Ok. lets say you are poorfag with 20k Link. You dont want to do Linkpool. You want your own nodes.
Do you do 20 nodes with 1k Link each?
4 Nodes 5k Link?
2 Nodes with 10K Link?
Has anybody done any math on this? Sure its very theoretical but potential income will very wildly im sure.
Yes im drunk but im not drunk user.
You literally dont even need LINK to run a node
Depends on how profitable/competitive each node is. If you're the only node pulling info for whatever particular thing, then you will obviously need less LINK.
One node.
You can run multiple jobs as long as you have liquidity to stake on a contract.
The bottleneck is processing power.
If you only have 1k nodes, you will never get to bid on the big contracts that will pay the best.
why don't you ask the team? nobody here will have an answer for you
how passive is managing a node? Can I leave it while i am wagecucking?
>poorfag
>20k Link
What?
Please answer this
It's a free market - so we won't know until mainnet drops and there's demand. As we compete with each other to provide the service the market will settle on positions for various amounts of risk absorbed by the staked amount. I suspect that running a few nodes that're low cost with a small stake will be enough at first, simply because a large chunk of traffic will be smart contract startups operating in a shoestring. Then there'll be a large gap to a few Enterprise-level projects that'll use giant nodes like Linkpool. As the market fills put the middle range will too - but by then we should all know how to price it better. Just my 10c of course, nobody really knows shit yet.
Yes, you'll want to monitor it regularly but you'll be fine to keep wagecucking and do it on breaks (Ssh in on your phone if need be). You can set up a node right now on the Ethereum testnet. Go give it a try, learn how it all works. You can get free Testnet Linkies from a faucet to play with. Like most passive income it's not really 100% passive, and of course part of it is absorbing some of the security risk for your customers, but it's not going to be 8 hours per day or anything. More like a little microbusiness that'll hopefully give enough to make wagecucking optional.
In terms of the work we will have to do to set up a node. When it comes to feeding information. Do I have to go find information like wether data, stock ticker data and such and feed that to my node? How do I do all this and will there be a /link/ general with all the major links to set up a decent node or are people gonna hide shit?
Or is that note the case at all, and all I need to do is run one and it automatically fetches data it needs?
Also how long before I can live off my nodes profits and quit working?
It's in everyone's best interest to not tell you how or where they get their data, so yes there probably won't be guides on anything except very general shit.
Also we're all probably going to be disappointed with the demand for data when mainnet finally releases. This is going to take a lot of time for people to trust.
So basically setting up a good node that earns a lot is going to be quite hard, or take a lot of research.
What if a company wants to perform a smartcontract but there are no nodes that currently know the revelent data? What if the government wants to do some smart contract with some other government agency in a different country? How the hell are we gonna get a data feed on that? What about for projects under NDA or classified stuff?
>You dont want to do Linkpool. You want your own nodes.
You could still use Node as a Service from Linkpool. It costs a fixed amount of Link per subscription time.
It's automatic. General API functionality comes working out of the box. However, certain external adapters won't necessarily be included, for example SWIFT has no reason to share its payment system for free, although they could offer access to their external adapters for a subscription fee or something like that. That's just an example, I don't really know if SWIFT will or won't ship their external adapter code in ChainLink nodes.
Unless you have some idea for an external adapter that is unique, you really have no reason to write your own code for node functionality. Just let it do its things for the most generalized API calls.
Oh that's what I was hoping for. Because if I had to find all those data streams myself I'd have no idea what to find or how to set it up.
Still, don't you get punished for feeding bad data to a smartcontract? Is that only if you set up your own external adapter and whatever info you fed to it was wrong. If you just leave it as default it wont fuck up and take your Linkies will it?
Basically external adapters are used for accessing authenticated API services, like for example if you wanted to use a smartcontract to trigger a trade on Binance, you would need the Binance account's API keys to initiate the trade with a ChainLink node. The external adapter would need to be able to accept the account's credentials as parameters to make the API call to Binance to initiate a trade. You would have to write that code yourself or hope another person writes it and shares it for everyone else. Then, you can offer Binance trades through your ChainLink node. If you're the only person who can offer that functionality... then why share that external adapter? That makes you the only person who can answer that API call, thus guaranteeing your use in that scenario.
You get punished for sending bad data, yes. The point isn't that you have to intentionally provide trusted data, the point is that you don't sabotage the data they request. They choose the API data they want by visually parsing the json responses from the data source. If you play around with the Make a ChainLink on smartcontract.com you can get an idea of how it works.
>What if a company wants to perform a smartcontract but there are no nodes that currently know the revelent data?
If there's no data the contract can't be made.
>What if the government wants to do some smart contract with some other government agency in a different country? How the hell are we gonna get a data feed on that? What about for projects under NDA or classified stuff?
It won't happen through a decentralised set of nodes on chainlink. So they will either just trust their own private node to send the correct data, or not even use chainlink and they will just trust their own datafeed.
>So basically setting up a good node that earns a lot is going to be quite hard, or take a lot of research.
Yes. This is a business with money involved. There's a reason why people keep trading strategies close to their heart, it keeps them profitable. If you had a very accurate datafeed on something that no one else had but many people wanted, you could charge much more for the data. If everyone else gets it then the profitability quickly goes to 0 because people start undercutting everyone else.
>don't you get punished for feeding bad data to a smartcontract?
You only get punished if they prove it's bad data.
This is true but also being the only node with a particular bridge results in centralization so it will be beneficial to have multiple nodes have your external adapter to allow for decentralizion otherwise no one may want to use your node
Yeah.. because if you offer fiat it'll market buy link for you.. much better to get more link while it's cheap so you have collateral
Thanks Thomas
bringo. it's gonna be an interesting balance of having enough nodes / decentralization to satisfy smartcontract creators but not so many that everyone and their mum can run your adapter on their node
Some tasks only require one initiator. A smartcontract wouldn't use three nodes to send a single offchain payment or to place a limit order on an exchange.
ChainLink is versatile in that it can serve both decentralized oracle needs and "centralized" oracle needs. It just depends on the job.
That doesn't discount what I said before at all. If you have a closed source adapter running only on your own node it will likely be used less than if u are running multiple nodes with said adapter or give access to it to others either for a price or for free. Decentralization in all transactions is always better than a single point of failure
Where's the logic in that? A smart contract developer needs a certain API job done, looks at the list of useful nodes, sees only one choice, then says: "sorry guys this contract can't be done there's only one node that can complete the task that we need to be done by only one node anyway. Cancel the contract!"
20k on one node.
>optimum rate before diminishing returns
If the transaction is worth a large amount why would they trust it? Why integrate chainlink at all if they're not looking to decentralized the api call? The point is it may not be best to keep close source adapters in your sole possession.
Just sell your LINKs. You'll have 20m.
The banks and corporations aren't going to use your neet nodes.