Jump to content
pucksterpete

To infinity and beyond! Ripple will only go to the moon when node validators are compensated

Recommended Posts

Thanks for sharing @pucksterpete. I had never really thought of this as a serious option until...

10 hours ago, JoelKatz said:

Incentivizing validators as long as is needed is not off the table.

With a bit of thought it makes perfect sense that it should be an option. In harmony with the Reddit author's post and the advantages listed above, here are a few reasons I believe a program like this could yield disproportionate advantages:

1. Incentivizing validators creates a new group of stakeholders. While PoW has its issues, I've always thought that one major benefit it creates is a group of organizations and individuals with a vested interest in increasing the price of the token in which they are compensated. 

2. It lowers the cost of experimenting with Ripple. We all have to make decisions every day on how we spend our most precious asset, time. Incentivizing validators would lower the monetary and time related opportunity-cost of spending time to understand and experiment with Ripple.

3. While the technical skills required to run a validators are likely minimal, they are nonetheless technical and as a result attract technical individuals with skills that are potentially useful to the community. By being compensated in XRP and growing in their interest and understanding of Ripple's unique qualities, I don't think it's a stretch to assume that some portion of those who test drive Ripple while running a validator, will find that they like the amenities and decide to stay (I'm no expert, but this is likely a run-on sentence). Remember those free burritos Chipotle gave out in the earlier days? I still eat there, so they worked. 

4. The obvious advantage is increased decentralization. I believe this would come at a very reasonable price. Imagine 250 validators compensated 500 XRP per month each, assuming satisfactory performance (approx $35k/mo at today's prices). Seems like a great return for the validators and for Ripple. 

5. I'm even less of an expert in tax policy than I am in grammar but I believe this compensation would provide an attractive tax write-off for Ripple as well. 

6. While this type of incentive wouldnt replace liquidity incentives, it would  be easier to verify. As validators are submitted a corresponding Ripple address could be included. As long as the validator meets certain performance levels, the XRP is sent. 

7. It's low risk. Participants would be required to submit additional info to satisfy all AML, KYC, and XYZ requirements. 

In general, I think this would help Ripple gain some of the communal advantages of PoW without the inherent drawbacks of adopting it. 

Share this post


Link to post
Share on other sites

My worry is that if Ripple is paying lots of people to run validators, it's hard to see that as real decentralization. What we want is lots of people and organizations that care about the network, are real stakeholders, and run a validator because they want a say in the network's evolution. But the perfect should not be the enemy of the good. Anything better is on the table, IMO.

Share this post


Link to post
Share on other sites
26 minutes ago, nikb said:

FWIW, I think that if you need to be "incentivized" to run a validator, you probably shouldn't be running one.

These individuals might not need the incentive to be financialy viable though the incentive may provide the motivation needed to participate. 

26 minutes ago, nikb said:

My servers (including my validator) will never trust a validator that is being financially incentivized directly.

How would you know the validator is incentivized? Would a grant to the MIT blockchain lab count as an incentive in your book? Most importantly, what makes you more wary of a financially incentivized validator than one that receives some other advantage by participating (assuming the paid incentive is based on some given performance requirement and the "other advantage" is not directly monetary in nature, like having say in how the network evolves)? Part of the beauty of consensus though is the ability to chose.   

P.s. Thanks for adding your thoughts. Always makes for a thorough conversation (which I miss around here).

Edited by Xi195
Detail added for clarification

Share this post


Link to post
Share on other sites
36 minutes ago, JoelKatz said:

My worry is that if Ripple is paying lots of people to run validators, it's hard to see that as real decentralization

Valid concern. I think I finally found an ideal to throw back at the BTC purists. 

36 minutes ago, JoelKatz said:

What we want is lots of people and organizations that care about the network, are real stakeholders, and run a validator because they want a say in the network's evolution.

Getting more people into the ecosystem increases the likelihood of attracting such individuals/organizations. Maybe the incentive could be revolving or diminishing and limited to a certain number to help avoid this type of free-loading. 

36 minutes ago, JoelKatz said:

But the perfect should not be the enemy of the good. Anything better is on the table, IMO.

It's nowhere near perfect but it's already getting better. PoW with all the disadvantages has a few advantages that I believe this scheme could help co-opt. 

Edited by Xi195

Share this post


Link to post
Share on other sites

By the way to preface my interest in commenting on this thread, I am less interested in the "to the moon" portion and more interested in discussing validator incentives as a means to increase overall adoption. 

The common "freemium" model is drawing some kind of reverse parellel in my mind though I can't quite articulate it yet. 

Share this post


Link to post
Share on other sites
8 hours ago, pyskell said:

Let's not kid ourselves, rippled isn't resource intensive and I can't imagine the bandwidth is much either (someone feel free to correct me).

Running a node with full history requires about 5 TB of SSD storage and will waste multiple TB of bandwidth until you are synced (in 2018 or 2019). Also you can expect a few TB of bandwidth on top of that for continuous operation, if you want to connect to more than a handful of nodes.

Since rippled by default is configured for relatively short term online_delete and very few node operators seem to change the defaults, it means nodes are constantly (a few times per day) re-syncing their full state after throwing out history. A few hundred MB each time.

