Jump to content
anichhangani

Incentivising running validators

Recommended Posts

I have looked at a lot of threads to find out why running validators is not incentivized "monetarily", but have failed to see any reasoning of why it is that way, although there are a lot of answers wrt to what are the incentives to run a validator namely, the users wanting to make sure the network stays healthy and so on, but I feel the XRP ledger can reach a higher level of decentralisation if the community votes for the incentives mechanism.

There are real costs of running a validator and although I want to run it, can't because I don't have the necessary resources. If the incentives mechanism is in place I am sure there will be a lot of nodes popping up and it could basically change the whole landscape for XRPL. 

The incentives can be made in the following ways from the top of my head, I am sure there might be many efficient ways, but what do I know:

  • An escrow account can be set up to release the huge amount of XRP's the company owns to people who run validators, basically 1 XRP / x TXN's.
  • By providing fees equivalent to or a percentage of whatever XRP is burnt. This way the asset remains deflationary but incentivizes the validator operator to at least offset/break even the cost of running a validator by this fees. One downside maybe the txn costs increases, but that won't be by that much...(hoping).

Although ripple as a company has dedicated itself to decentralize the XRPL much further in 2018, but what I have seen is that the validators are being mostly run by institutions which have vested interests and although that is good, I also want to see validators run by individuals and if such incentives are in place I totally believe more and more people will start hosting validators.What are the long/short term implications of doing this, Also, what are the pros/cons? Also I think its technically possible, but still I am not sure.

PS. I am a total newbie, so probably I am making a complete fool of myself here :)

Also, would like answers from @JoelKatz  and @nikb

Thanks

Edited by anichhangani

Share this post


Link to post
Share on other sites
13 minutes ago, anichhangani said:

Also I think its technically possible, but still I am not sure.

Please describe this mechanism you have in mind in more detail. How much should I earn for my "validator of love" (https://xrpcharts.ripple.com/#/validators/n9LoveB8ztAxoJYbL49zrPp6e5KJqKXCqVrte3FskXLuRJMzYxEf) for example and how would the process look like?

Share this post


Link to post
Share on other sites

I am also in favor for this idea. This would help further in distribution of xrp and more validators, which is good for overall network availability of the ripplenet.

Share this post


Link to post
Share on other sites

While I agree its a may be good idea in principle, what's difficult, and what's mired discussion in the past is the following conundrum:

Any incentive based on validator performance is also an incentive to attack/spam the network to achieve the moneys attributable to such performance, and yet, how can validator performance be suitably rewarded without creating such a dichotomy?

For example, rewarding validators for the # of txn's validated will incentivise validators to collude to create/validate huge numbers of transactions in order to claim larger and larger amounts of any available fee's or escrowed proceeds.

So, while validators being expensive to run limits expansion, it also protects the network against bad actors.  Currently, in order to run a validator you have to have a real, tangible business-oriented reason to do so.  It must provide a reward in and of itself that counters the expense.  In that sense, its a good thing for the integrity of the network as it encourages only those who receive significant direct benefit from some legitimate aspect of running a validator to do so.

Therefore, to me - aside from the question of whether or not its a good idea to reward validators, there is the more pressing question of how it might be done without disrupting the integrity of the network.  Personally, I think its a great question to try and figure out an answer regardless.  Resolving this kind of conundrum may have many potential applications.

(There's also one other concern - it's been previously stated its not in Ripple's interests to grow the network significantly.  I remember David specifically saying a while ago that 15 - 20 validators was an ideal number for that moment, when someone suggested increasing it.  Perhaps that position is changing now the network is mature enough, but I still suspect they may wish to have tight control over the process of expansion for the same reasons they wanted to limit expansion in the first place.  Being able to easily adapt the network to potential future changes/updates without much fuss was the main reason IIRC.)

Share this post


Link to post
Share on other sites

I don't think that incentivizing validators by having the Company compensate operators in XRP helps to meaningfully make the validator network both more robust or more distributed. In fact, I think it's a bad idea.

If people run a validator only to get compensated, then I wouldn't trust them because I'd worry that they would be willing to collude against me if someone paid them to do so. Validator operators need to have a stake in the long term health of the network, not in a monthly paycheck. Presumably the compensation would be enough to making running a validator profitable (otherwise, why do it?), in which case the plan could be easily gamed: simply setup multiple validators, under multiple personas.

I am not against operators seeking to defray their costs. If someone wants to operate a validator and publishes a policy that explains how they'll operate it and then chooses to solicit donations to cover the expenses associated with running the validator, I have no problem with that. A program, operated by the Company, to compensate validator operators? Sorry, I don't like that.

Although, I'll be honest: the marginal cost of operating a validator for an entity that already operates at least one server is minimal. Even if you're a user and want to host a validator in your home, the cost is likely pennies a day.

Share this post


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

I don't think that incentivizing validators by having the Company compensate operators in XRP helps to meaningfully make the validator network both more robust or more distributed. In fact, I think it's a bad idea.

If people run a validator only to get compensated, then I wouldn't trust them because I'd worry that they would be willing to collude against me if someone paid them to do so. Validator operators need to have a stake in the long term health of the network, not in a monthly paycheck. Presumably the compensation would be enough to making running a validator profitable (otherwise, why do it?), in which case the plan could be easily gamed: simply setup multiple validators, under multiple personas.

I am not against operators seeking to defray their costs. If someone wants to operate a validator and publishes a policy that explains how they'll operate it and then chooses to solicit donations to cover the expenses associated with running the validator, I have no problem with that. A program, operated by the Company, to compensate validator operators? Sorry, I don't like that.

