Author Topic: DAC development incentive models  (Read 7378 times)

0 Members and 1 Guest are viewing this topic.

Offline bytemaster

I think that having some hard-coded limits is in place.  Nothing stops the delegates from hard-forking to get around those limits, but it would take a far more concerted effort.  The difference between passing a new bill in congress and passing a constitutional amendment. 

Market always wins, so our goal is to move with the market not against it.
For the latest updates checkout my blog: http://bytemaster.bitshares.org
Anything said on these forums does not constitute an intent to create a legal obligation or contract between myself and anyone else.   These are merely my opinions and I reserve the right to change them at any time.

Offline Troglodactyl

  • Hero Member
  • *****
  • Posts: 960
    • View Profile
...
I at least am in agreement with you on that, but the time released shares your suggesting are still preallocated and limited, which is the key point for me.
Trog, not sure what you mean by "preallocated" but I think bytemaster was essentially saying the opposite:  time-release allows shareholders to make allocation decisions later as the picture on how to best allocate it becomes clear.  So the funds are not "preallocated"

By "preallocated" I meant that a finite predetermined quantity of shares is allocated to be distributed gradually through the delegates.  If the quantity has a hard upper limit, it makes no technical difference whether those shares are described as "new" diluting shares or as a preallocated delegate trust fund.  The preallocation is to "the delegates" generically, leaving the specifics of distribution to individual delegates more flexible.

Offline Agent86

  • Sr. Member
  • ****
  • Posts: 471
  • BTSX: agent86
    • View Profile
I think the #1 thing that everyone here is missing in the pre-allocation vs time-release allocation is that under the time-release the shareholders have an opportunity to change who receives the funds and how much after getting to observe performance.   With pre-allocation you are effectively stuck with one development team and a fixed fee.

So from my perspective an allocation of 20% AGS/PTS and 80% delegates over the life of the DAC is greatly preferable to 20%/80% preallocation.  It is certainly more flexible, keeps the majority of the pre-allocation off of the market  and gives AGS/PTS holders 100% control over the allocation of these funds without violating the social consensus of 10/10/80

In fact the only thing that I think needs to be decided is the release schedule for the 80%.
I agree that defined time-release schedule is better than preallocation.  I can also see how this is more "understandable and palatable" to many current market participants.  Maybe in the same way that we started with a mined proto-coin for share allocation; it made sense to the current market participants even though it was wasteful.  I mean people still love counterparty proof of burn …

I think this could be a compromise "transition system" while people get used to it and start to understand how this works better.  They'll start to like having a say in allocation and realize their fears of secret colluding majorities don't happen. 

In the long run a hard limit at 20% or defined rate of dilution has no real justification. I think we could probably skip the transition even if it means losing appeal to some people until they see the results.

Some people may disagree but I think the "industry" we are working on has multi-trillion dollar potential.  Lets keep that in mind when thinking how far we can predict the need for growth/development/re-investment.  The idea that we can pre-decide the right amount of dilution we need and when we'll need it to get this whole thing wrapped up in a neat bow is not convincing.  If we don't allow for it, dilution of your stake in the industry will just happen in other ways, such as the introduction of more new chains and competitors.

Do you honestly think if you got a 1% stake of the industry now and then just park it in a cold wallet that you will wake up 20 years from now with a 1% stake of a new multi-trillion dollar industry with all these people working their butts off to increase the value of your "non-dilutable" shares?  Sorry to pop your bubble but that's not how things work.

I at least am in agreement with you on that, but the time released shares your suggesting are still preallocated and limited, which is the key point for me.
Trog, not sure what you mean by "preallocated" but I think bytemaster was essentially saying the opposite:  time-release allows shareholders to make allocation decisions later as the picture on how to best allocate it becomes clear.  So the funds are not "preallocated"

Offline Troglodactyl

  • Hero Member
  • *****
  • Posts: 960
    • View Profile
I think the #1 thing that everyone here is missing in the pre-allocation vs time-release allocation is that under the time-release the shareholders have an opportunity to change who receives the funds and how much after getting to observe performance.   With pre-allocation you are effectively stuck with one development team and a fixed fee.

