What If Everybody Did That? A New Standard for Delegation Bots

TL;DR: A good delegation bot rewards you today without devaluing your investment tomorrow.

Introduction

One fundamental difference between Thoth and the earlier generations of delegation bots is critically important for the Steem ecosystem. The reason for this can be easily understood by application of a simple philosophical and ethical test.

What if everybody did that?

At scale, this question delineates the boundary between growth and collapse.

I am aware that discussions of Thoth's incentive system tend to get long-winded and many people have been confused by them, so I want to step out of the weeds and look at the picture from a more holistic perspective. What is the background for the so-called "Everybody Test", how does it apply to the legacy delegation bots, and how does it apply to delegation bots that make use of an incentive system like Thoth's?

Philosophy's categorical imperative

The Everybody Test or the universalization test is a philosophical test that people often suggest as a way to guide human behavior. We describe a particular course of action that we might want to follow, and then we ask ourselves the question, "What if everybody did that?".

If it turns out that the result would be dysfunctional if everybody engages in the specified behavior, then we should probably consider following a different course of action.

This test is often applied in the areas of morals and ethics. Additionally, Immanuel Kant proposed it for philosophy in the form of his categorical imperative, which was central to his philosophical framework. In Kant's words, people should

Act only according to that maxim whereby you can at the same time will that it should become a universal law.

In modern English, this would be something like:

Only do something if you'd be happy for everyone else to do it, too.

Philosophers have applied this idea to demonstrate that things like theft, deception, suicide, laziness, charity, and even cruelty to animals should be considered immoral.

For today's article, however, I'm not interested in arguing about morals and ethics. There's also a practical nature to the Everybody Test, and that's what I'm interested in now. When people in a community don't let the Everybody Test guide their behavior, that's when various forms of economic dysfunctions, like the Tragedy of the Commons emerge.

In the Tragedy of the Commons, a community has access to a shared and exhaustible resource. It is in everyone's long-term interest to preserve that resource, but everyone's short-term interest depends on consuming some portion of it. Without proper discipline or controls, the short-term interest wins out and the resource gets drained. Examples of this have included shared grazing areas for cattle ranchers or depleted fishing basins for fishermen. In both cases, short-term incentives drove people to use up the grass or the fish before the other participants in the ecosystem might do the same. In Steem's case, the shared resource is the market cap of the blockchain itself, and it grows or declines according to the audience that Steem's creators attract.

As far as applicability, EthicsOps puts it like this:

We are all equal as ethical actors, so whatever is ethical for me must be ethical for others in the same circumstances. This is true for individuals and companies.

and:

(The Everybody Test) Addresses the free rider problem

What if everybody did that: Legacy Delegation Bots?

By "legacy delegation bot", I mean a form of delegation bot where an author delegates SP to the delegation bot and receives a daily upvote in return for the delegation.

If every SP holder were to make use of this mechanism, what would happen?

1. Authors would all post daily in order to maximize their ROI.
2. Authors would put as little effort as possible into each of their daily posts.
3. The attractiveness of Steem's content would deteriorate.
4. The value of STEEM would fall in proportion with the decline in eyeballs.
5. In the end, the value of STEEM would not be high enough to retain witnesses, and the blockchain would collapse.

I suppose it will not surprise anyone if I claim that the legacy delegation bots fail "the Everybody Test".

The people making use of these services are free-riders (not a value judgement, just an observation), and they can only succeed for as long as there are enough other people to create content that attracts visitors. As soon as the percentage of organic authors drops below some unknown threshold, Steem enters into a death-spiral. It's a classic Tragedy of the Commons.

What if everybody did that: Delegation bots like Thoth?

When I say "delegation bots like Thoth", I'm talking about a delegation bot that does not require (or even encourage) daily posting from its delegators. The bot directs passive rewards towards delegators and also to the authors of the posts from the past and present that make it through screening and AI evaluations. It's important to note that I'm not thinking of one centralized bot... I'm thinking of a competitive ecosystem where bots compete for delegations and upvotes by picking out the most attractive content and delivering the best mix of rewards for authors and delegators.

Now we ask the question, what if every SP holder delegated all of their holdings to a generation of delegation bots like Thoth?

1. Authors would continue creating articles that are sufficiently attractive to pass Thoth's screening and LLM evaluations.
2. Vacuous posting from delegators to legacy delegation bots would disappear.
3. Each new post has a potential for lifetime rewards, so authors would be motivated to post regularly, but not so often as to degrade attractiveness.
4. Delegators who don't particularly want to be authors would have no need or benefit for posting daily.
5. The overall attractiveness of Steem's content would improve.

I think it's clear that Thoth and (hopefully) future bots that harness similar incentive structures pass "the Everybody Test".

Side by side

Here's a comparison table that shows how these technologies fare against the "Everybody Test":

Feature	Legacy Bots (Fail)	Thoth-Style (Pass)
Requirement	Daily posting (Low effort)	Attractive content (Higher effort)
Market Impact	Floods Steem with "noise"	Curates "signal"
End Result	Ecosystem collapse	Sustainable growth

Conclusion

In the end, we don't need to understand Thoth's technical details. All we need to understand is that it represents a fundamental paradigm shift in delegation bots.

The legacy bots fail the "everybody test" because they are dysfunctional at scale, and they can only survive as dependents on the creators who produce content that attracts eyeballs. In contrast, I believe that Thoth passes the test because it creates a form of interdependence among the content creators, the delegators, and the operator of the bot - a sort of a "mutual aid society".

If we want Steem to succeed, then, delegation bot clients need to insist that their service providers move to this new paradigm. They don't need to adopt Thoth, necessarily, but as stewards of their clients' delegations, they have a fiduciary obligation to protect the value of their delegators' investments. They can do this by redesigning the bots for "everybody test" compliance. Thoth has proved that it's possible, so inaction should not be an option