Although, I'll be honest: the marginal cost of operating a validator for an entity that already operates at least one server is minimal. Even if you're a user and want to host a validator in your home, the cost is likely pennies a day.

Sure for pennies... nothing is for free.

Thats why you have PoW & PoS. They are the gamification within the crypto world.

Share this post


Link to post
Share on other sites
On 2/8/2018 at 5:32 AM, Sukrim said:

Nice vanity address! You are awesome, @Sukrim.

I personally don't favor incentives (other than soliciting donations from non-Ripple sources). As @nikb indicated, there are a lot of independent operators. I imagine that, as xRapid progresses, this number will include more institutions.

 

7 hours ago, nikb said:

If people run a validator only to get compensated, then I wouldn't trust them because I'd worry that they would be willing to collude against me if someone paid them to do so. Validator operators need to have a stake in the long term health of the network, not in a monthly paycheck.

I also worry that incentives ultimately increase vulnerability and lower long term health.

Similarly, I think that incentives create a false illusion of a viable network. Right now we see real people deriving real use who want to support the network. I can't say with confidence the same is true for PoW or PoS coins.

Share this post


Link to post
Share on other sites
7 hours ago, nikb said:

I don't think that incentivizing validators by having the Company compensate operators in XRP helps to meaningfully make the validator network both more robust or more distributed. In fact, I think it's a bad idea.

If people run a validator only to get compensated, then I wouldn't trust them because I'd worry that they would be willing to collude against me if someone paid them to do so. Validator operators need to have a stake in the long term health of the network, not in a monthly paycheck. Presumably the compensation would be enough to making running a validator profitable (otherwise, why do it?), in which case the plan could be easily gamed: simply setup multiple validators, under multiple personas.

I am not against operators seeking to defray their costs. If someone wants to operate a validator and publishes a policy that explains how they'll operate it and then chooses to solicit donations to cover the expenses associated with running the validator, I have no problem with that. A program, operated by the Company, to compensate validator operators? Sorry, I don't like that.

Although, I'll be honest: the marginal cost of operating a validator for an entity that already operates at least one server is minimal. Even if you're a user and want to host a validator in your home, the cost is likely pennies a day.

Is there any type of, say, novice set up for validators? 

Is there anyway for someone who isn't programming literate to run one for the hell of it? 

What are the baseline minimum requirements both in hardware and knowledge of software to run one safely?

Share this post


Link to post
Share on other sites
57 minutes ago, Hodlezerper said:

Is there any type of, say, novice set up for validators? 

Is there anyway for someone who isn't programming literate to run one for the hell of it? 

What are the baseline minimum requirements both in hardware and knowledge of software to run one safely?

@WietseWind on twitter (sorry, don't know his name in here - he made xrptipbot.com) has docker images for running validating/nonvalidating nodes: https://hub.docker.com/r/xrptipbot/rippledvalidator/

I have developed a video tutorial with written instructions (still need to edit, post, and organize the last three videos) here: https://rabbitkick.club/rippled_guide/rippled.html
I literally screen recorded myself deploying and securing a CentOS 7 install, enabling validation, and then viewing the new validator's status on the Ripple Charts site.

Running any rippled node takes some amount of technical expertise, however, it is certainly manageable for someone w/o any programming skills. For most people, I imagine they would  need a non-validating ("stock") node, as running a validating node requires trust building and doesn't really offer any benefits over running a non-validating node w/o trust building.

@Sukrim and others have offered me invaluable support while I was learning to set up nodes, and there are a lot of community resources.


Minimum hardware requirements are a solid state hard drive (size doesn't really matter, unless you need to store a lot of history) and 6+ GB of memory. Rippled uses minimal CPU processing power. To run a node safely, I think that having some experience with Linux (Ubuntu, Debian, CentOS) would be very helpful. One would likely need to know how to SSH into a remote server, setup a firewall, and transfer files to/from a remote server using something like rsync or scp.

Gaining trust for a validating node would probably require more technical expertise, as up time and security are important.

Edited by Rabbit_Kick_Club

Share this post


Link to post
Share on other sites

If the validators could run on some sort of Rasberry Pi like system,  I would run one in each of my apartements. Anyone know of any Pi clone with hardware capable of tackling it?

Financing an easy-install plugin for Synology and Qnap could also be an idea. I guess the hardware in most of those units currently aren't up for the task either though.

Edited by mandelbaum

Share this post


Link to post
Share on other sites

These are far too underpowered and I would also question the abilities and commitment of the operator if the validator is just installed via a random script on minimal hardware. In the future it will be potentially necessary to manually or semi-automatically ensure that your UNL is good for example. If you just run a non-validating node, there's not much harm if it isn't perfect since you could re-join the network easily. If you are validating, others might depend on you (and they should, otherwise there wouldn't be much point in validating).

You can take a look at http://up-shop.org/4-up-boards by the way, if you need to have a very small system that can run rippled.

Share this post


Link to post
Share on other sites

This is a topic of great interest to me, too. The key here is perhaps finding a proper views of the indirect benefits and promote it.

Agreed with @mandelbaum's concept, make it easy for supporters to contribute as honest validator nodes. Enthusiasts are willing to contribute to good causes like Folding@home or SETI@home because the amount of resources contributed is voluntary, though these are more for humanitarian achievements.

update: there's Docker image available and it's definitely a step towards the above.

Edited by ripplex

Share this post


Link to post
Share on other sites

Create an account or sign in to comment

You need to be a member in order to leave a comment

Create an account

Sign up for a new account in our community. It's easy!

Register a new account

Sign in

Already have an account? Sign in here.

Sign In Now

×
×
  • Create New...