So from my perspective an allocation of 20% AGS/PTS and 80% delegates over the life of the DAC is greatly preferable to 20%/80% preallocation.  It is certainly more flexible, keeps the majority of the pre-allocation off of the market  and gives AGS/PTS holders 100% control over the allocation of these funds without violating the social consensus of 10/10/80

In fact the only thing that I think needs to be decided is the release schedule for the 80%.

I at least am in agreement with you on that, but the time released shares your suggesting are still preallocated and limited, which is the key point for me.

I think for completely new development with no live network, AGS was brilliant for providing time released dev funding, and allowing the community to watch development progress and fund more or less depending on promise, rather than going in blind.  Preallocated time-released shares can do the same thing for a network post launch, with the added benefit of being able to divert funds to competing groups.

There still may be some delegates who take the pork barrel spending approach, but the impact is more limited.  Before the network reaches near saturation the shareholder's interests should all be closer to aligned anyway.

Offline bytemaster

I think the #1 thing that everyone here is missing in the pre-allocation vs time-release allocation is that under the time-release the shareholders have an opportunity to change who receives the funds and how much after getting to observe performance.   With pre-allocation you are effectively stuck with one development team and a fixed fee.

So from my perspective an allocation of 20% AGS/PTS and 80% delegates over the life of the DAC is greatly preferable to 20%/80% preallocation.  It is certainly more flexible, keeps the majority of the pre-allocation off of the market  and gives AGS/PTS holders 100% control over the allocation of these funds without violating the social consensus of 10/10/80

In fact the only thing that I think needs to be decided is the release schedule for the 80%.
For the latest updates checkout my blog: http://bytemaster.bitshares.org
Anything said on these forums does not constitute an intent to create a legal obligation or contract between myself and anyone else.   These are merely my opinions and I reserve the right to change them at any time.

Offline amatoB

  • Full Member
  • ***
  • Posts: 62
    • View Profile
I have another idea of how to handle it... We can make a DAC that honors PTS/AGS 50/50 with no dilution, then a new DAC or version of this DAC is quickly proposed that allows dilution.  A delegate is created (or other stake voting mechanism) and announce that in 2 months there will be a snapshot and every stake/share that is approving this delegate at snapshot time will be honored in the new chain with dilution.  All others not approving can honor themselves in a separate chain without dilution.


My first impression is that this would be viewed as a coercive mechanism. It would certainly be divisive.

Regarding dilution, I do see that it might have some advantages over a one-time share allocation. Some people here argue that dilution could be good for fast-growing DACs with very uncertain future funding needs. But for these cases, a pre-arranged schedule of dilution (which is probably the only one that most shareholders would ever find palatable) doesn't seem to offer much additional flexibility anyways.

So, really, what can a pre-set schedule of dilution accomplish that a pre-allocation can't? It may be that dilution can be useful because it provides a "vesting" feature--not all of the funds are made available upfront, so there is less danger that a developer can walk away with 30% of the DAC. I think that is a valid point. However, pre-scheduled dilution has its own drawbacks. It is like paying someone in the hopes/expectations that they will do what's good for the network. I can see that there might be a lot of inefficient spending in such a system. At least with a pre-allocation, an honest group of developers or delegates can take a bounty-like approach where they can make sure they don't get much less than they paid for.

One thing that would make me a bit uncomfortable as a PTS/AGS investor is that the dilution rates that have been discussed here seem quite high to the point of disenfranchising early investors. For example, many investors probably bought into AGS, PTS thinking they would get a 35/35 allocation in some of the first round of DACs. By changing to a 35/35 allocation with dilution, they now are effectively getting 15-20%. Perhaps that will eventually go to 10%. Where is the original 30% pre-allocation to non-investors going to go? If high rates of dilution are being considered, then perhaps there should be fewer (or no) shares pre-allocated. So, maybe it's a question of magnitude. If the benefits from dilution greatly exceed the amount of value transferred, then that would be great. But, as some here have argued, the true benefits (above and beyond pre-allocation or additional fundraising) have yet to be proven.

