This is the End of the Link Token FUD

This write up vastly understates the value of the LINK token. Yes, the LINK token can be used by smart contract owners to pay Chainlink nodes for getting data. Yes, LINK is used as part of the reputation formula for node operators and will factor into how the contract listing service rates them.

HOWEVER, most importantly, LINK can (and will be) used for data request penalty payments to ensure that node operators provide the requested data. Penalty payments are LINK tokens that are required to be held in escrow by the smart contract. They are paid to the smart contract creator in the event any of the node operators do not meet the required data requests as stated in the smart contract. This provides an incentive for smart contract creators to trust node operators, knowing that they have a form of financial insurance (the penalty payment) in the event a node (or nodes) submit bad data.

Other urls found in this thread:

discord
twitter.com/SFWRedditImages

For information that will trigger high value smart contracts, smart contract owners will want to require a proportionate amount of link to be held in escrow as penalty payments by the node operators. When link is tied up for penalty payments, it is released over the life of the contract. For example, let’s say party A wants an API snapshot sent every day for 30 days. If the penalty payment for the contract is 300 LINK (per node operator), then each node operator will have 10 LINK released to them at the end of each day – receiving the full 300 LINK at the end of the 30 days if they successfully performed the data request the smart contract asked for. Now imagine the smart contract creator wanted 10 node operators. That means 3000 LINK is taken off the market immediately, and 100 of that 3000 is released each day from the smart contract to the individual node operators (10 each per operator, assuming they provided the requested data). A cycle will be created where more and more smart contracts will make requests and node operators will be limited only by the availability of their LINK tokens to be used for penalty payments.

Add it all together and you have a singular payment method for a desired network (the most secure external data oracle), lots of supply constantly locked up to have enough link for signaling purposes (the reputation boost for a node operator), financial insurance for smart contract creators (penalty payments) for increasingly valuable triggering data in a wide variety of smart contracts, and a network posed for growth as more adapters are built and more API’s become available so that dapps can thrive on any blockchain network. Yes LINk is an ERC20 token, but it is blockchain agnostic and the adapter network can continue to grow.

1.

Smart contracts are basically pieces of software that can oversee and execute a trade, without any third party having to press the “trade” button. So if we want to trade, say, 1000 LINK for 10 ETH, a smart contract can “look” to see that we have both put up our ends, and when we have, it automatically performs the swap. It’s great because we can never rip each other off (the swap only happens when both sides have fronted) and we don’t need a third party to do the swap for us and take fees.

But smart contracts are limited at the moment because they can only perform swaps for very specific things, and that is, things that are coded in terms the smart contract can understand. For example, an Ethereum smart contract can trade LINK for ETH because it can “see” both of those things on the Ethereum blockchain. But it hits a wall when we want to trade, say, ETH for a concert ticket. The concert ticket isn’t coded in a way that Ethereum can natively understand, so there is no way for the smart contract to know that it is a legit, non-cancellable, version.

This is where oracles come in. An Oracle is a company which has recently appeared in the smart contract space, whose job it is to translate data into a form that a smart contract can understand. So, you could hire an Oracle to program your concert ticket in a form that could be read by the smart contract, so that the smart contract could properly execute a trade when you have fronted the ticket, and they have fronted a couple of ETH.

2

But there’s a pretty serious problem here. If you’re paying a private company to translate your data into a smart contract readable format, then what’s the point of even using a smart contract in the first place? The whole point was that we didn’t want to have to trust a middleman in order to make our trade, but now we’re using a trusted middleman and both of us have to believe that the Oracle is doing an honest job. The whole reasoning behind using a smart contract is wasted, because if we’re going to be paying private Oracles to translate data, we might as well just pay a private mediator to oversee the trade and take a fee.

This is where ChainLink comes in. ChainLink creates a decentralised network of oracles so that the entire process of smart contract execution is trustless. Instead of trusting a single private oracle to translate the data honestly, we are now using a trustless network of nodes to move information on and off blockchains. This is massive, but to get an idea of how massive, you need to think bigger than concert tickets.

Big business, and I mean pretty much any big business, will love smart contracts. Billions if not trillions of dollars a year are spent on exchanging value, and huge money is paid so that the value is exchanged properly and accountably and fairly. Smart contracts automate the process of value exchange and therefore offer billions and billions and billions in savings to companies that use them. But companies aren’t diving all over them yet because of the problem listed above. If a number of entities want to perform large exchanges of value, but are all using their own private oracles, then the trustless exchange is not achieved and the smart contract is pointless. ChainLink, and ChainLink alone, offers up the possibility of bringing end-to-end trustless smart contracts to the world, and this is a revolution so profound that at this point it is almost impossible to see the future extent of it.

