First Mover Advantage is a huge benefit for any cryptocurrency, but especially so in the case of ChainLink...

First Mover Advantage is a huge benefit for any cryptocurrency, but especially so in the case of ChainLink. This is due to the fact that any decentralised oracle system needs a lot of hardware and IRL adoption to be functional. Any subsequent project that tries to enter the space as a competitor will have to do one of two things. Either; 1) They will have to create a whole new group of node operators, which will be extremely difficult considering everyone with the means and intention to run an oracle node will already be doing so through the ChainLink network. 2) They will have to incentivise node operators to switch from the ChainLink network to their network. This will be equally difficult as the incumbent network will have all of the inertia, adoption and users.

The upshot of this is that it is likely that ChainLink will operate as the unchallenged and premier decentralised oracle service. The bigger it gets, the harder it becomes to challenge it. FMA is particularly useful for ChainLink due to the unique nature of its real life infrastructure. Nothing else (no, not Aeternity) is meaningfully competing with ChainLink for the position of premier decentralised oracle service. With the increased global adoption of smart contracts (which is inevitable, because a trustless system is by definition more efficient than a trusted one), ChainLink is poised to be the global bridge between any legacy business technology and blockchains. It is the most exciting opportunity for spectacular growth and real life use in the cryptocurrency space over a 1-3 year time period. Don't miss out.

Other urls found in this thread:

create.smartcontract.com/#/contracts/fa4703cb68e3c152a9f47bafd57fe1fa?tab=info
docs.corda.net/key-concepts-oracles.html
twitter.com/NSFWRedditVideo

>it is likely that ChainLink will operate as the unchallenged and premier decentralised oracle service.
indeed. and we early LINK adopters are gonna be fuckin, HELL ASS FUCKIN RICH

Meh sergay is a bitch. Will get in the way of success

Why can't node operators sell their data to multiple different entities? Also so what will come first the node operators or the people paying for their information? Will businesses that need data set up node operators and if so doesn't that just make it centralized if they're pulling data all from nodes they've created?

I was an uninformed linkie and still am but the more I learn the more dizzying the potential of LINK becomes clear.

>1) They will have to create a whole new group of node operators, which will be extremely difficult considering everyone with the means and intention to run an oracle node will already be doing so through the ChainLink network
who?

>2) They will have to incentivise node operators to switch from the ChainLink network to their network.
or simply hire new node operators using a different solution

>The bigger it gets, the harder it becomes to challenge it.
its small

>Don't miss out
im hodling btc so while your hodling bags of shitcoons

do your own reserach brainlet

yeah your whole 0.11 btc. we're so jealous

Feels good to get on this early. Seriously gonna hit like 100 in 2018

fortunately the only people hyping link seem to be here on Veeky Forums, letting the link team continue to improve the product without attracting competitors by the hype

Damn, I have the exact same picture in my head of how things will play out, but never said it out loud. I'm assuming By your post that (((they))) must know that too.
Fuck, I'm off to binance to fill my bag to the fullest

screencapping for pink wojack usage later
t. brainlet

>Why can't node operators sell their data to multiple different entities?

I don't actually know the terms of node operation or if there are any exclusivity clauses, so it may potentially be possible for a node operator to process data for multiple decentralised oracle services. The fact remains that even if that occurs, ChainLink will still have the largest and most robust network, due to FMA.

>what will come first the node operators or the people paying for their information?

Node operators. The software has already been shown to work, so the company is currently working on creating a robust enough network of node operators for a functional main net launch next year. There is every reason to believe that very large financial institutions are watching the launch closely. A successful launch cause an unbelievable value increase. A failed launch could seriously injure the project.

>Will businesses that need data set up node operators and if so doesn't that just make it centralized if they're pulling data all from nodes they've created?

No other party would choose to transact with a business who was pretending to use a trustless service but was trying to game the system by loading the network with their own nodes. And even if they did, any discrepencies in their data due to manipulation would be observed by the network and instantly red flagged. Manipulation would not work due to inconsistencies that would arise in the data feeds.

