an actual proposal, not just "it can be adjusted."  You must specify something and defend its logic.

I'm very interested in this topic, but feel I missed most of the context.  Is there a source of informaiton where I could learn more about what all this means to Bitshares?  What is the current methods that are being used?

And what kind of deligation is there?  I like infinite delegation, where anyone can delegate their votes to anyone.  Making large hierarchical trees with significant power to move an large organization on a dime, out performing traditional hierarchies, yet of the leader screws up, his delegated tree of power will vanish instantly.

Is anything like that being used for this?

To the extent I've considered it, large hierarchical trees are very susceptible to vote buying / influence peddling.  Approval voting provides an incentive structure that allows representatives to put the interests of whole organization/DAC ahead of a constituency.  I think it works well for our application.

This is a good idea but it is only meaningful if it can be properly weighted by stake. So you'd either need to have this site read the chain and require signatures from stake.. or we can just build it into the client =D

I think a way to do this would be very useful and fun.
I would LOVE it if it was possible to register "proposals" or other statement, in the same way that we register IDs, and then everyone could vote up or down on it (weighted by stake with transactions).  Of course it is completely non-binding and is really just a pol of shareholder views.  I think BM may have implemented something like this that is only available to delegates?

The thing holding me back from pushing it is that it seemed to be another example of being able to vote on something that is "unbounded" because there wouldn't necessarily be a limit to how many things you could vote on.  I don't see why this is impossible but it seemed unbounded voting was an initial objection to approval voting.  Also if your voting profile was tied to your stake it could create even more of the kind of privacy issues that bytemaster was trying to avoid.

Are there ways to do it that get around large data requirements/blockchain bloat or privacy issues?  Can you do it "off chain" but still have it verified and weighted by stake?  Maybe we need "anonymous" voting similar to the voting DAC with zerocoin tech that you guys discussed.

Not so fast, my friend. "Dilution" in the VC world is actually a form of financing, i.e., capital-raising. That means the VC gets shares for providing money to the company company. Often, anti-dilution clauses are put in place as a safeguard to protect early investors. This is very different I think from the dilution in a crypto chain that you have in mind.

You can try to draw a distinction but the result is the exact same.  New shares are issued, prior shares are diluted, funds raised from the sale of new shares are used to pay company expenses.

I'm not sure if you understand what a standard anti-dilution clause says:
"An anti-dilution provision is a clause in an option, security, or merger agreement that gives the investor the right to maintain his or her percentage ownership of a company by buying a proportionate number of shares of any future issue of the security."
Guess what, under my plan you get a built in anti-dilution provision! Lucky for you!  All shares are sold on the open market and you are free to buy them so that your total stake is not diluted!

Your proposed "algorithm" would be very inefficient and wasteful. It defies economic principles. The fact is, a diffuse set of thousands or millions of shareholders doesn't have the information and unity of objectives to efficiently make lower-level employee pay decisions. Why doesn't the median voter in Apple decide the salary of the CFO or a senior vice president? Because it would be terribly inefficient for shareholders to coordinate on a business decision like this, leading to waste and agency problems. That type of decision is best left to management.

"The salary of management is a decision best left to management"  ... I'm not sure where to start with that one.

You have many times said I am conflating day-to-day operation with oversight.  This is not the case.
Even though I called it "DAC employees" in many ways it's analogous to CEO/upper management/BOD, I just don't like those terms for a number of reasons I won't get into.  Call it what you want, but these are people that have broad community trust and can in turn delegate or pay other contractors, less critical employees etc.  Quite likely a new crypto may only have one "DAC employee":  The core dev.

You say my algorithm defies economic principles…  Why, please be specific and elaborate?

Setting aside shares upfront and raising capital are but two examples of tried and true ways. Do you think there is a reason that these two methods have been commonly used in the capital markets for decades, while outright dilution has not?

Completely and provably false. see post: https://bitsharestalk.org/index.php?topic=4713.msg61701#msg61701

There is another problem with dilution strategies that has not really been discussed. That is the following: who will decide exactly how the funds are allocated and used? Which employees get the spoils, and how much they get? Who will decide how much to spend, and on which projects? Which marketing initiatives or infrastructure projects should get priority?

Yes, as I said in my previous post, a smart dilution algo with the right incentive structures is not an immediately obvious problem to solve... it's "not easy"  And that's why it's such a big deal that BitShares is on the verge of cracking the nut.

I submit that the following algo works and creates the right incentives:

Employees are elected by "approval voting."  Along with your "approval" you indicate an appropriate annual salary in bips (gets paid out daily).  Only an employee with over 50% support of stake ends up paid.  Their salary is the median voted by stake (people who did not vote for the employee are included in the median as voting for a "0" salary).

Any "inactive stake" (no transactions for over 1 year and paid inactivity penalty) should be removed from the voting algorithm.  Stake may voluntarily "abstain" and thus remove their stake from the voting algorithm.
https://bitsharestalk.org/index.php?topic=4660.0

It will be governance by the masses

uh... it's governance by the shareholders... the only governance that makes any sense.

It seems that you think that a diffuse group of shareholders can coordinate effectively on the best way to invest and to compensate employees. That seems awfully naive.

yea... except that's how every single company works. 

Every company is ultimately governed by the shareholders; the shareholders elect the board of directors, they can fire the CEO and vote on his comp plan.  They can hire people to make informed decisions on their behalf but the "the diffuse group of shareholders" are ultimately the ones who voted in the leadership and call all the shots.

Investors are capital-providers. They are not managers.

Wrong again.  If you think your job is done as soon as you write the check, in most cases you can kiss that money good-bye.  You, as the shareholder, are responsible for your investment.  You abdicate this responsibility and power to your own peril.

You seem to somehow acknowledge that investors are smart enough to find a capable trustworthy developer (such as bytemaster) and fund this developer.  And yet as soon as they do this they all of a sudden become incompetent idiots who can never be trusted to make a smart decision again.