3

But is ChainLink the project that will do it? You might have been reading FUD about 2 man teams or absentee CEOs or shit like that, but here’s what we know.

The ChainLink team were the only blockchain related team to be invited to present at SIBOS two years running, and in their proof of concept this year they showed that the ChainLink network could automate messaging for certain aspects of SWIFT’s bond processing. Sergey’s EOY announcement mentioned SWIFT by name twice in giving examples of the sort of work that ChainLink could do, and the work that was ongoing. While the market seemed to want to hear the word “partnership”, this was proof of a close and ongoing relationship. SWIFT manages messaging standards for more than 11,000 banks worldwide.

Also, ZeppelinOS recently announced a partnership with ChainLink. ZeppelinOS stands to be the base development layer for future dApps and smart contracts on the EVM (Ethereum network). They provide a standard interface for dApp/SC development and will use ChainLink as their default oracle service. This means that the vast majority of new projects built on Ethereum will use ChainLink in the background without even knowing it.
And this is just the beginning. ChainLink is probably the most profoundly anti-hype project in the crypto space right now. They have said that they will let their partners announce all partnerships, and they just want the tech to speak for itself. If ZeppelinOS and SWIFT aren’t enough, the EOY announcement hinted at ongoing work with everyone from small SC startups to the absolute monsters of fintech and insurtech. ChainLink’s demo at SWIFT had data provided by organisations (like Barclays, Santander, BNP Paribas) with literally trillions of dollars in assets.

Read between the lines. Consider the complete absence of hype. Look at the monster information that has come out already. DYOR.

4

Pajeet shill


(Kidding - good info)

TL;DR: LINK tokens are pretty much worthless, and SWIFT partnership will never actually use the token. It will remain tied to $1, effectively replacing USDT but never seeing a moon. Sell at $1 or watch moon missions pass you by forever.

user, this is a golden opportunity for all of us to make it. Don't let it go. Join this PnD server.

discord dot gg/JWWn6Qr

tldr just bought 100k

since the node operators will have a certain return on capital they will need to hit to maintain economic viability, meanwhile the SC owners will need to keep fees as low as possible, the economics will be a bit of a tightrope walk. i view getting the economics right as the biggest challenge to LINK earning me a lambo long-term

No you didn't. There's barely 100k on the orderbooks kek

Still doesn't mean the token won't rise. They could just lower the amount of link needed since its value is higher

the question should be: what will the aggregate amount of LINK held across all nodes be? and in the future is it conceivable that there will be enough fees payable to node operators to give them a sufficient return on their capital for operating their nodes? if no, its not economically viable long-term, imo

you are right :(

Im poor and hold only 1,5k link and thats like 70% of my portfolio.

Link is my only chance.

I save all of these threads, quality original content!
Very Stinky
Much Linky!

55k LINK since 0.17$. Not all in however. Only 30% of my portfolio.

...

we gonna make it bruh

It's fucking ogre guys time to sell these heavy ass bags. Chainlink is doomed to sideways movement, will ping pong between 60-70 cents until the bull market dies and the market crashes this shitcoin to 3 cents. Chainlink will never reach 75 cents. It will never breach 70 cents ever again. Market sell all Chainlink tokens (LINK) ASAP. Do the right thing, brothers.

Sergey said that SWIFT member banks pay clearninghouses 10-20 million dollars a year for bond interest rate data. Assume Chainlink cuts that in half. Across 1k nodes thats $5k a year.

Another number we know is zeppelin dapps are managing 1.5 billion in smartcontracts. If Chainlink charges .1% of the contract value that's another 1.5 million, or $1.5k per node per year.

From there you could try and determine rouglly how much link you would have to have to run a node, assuming the ROI would settle somewhere, maybe 5%.

Initially it might be ugly, but if they launch with big implementations right off the bat, it's definitely conceivable the tokens would have real nonspeculative value immediately. Also keep in mind dapps and smartcontracts are in their infancy and zeppelinOS is highly likely to get much bigger in the very near future.

I also can't speak to contract staking, I'm not sure how that would affect the ecosystem.

Ok thanks for the tip, just sold everything.

Those numbers are pretty dismal IMO...

