This topic can cover techniques/ideas for steaking
Sounds delicious, but I think you mean staking. I’ll remove this comment if you want me to, or if you make the correction.
I think staking has great potential, as long as it’s not the only engine of consensus on the network. If it is the only way to have a vote, it would get prohibitively expensive for most regular or new users. Staking with time, spending it volunteering for the community, may be an alternative. All alternatives could tie in to a standard trust protocol. But we need immutable ID’s before we can really get there I think.
That looks like a master node to me
Sorry to correct you, @jirusu, but in the official “steaking” documentation, masternodes are defined as follows:
Just wanted to make sure no misinformation is spread on our trusty new forum. Thanks for understanding.
Lol yeh its OK leav it up.
I’m not ashamed lol.
Thanks man, I’ll try to interpret my understanding as best I can.
please don’t hesitate to correct me.
And yes I know my English is awsome.
That’s how you own it like a boss
(steaking sounds right because there is chance of loosing what u have stacked)
A while back I was having a conversation with Tim about which agent he would be steaking on. The idea of stacking on a contracted agent seems like a really good strategy.
P. S. I do apologise if I am overstepping my mark on this topic.
Master node? I thought that was curator
Don’t worry about that master node thing. Dash has it for example. It’s just a name for a role in a consensus machine.
It’s definitely the strategy I will be taking. We need to analyse which services the AI themselves are sub contracting to and then stake those.
We are hodling so we can be steaking, right?
My question: what happens when I stake, the node is successful but later on there is a new node added to SNET with much better functionality so the older one I staked would no longer be useful?
A bit of both, the AI contracted originally will be able to put to tender or outsource an element of the service requirements that it alone cannot manage.
What the user sees is a list of services that fulfills their requirements ranked by cost and reputation.
It’s the users choice which they choose.
Why not stack on the service that is offering the sub contract?
I think the key will be to analyse if there are any nexus agents that multiple services use. The more revinue that agent generates, the more reward pool percentage is open to those that stake them.
I brought it up in the telegram channel before but maybe someone else can explain what is in place to safeguard against “stake pumping” a node the same way people agree to pump something on an exchange? I think it was talked about the value of your staked tokens decreasing if the node under performs but it won’t matter, it would be exactly like pumping a token on exchanges. As long as you make some $ quickly who cares if the node sucks. They’d just dump it/stop staking t and move to the next one.
I would have thought staking doesn’t influence the reputation, so if ppl pump a stack they are just throwing tokens away.
So when you request or search for a service on SingularityNET, what comes back is a list of available services that fits with your requirements.
These services are ranked by popularity, reputation and staked value.
Staking is meant to be a way to steer toward ethical services btw.
In your example, what the user sees is a service reputation values being out of line, very high stake but perhaps the popularity or reputation is not in line .
It’s then up to the user which service to chose to use.