The pro-dilution folks seem to ignore (or at least downplay in importance) that dilution can give a negative perception to shareholders. And that is perhaps the chief disadvantage of dilution. It makes the optics look bad and, rightly or wrongly, is easily misperceived by shareholders. The only way to get maximal acceptance and investor interest is to make sure everything is clear, transparent, and palatable to shareholders from the start. Perceptions are really important. There should be nothing to suggest that someone in power can game the system to extract or misuse wealth. I believe that misperceptions are one of the things that crippled Ripple early on. Many criticized Ripple heavily and labeled it a scam on account of the fact that they retained about 95% of the supply upfront, even though they promised to distribute 50% or so to grow the network. (Of course, Ripple was also closed-source initially, but most of the criticisms seemed to center around the initial and planned share distribution.) Rightly or wrongly, investors balked at what was perceived to be a grossly unfair system. A concern I have is that if shareholders in a DAC feel wronged and get sufficiently PO'ed, scrutiny and intervention by a regulator could become more likely, which is something that probably few people here want...

Whatever solution is chosen, I hope that developers always remember that every DAC is different and that public perceptions do matter--they matter a great deal. Let's learn from the experience of Ripple and others. Simply put, the optics have to look good from the start, otherwise a DAC will have a hard time gaining widespread acceptance and investor interest.

Offline Agent86

  • Sr. Member
  • ****
  • Posts: 471
  • BTSX: agent86
    • View Profile
So as long as some AGS holders have at least a little PTS, they can issue new shares to AGS holders only and take away most of the PTS based stake?
Trog, there's no secret illuminati AGS society that coordinates in secret without PTS holders knowing.  Where the hell do you get these ideas?
...

That was a quick example, and no secret societies are required.  The point is that in a direct democracy with unlimited power, you're very likely to end up with tyranny of majority and oppression of minorities as soon as a possible choice comes up that reveals that not everyone's interests are perfectly aligned.
Yes, they are required; you just haven't thought it through.  anyway I may be afk for a bit, I will see where this stands later.

Offline Troglodactyl

  • Hero Member
  • *****
  • Posts: 960
    • View Profile
So as long as some AGS holders have at least a little PTS, they can issue new shares to AGS holders only and take away most of the PTS based stake?
Trog, there's no secret illuminati AGS society that coordinates in secret without PTS holders knowing.  Where the hell do you get these ideas?
...

That was a quick example, and no secret societies are required.  The point is that in a direct democracy with unlimited power, you're very likely to end up with tyranny of majority and oppression of minorities as soon as a possible choice comes up that reveals that not everyone's interests are perfectly aligned.

Offline Troglodactyl

  • Hero Member
  • *****
  • Posts: 960
    • View Profile
I would only consider a 49/49/2 (pre dilution) allocation fraudulent if the initial allocation was advertised to investors, while the fact that this was not the final real allocation because of dilution was left out.  When you asked if I would prefer 10/10/80 without dilution or 49/49/2 with dilution, the final distribution was omitted.  Without stating the final distribution, or even the maximal rate of any additional distribution, the 49/49/2 initial distribution is completely meaningless.  It's all marketing and no substance, which is a good indication of fraud.

Actually I said with "49/49/2 with reasonable dilution".  I think we can reasonably agree what I meant by that without arguing details.  You're now insisting on final allocation.  Ok....  but I did at least address that the dilution wasn't just something fixed rate without bounds.  You've ignored that though or don't remember it.  I think this points out how much people lack objectivity when looking at these things.

Sorry, I was overly harsh there.  I agree that preallocated/limited dilution can be a good way of funding ongoing development and marketing in the early stages.

Offline Agent86

  • Sr. Member
  • ****
  • Posts: 471
  • BTSX: agent86
    • View Profile
So as long as some AGS holders have at least a little PTS, they can issue new shares to AGS holders only and take away most of the PTS based stake?
Trog, there's no secret illuminati AGS society that coordinates in secret without PTS holders knowing.  Where the hell do you get these ideas?
For the record, are you in favor of a fixed limit to the dilution rate, or completely discretionary dilution?
I wouldn't care that much if there was a cap that seemed reasonable.  In principle I don't believe in the need to babysit shareholders and have no problem with completely discretionary dilution.

Offline Troglodactyl

  • Hero Member
  • *****
  • Posts: 960
    • View Profile
I've added two more options to the poll, marked with "EDIT1".  None of the existing options were altered.