For the system to work customers have to convert fiat to LINK to setup smart contracts. Node operators will then have to convert LINK to fiat to pay data providers, buy hardware, pay people, etc. The question then is how much LINK do you need to make a profit? If you see it that way then 10K LINK at a $1 doesn't seem like a whole lot to run a node for a year. If successful, LINK is going off the charts.

yea but there are literally no dapps right now other than cryptokitties. Also sergey himself seems to believe more money will move over blockchain than over the internet right now.

I'm just giving estimates for what might happen in the first few months, and numbers we already know, which is very little.

Haven't seen good info on link since the heavy fud and explaining-is-over era. This is highly appreciated. Very well thought and written. Thank you user!

Thank You user. This is the ultimate anti-FUD weapon.

Yea I think sergey envisions many nodes as not being something you set up in your bedroom.

Look at some of the numbers. If node operators make $10k a year, there would probably be growth in the number of node operators until the ROI hit something reasonable like 5%, which would mean $200k in LINK and/or the cost of running and setting it up.

High value contracts will probably involve significant amounts of "collatoral" so the nodes are accurate and reliable.

Is there any info on how many nodes will be needed for the main net at the time of launch? I've seen rumors saying 10k

I have a PhD in crypto economics and mathematics. Crypto incentives in Chainlink are a legitimate concern. I saw Ari Juels speak at a conference recently where he mentioned tokens and asked him about the token economics of a node staking system like the Chainlink network is planning to use. The problem is that node operator incentives are fuzzy at best and not even figured out fully by the team (see the gitter for Steve stuttering about this). When I brought it up to Ari Juels, I told him that in the way the network is expected to be used, the fees payable to node operators would actually decline as requests become more ubiquitous because as the network grows it becomes cheaper to use. This makes sense if you took a few advanced cryptoeconomics courses. Ari admitted that it was a great question but that they were "actively pursuing research in that area." I sold my LINK immediately after that and saw a significant dump on the binance charts. It's pretty clear these guys are pulling you along making you think they're doing something revolutionary when the incentives aren't even fully determined yet.

Do I need to be haxx0r to set up a node or could I rent a cheap azure vm?

>"actively pursuing research in that area."

Except that's literally a quote from Steve in the Gitter where he was just trying to placate some neckbeard who was harassing him for updates.

I'm not a phd in cryptoeconomics (nor are you I suspect as theres probably like 13 of them in the world and they are not posting on Veeky Forums) but I think it's more reasonable to assume Ari Juels and Sergey can figure it's possible out than some dumb larper can prove it's impossible.

Anyways steve is just a proggrammer, it's not his job to figure this out.

shit a 5% ROI? that's abysmal, how does linkpool expect to work if all they'll offer is something disgusting like 0.001% a year to their stakers

Even if you don't believe that I'm right, it's very clear from Steve's comments that none of the economics has been figured out yet.

It hit 84 cents the other night...

Also Binance doesn't give correct price valuations

Setting up is pretty much copy-paste, but in order to connect to APIs and to communicate with different blockchain platforms you may need to be able to build adapters. I'm hoping things will be pretty straight forward when connecting to open APIs (this will be a lot of the SWIFT stuff), maybe just running a node in SGX enclave without having have to really look into it too often

After holding link for ages, i finally sold today to get in the vechain train. I am sorry bros.

...

Damn bruh seriously nice fud. Even got me to second guess myself there. Guess those who make it truly do deserve it. Keep up the good work men

I see everyone is expecting to get paid insane amounts of money for running a node.
This is not a pajeet project with 50% stake rate, compare it to other legit staking projects like NEO, NEM, ETH in the future etc.
You think they have more than 5% ROI yearly?
And I see they're doing just fine price-wise.

If you're not satisfied with node ROI, then don't run it. There will always be people who'll want to run nodes and don't care about the ROI.

>the economics will be a bit of a tightrope walk.
Lol, this is how free marketplaces work. Askers want to pay as little as possible, and providers want to be paid as much as possible.
What are you, 12?

Why are people still pushing this agenda on 4cham? Fuck the nolinkers. They will learn too late that they missed the boat. Enough people have explained this enough times that by now they can honestly get absolutely fucked

Someone post the business card pic, dont have it saved on this pc

>is it conceivable that there will be enough fees payable to node operators to give them a sufficient return
Of course. Nobody can predict what will actually happen in the future, but it's absolutely conceivable.

LINK is such a fucking shitty coin to hold. Jesus christ can't even break top 100 on cmc, but shit like verge is over the moon. This token will be the death of me because it is such a pile of shit.

newfag

You realize you're complaining about being early to the party, right?

Way to miss the point.

I didn't.

if the economics between node operator and SC user don't work out the project is DOA and the token is worthless