• Forum has been upgraded, all links, images, etc are as they were. Please see Official Announcements for more information

Variable Amount Proposal - A new budget proposal type that request a % of the unawarded treasury

Bitedge

New member
Background

Large amounts of the treasury going unawarded every month is a missed opportunity to use those funds to dash's advantage.

At the same time we reject the idea of approving lower quality proposals or putting the unawarded funds in a general "rainy day fund" because the dash should be earnt and accounted for. We dont want the wasteful government mentality of having to spend all the funds or having guaranteed fee money because these devalue dash and result in misallocation of resources.

The concept of variable amount proposals for a % of the unused treasury solves both of those problems :)

What is it

A new type of budget proposal that works in the same way as other proposals except the amount is a stated % of however much dash is unawarded for the treasury cycle.

So if the proposal is for 50% of unawarded dash and there are are 10 dash unawarded it would get 5 dash, if there are 1000 dash unawarded it would get 500 dash and so on.

Advantages

Allow us to make take more complete advantage of the monthly treasury to help dash.

It avoids falling into the wasteful government mentality of having to spend all the funds because these proposals could easily be rejected if they are not advantageous to dash.

No dash would go to proposals that are not specifically approved so there is no guaranteed free money and there is still a real probability of the treasury not awarding all funds in a cycle.

Disadvantages

Time and effort to write, test and implement code and documentation (like any change to the system).

Added complexity for users to the proposal and treasury system (like any new feature).

Examples

Please don’t get hung up on if you like these examples or not, If these proposals would not be advantageous to dash they would not be approved, just like standard proposals.

The point of this threads is to discuss the concept of proposals for a % of the unawarded treasury. The examples are to show that some proposals would not suffer from receiving a variable amount each month that is not known ahead of time.

Example 1
Variable spend google adsense campaign
50% of unawarded treasury


Set up a google adsense campaign targeting the optimal key words and type of users for dash.

Let’s say the cost will be 10c per click. In a month where this proposal gets $1000 worth of dash we buy 10,000 clicks. In a month where this proposal gets $10,000 worth of dash we buy 100,000 clicks or whatever it is.

The varying amount is not a problem for this proposal.

Example 2
Variable contribution to Dash Core general savings account
50% of unawarded treasury


I gather the dash core team already have something like a savings account they manage allowing them to pay consistent UDS denominated wages in times of dash price swings. This can go to any areas of the dash core project with accountability and transparency.

This proposal would better fund that account, and again, the variable amount is not a problem, just something they would manage and budget for.

What’s it called

"Variable Amount Proposals"

Okay I might not have come up with a good name yet. Other literal names could be

“Proposal for remaining treasury”
Or
“Proposal for unawarded treasury”

Let us know if you can think of a better name.

Feedback please

Please let us know any ideas to improve the concept, any further advantages or disadvantages you can think of et cetera.

Thanks for reading :)

Cross-posted to reddit https://www.reddit.com/r/dashpay/comments/61nbe0/variable_amount_proposals_a_new_budget_proposal/
 
Last edited:
A while ago the core team would made a bunch of "catch-all" proposals that would be up-voted just enough to pass so as not to get in the way of real proposals and would go to something like a "core savings account" - it would actually go to attend conferences they didn't expect to have the money to go. My guess is, if it was easy to code they would have done something like it already.
 
Back
Top