Offline gamey

  • Hero Member
  • *****
  • Posts: 2253
    • View Profile
I would only consider a 49/49/2 (pre dilution) allocation fraudulent if the initial allocation was advertised to investors, while the fact that this was not the final real allocation because of dilution was left out.  When you asked if I would prefer 10/10/80 without dilution or 49/49/2 with dilution, the final distribution was omitted.  Without stating the final distribution, or even the maximal rate of any additional distribution, the 49/49/2 initial distribution is completely meaningless.  It's all marketing and no substance, which is a good indication of fraud.

Actually I said with "49/49/2 with reasonable dilution".  I think we can reasonably agree what I meant by that without arguing details.  You're now insisting on final allocation.  Ok....  but I did at least address that the dilution wasn't just something fixed rate without bounds.  You've ignored that though or don't remember it.  I think this points out how much people lack objectivity when looking at these things.

« Last Edit: July 05, 2014, 10:38:09 pm by gamey »
I speak for myself and only myself.

Offline Troglodactyl

  • Hero Member
  • *****
  • Posts: 960
    • View Profile
Meh, I don't even care to argue my points as I've said them elsewhere.
I can see how you can start to feel that way.  At some point it's not your job to convince everyone.
For everyone who is opposed to this concept after reading the many posts explaining it.  There is an easy solution:  when DACs that have this property are released PLEASE PLEASE PLEASE sell your shares.  Then we are both happy!!  I actually believe in empowering shareholders to make decisions and put their fate in their hands.  And for this reason, I am actually a little particular who my co-owners are.  I wouldn't want to run a business with co-owners that I have to constantly explain simple concepts to and who are going to make bad decisions that affect me, or who will vote to sit on their hands when we should be reinvesting in growth.

I also want to go on record as being in favor of dilution with no bound: dictated by business needs and shareholder vote;  no limit @ t=infinity.

If a DAC starts off 50/50 PTS/AGS but those original shares end up representing less than 20% in the long run that was a decision made by the shareholders.  We must allow our thinking to evolve with what is right and not box ourselves into corners because some people like to bitch about "pray we don't alter the deal" whenever a development in thinking happens.

So as long as some AGS holders have at least a little PTS, they can issue new shares to AGS holders only and take away most of the PTS based stake?

For the record, are you in favor of a fixed limit to the dilution rate, or completely discretionary dilution?

clout

  • Guest
Investing in a developer and his dream before there's a working product or any guarantee of value is not free riding.  If the funding is all in advance, the investors are trusting the developer to be honorable and see the project through, and they are taking considerable risk.


Exactly !  Considerable risk.  With dilution this risk is greatly reduced.

Not sure why you would think 49/49/2 with dilution means you'd being defrauded.  You are not looking at it objectively.

I also think a developer will put far more time into a DAC than random shareholders.  Expecting voluntary contributions to fund the thing just introduces freeriders which I despise.  That is a whole mess in itself. 

shareholders are not freeriders. whether they directly fund development or not. by holding their shares instead of selling them they give everyone else's shares more value.

Within the context of expecting continual development/marketing support to be paid by donations or volunteers there will be freeriders.  That seems to be the most popular alternative I am seeing - a voluntary donation towards development. 

We can come up with advanced voting systems and offices being elected etc that would likely be better than either.  Thats assuming the developer wants such a thing.

No they are not free riders... by holding shares they make sure the share price doesn't drop which means fewer shares buy more development support. This is one of the reasons why the argument can be made that bitcoin holders should receive a pre-allocation in dacs. If it had not been for the increase in the bitcoin price from the fall of 2013, I3 would not have raised as much money as it has for the development of the bitshares_toolkit.

Offline Agent86

  • Sr. Member
  • ****
  • Posts: 471
  • BTSX: agent86
    • View Profile
I have another idea of how to handle it... We can make a DAC that honors PTS/AGS 50/50 with no dilution, then a new DAC or version of this DAC is quickly proposed that allows dilution.  A delegate is created (or other stake voting mechanism) and announce that in 2 months there will be a snapshot and every stake/share that is approving this delegate at snapshot time will be honored in the new chain with dilution.  All others not approving can honor themselves in a separate chain without dilution.