Run a node yourself, before you make such blanket statements please.

 

About the topic: I heavily disagree with handing out hosting coupons or incentivizing node operators in any other way. If you only distribute destroyed XRP, this likely runs into the issue where people will happily volunteer e.g. cleaning up a road, but will not do it any more if you pay them 50 cents an hour on top. If subsidies are substantial, then operators are not independent. If subsidies are not substantial, they can be 0 as well.

Rippled needs better documentation and best practices for setup and operation, not money thrown at "cloud" providers to pay for hosting.

Share this post


Link to post
Share on other sites
1 hour ago, nikb said:

FWIW, I think that if you need to be "incentivized" to run a validator, you probably shouldn't be running one.

My servers (including my validator) will never trust a validator that is being financially incentivized directly.

Well, your validator is indirectly financially supported by RL because they give you a salary,. stocks, etc., they "give" you a knowledge which is needed for reliably running a validator,  they give you enough freedom so that you can run validator in your free time, etc. Your validator is actually heavily subsidized.

 

Share this post


Link to post
Share on other sites
1 hour ago, Xi195 said:

Most importantly, what makes you more wary of a financially incentivized validator than one that receives some other advantage by participating (assuming the paid incentive is based on some given performance requirement and the "other advantage" is not directly monetary in nature, like having say in how the network evolves)? Part of the beauty of consensus though is the ability to chose.

I'm thinking aloud here, so my concerns will likely be invalidated by better reasoning but here's the deal:

Say we have a system running that has validators who appear to be validators without having a precise, legislated, guaranteed direct monetary incentive to be validators. Why they are participating is unique to every validator and each one may have a different underlying incentive, belief, commitment to participate or they may share the same. Let's presume that this  fuzzy incentivization paradigm has some properties that restricts "growth" or at least the potential growth of the pool of validators.

Now, a new theoretical model or belief is widely adopted that:  it is the growth of the pool of validators (or greater decentralization) that is directly proportional to the growth of the value of a token associated with the network (it's native digital asset). While this may or may not be true, it is implemented in the system and now the validators create via consensus a new paradigm where they get some direct monetary benefit of being validators.

Let's presume that now the number of validators have more than doubled making the original set of "fuzzy logic" incentivized validators a minority. Now the new set of validators decide for example that they want to be paid directly in USD or some other asset that doesn't ensure that the network's native token isn't valuable at all. Since the new validators likely joined the system for direct monetary benefit, they are now incentivized to guide the network and it's properties in a direction that ensures that they keep getting at least whatever they believe is a fair piece of the action. They may now choose to ignore the potential value or growth of the native digital asset associated with the network because they can simply accumulate a greater share of it (knowing full well that they don't have to care for the future of the network in the extreme long term because they never held any significant native assets or they never cared about those "fuzzy incentives" like technology, faster payments etc that original validators did) and ultimately cash out of it at anytime they'd like.

Does that sound like a nightmare? It does to me.

 

EDIT: 

Basically: if majority people are purely doing validation because it directly "pays", then there is little incentive to grow the underlying value of the asset as a form of "payment", because the vast majority of new validators will likely never hold an "excess" of the underlying asset, creating at least an equal incentive to want to accumulate a greater share of the total assets as opposed to working to increase it's underlying value. This is in stark contrast to the majority stakeholders who gain little by accumulating more of the assets but rather are much more interested in growing the underlying value of the asset (Example Ripple Labs)

So, to a random validator holding 1000 XRP at 1 USD makes little difference  to holding 10,000 XRP at 10c, except 10,000 looks like a greater stake. If you hold 60 Billion XRP of 100 billion however, you can't possibly be interested in "increasing your stake" and would exchange ownership of stakes if it meant a large increase in underlying value of asset.

Edited by hesque

Share this post


Link to post
Share on other sites

I think that if the incentives are enough to pay let's say 80-100% of the cost of running a validator, no validators will pop to gain money and screwing everything up (because the incentives are less than the cost), but it will encourage people that cannot afford the full cost of running a decent validator (100-200$ /mo).

Edited by tulo

Share this post


Link to post
Share on other sites
4 minutes ago, tulo said:

I think that if the incentives are enough to pay let's say 80-100% of the cost of running a validator, no validators will pop to gain money and screwing everything up (because the incentives are less than the cost), but it will encourage people that cannot afford the full cost of running a decent validator (50-100$ /mo).

Gotta say, due credit - this is indeed a brilliant idea. "subsidization" of cost to promote healthy ecosystem. People don't get to make profit, but if they want to run a validator, they just get a helping hand in reducing their costs. 

Share this post


Link to post
Share on other sites

Thanks for raising the subject. It appears that the validator network is crucial.

Some brief and naive questions from someone who is primarily interested in XRP and other cryptos as an investment / object of speculation:

- What is the current status of that validitor network, and is it possible to verify it for "outsiders"?

- Can normal people become a "validator" - just like people running a Bitcoin node - or would banks themselves assume this role?

- How long will it take for this network to reach the minimum size / distribution / other characteristic?

What I am trying to get at: Are we looking at a central enabling aspect that requires a lot more work and time to become a reality? Could this explain the reservation on the part of investors?

Share this post


Link to post
Share on other sites

×