>>Why can't node operators sell their data to multiple different entities?
>
>I don't actually know the terms of node operation or if there are any exclusivity clauses, so it may potentially be possible for a node operator to process data for multiple decentralised oracle services. The fact remains that even if that occurs, ChainLink will still have the largest and most robust network, due to FMA.
long term, the vision is that nodes will be running hardware with the intel sgx extension set that will not the node be able to examine the code/memory being executed. this means you could pass, e.g., your ethereum private key, and have zero risk of it getting ripped off. under this scenario there's no reason to mandate exclusivity imo

create.smartcontract.com/#/contracts/fa4703cb68e3c152a9f47bafd57fe1fa?tab=info

unfortunately a lot of people are under the perverse impression that because LINK is bleeding, the project must be a scam

reminder that less than 5% of total coins are being actively traded. Meaning the price is dictated by 5% of coins. In essence, the price has no meaning

if LINK were at say, 60 cents right now, Veeky Forums positive sentiment would be much higher despite the project being in the exact same position. It shows you that most traders are cognitively biased and emotion driven.

>the price is low because the price is low

>reminder that less than 5% of total coins are being actively traded. Meaning the price is dictated by 5% of coins. In essence, the price has no meaning
that really gets the almonds rustlin'

>Will businesses that need data set up node operators and if so doesn't that just make it centralized if they're pulling data all from nodes they've created?

You could create a contract that only uses 1 or a few Oracle nodes, and attempt to manipulate the terms so you would use nodes that you have created.

While there is the possibility to commit fraud this way, the terms of the contract are transparent and the fraud is easily recognizable to anyone paying attention. In particular, a contract calling for only 1 or a few nodes should raise red flags right away

exactly. the smart contracts are going to be written in such a way that they are calling and verifying data via the access channels of the chainlink decentralized oracle network. this is the MOST efficient way to do things and thus it's gonna be WIDELY adopted.

Hitting 100?
That's like 588x gains. Isn't that too much?

nigga you think we was kiddin?
this is your lambo ticket, full stop.

I have 344 link. And I don't intend to sell till the end.

If it hits Ethereum's current market cap tokens will be $30. I think given the broadness of the use case then this is a reasonable value to expect in the mid term (3-5 years).

Thanks m8s that makes it a little clearer.

I can only imagine how many nodes there would be and how much link they would have to have to staked (I know that's only part reputation but given how lucrative it could be I could imagine they'd want to have as much as it would take) would do to the price. Millions of nodes with tens of thousands of LINK tokens would push the price up to lambo levels

my almonds have been rustled long enough

excuse me, he is such not a bitch he made vitaliks pussy wet

Also despite the FUD caused by LINK going sideways as every other coin has mooned there is no actual fundamental change in the project. This is meaningless FUD.

You also forgot the fact node operators will have spent time building up their reputation and won't want to switch to another Oracle platform and start from zero.

lol no wonder vitalik bad mouthed LINK

he got a boy crush

what about corda? They just released their 2.0 network and they are working with banks. Did some research but I'm still not sure if they have their own oracles or if they are planning to use an "external" network. If some smarter user could have a look: docs.corda.net/key-concepts-oracles.html

Not buying your bags

There is fundamental change in the project
It turns out they are are lot further behind than we originally thought
Most people thought they had a fully finished product for sibos... turns out it was just a poc and nothing more
They haven't even hired devs yet...
We won't see an actual finished product until q2 2018

That's not new as of the past two days though.

Try your maths again, it would be 150 or so per link. Based on the same circulating supply

>q2 2018

So buy now. Got it.

My fellow LINKmarines are the funniest mothafuckas on Veeky Forums

Greed is the only thing moving this coin

If u don't trust the link Google corda blockchain oracles. what do you guys think. I know for a fact that there are banks developing real estate solutions in blockchain with corda

holy shit vitalik is such a fucking fairy boy