Why there can be only a one oracle standard?

Why there can be only a one oracle standard?

Attached: 11link.jpg (319x960, 164.78K)

Other urls found in this thread:

blockchainconferences.io/speakers/longfei-wang
thedefiant.substack.com/p/defi-builders-are-neglecting-data-33a
eth.wiki/en/fundamentals/design-rationale
twitter.com/SFWRedditGifs

Attached: Link 19.jpg (800x448, 102.47K)

bump

Attached: 1595583033955.jpg (1080x1102, 321.36K)

Attached: 20200803_120227.jpg (1280x853, 332.73K)

The centralized shitoracle

there can be multiple but do ((you)) use bing or google to search for trannies?

Attached: 20200805_103503.jpg (800x759, 204.27K)

One large network is infinitely more valuable than many small ones

One oracle network to rule them all,
one oracle network to find them,
oracle network to bring them all
and in the darkness bind them

Because first mover advantage, like always

and why is it flare?

flare.ghost.io

Attached: Muh Meme.png (1200x756, 646.8K)

cause real platforms has built in oracle function.
so will eth soon

Attached: 456456yrtgyh.jpg (374x642, 71.29K)

Anyone can run a Chainlink oracle any way they want.
Whatever contract they want, whatever security features, whatever smart contract backend (ETH, Cosmos, EOS, Tezos, ...), however many nodes (from 1 to infinity), ...
Nodes can even choose to work for free, charging absolutely nothing.

Every node is basically their own oracle solution.

Wrong

How are you this ignorant.

Attached: chainlink neo.jpg (1001x955, 238.62K)

that linker on github has no idea what he's talking about. neo has had built in oracle since 2017

Attached: 56utyhg56.jpg (750x656, 136.34K)

And how does "NEO built in oracle function" works? Space magic?

He's a higher-up at NEO, pleb.
blockchainconferences.io/speakers/longfei-wang

Meanwhile that Kyleleblanc guy doesn't seem to be connected to NEO at all.

there is only one lord of the cube. and he does not share power

Attached: file.png (829x515, 711.82K)

same way they achieved 10000 transactions per second vs eth 15 and multiple coding languages i guess.

you know eth 2.0 is just coping neos old proof of stake mechanism right? now they're also trying to make the same gas model and the whole deal. the price only represent how many memes was made literally. the market is that retarded.

god i hate reddit and idiots like this

To be fair, you have to have a very high IQ to understand how Chainlink works.
Its superiority is only hinted very subtle, and without a solid dose of advanced autism most of the Lambos will go over a typical investors head. There's also Sergey Nazarovas very nihilistic outlook, which is deftly woven into his characterisation- his personal philosophy draws heavily from Zig Forums shitposting.
The real frens understand this stuff; they have the intellectual capacity to truly appreciate the depths of this coin, to realize that it’s not just funny- it also says something deep about society.
As a consequence people who don’t trust Chainlink truly ARE idiots- of course they wouldn't appreciate, for instance, the genius in Nazarovs iconic catchphrase "Get in here faggot, we’re gonna save the blockchain," which itself is a cryptic reference to the Russian epic 'Crime and Punishment' or the struggle of life itself.
I'm smirking right now just imagining one of those addlepated simpletons scratching their heads in confusion as Sergeys genius wit unfolds itself on their screens. What fools.. how I pity them.

And yes, by the way, i DO have a QR Code tattoo. And no, you cannot see it. It's for the ladies' eyes only- and even then they have to demonstrate that they're within 5 IQ points of my own (preferably lower) beforehand. Nothin personnel kid.

Attached: 71149BF3-1B2D-4EAF-86BD-8D9931E08B5A.png (2732x2048, 3.56M)

tfw it's already the age of orks

oops this one but yeah..
god i hate reddit and idiots like this

yeah this complex issue, neo is just gonna fix on the side with no authenciated api access and no security mechanisms
its the smae with all these projects, tezos, cardano too, they have a couple of young engineers who go oh yeah easy we can easily figure out the oracle problem by ourselves by... applying blockchain concpets to it!!
sergey rips this thking apart here: thedefiant.substack.com/p/defi-builders-are-neglecting-data-33a

>built in oracle function
Now imagine you have a contract that involves two parties who each want to use a different blockchain for the same contract: a public one (e.g. NEO, ETH, ...) and a permissioned one (e.g. Hyperledger, Corda, ...). Or multiple variations thereof in a single contract.
Good luck getting any kind of coordination.

Plus, a large use case for Chainlink is cross-blockchain communication. Again, good luck with that if you want to use proprietary oracles for each chain.

Not to mention the fact that an oracle system that is used for ALL blockchains will obviously have a lot more experience, reputation, stress testing, bug finding, options, scope, ... than an oracle system that only serves one blockchain.

Go cry in the corner.

here for lazy:

The second issue is that —and this is something we take very seriously and I've done a presentation on recently and I advise anybody building an Oracle mechanism on their own internally or whatever system they use to look at very seriously— is I've seen one or two oracle mechanisms where people do not take into account data quality. So, they basically ignore the fact that in traditional finance there is a large group of data companies like Bloomberg, Reuters and others, that are hugely successful companies in hugely competitive markets, because they smooth out risks related to data.

And there's people that basically say, I'm going to completely ignore all the risks from data, I'm going to both make a data aggregation methodology and I'm going to build an oracle mechanism, so I'm going to do both. I'm going to create a data company and I'm going to create a piece of software that's meant to provide security about the transportation of data. And then they begin to make all the mistakes that people who have never made a data company make, which is why we don't do that.

oops wrong paras

There's these two approaches that I see the people solving this problem. One approach is I'm going to take blockchains and I'm going to replicate everything I do in blockchain-land on to oracles. That usually misses the point of the key problem you're solving. The key problem you're solving is highly reliable, highly secure, highly available access to external systems, which you don't even know if those systems are secure, so you then also need to deal with that problem. And so that blockchain approach of, “it's just a blockchain but different,” is one of the first issues.

The second issue is that —and this is something we take very seriously and I've done a presentation on recently and I advise anybody building an Oracle mechanism on their own internally or whatever system they use to look at very seriously— is I've seen one or two oracle mechanisms where people do not take into account data quality. So, they basically ignore the fact that in traditional finance there is a large group of data companies like Bloomberg, Reuters and others, that are hugely successful companies in hugely competitive markets, because they smooth out risks related to data.

fucking neo idiots
eth has no features, point 4
eth.wiki/en/fundamentals/design-rationale

In order to understand the answer to this question you have to look into the future
Let’s say it’s 5 years from now and we are storing digital assets in our bank accounts. Large financial institutions are issuing digital assets and they’re trading on large exchanges like the NASDAQ.
The USA issues it’s own digital dollar that’s backed by gold and silver, and it uses ChainLink oracles under the hood.
Mumbai, wanting their own digital rupee, decides to follow suit. They are using band protocol to make their rupee an exact replica of the USA digital dollar, with an equivalent mix of gold and silver.
But because they are using different oracles, the band protocol oracles apis occasionally tick for gold prices at different than the ChainLink ones.
This doesn’t work. If the dollar says gold is $2000, and the rupee says gold is $2001, and I buy a shit ton of rupee, I am getting ripped off. Oracles are a winner take all and ChainLink has first mover advantage. We won a long time ago.

retarded-tier thinking
take a look at maker's post about why chainlink is useless to them and why they're focusing on their own oracle service
this is a sector growing exponentially right now and no one's going to give up and give their business to chainlink just because they have good data feeds and more experience thus far, when they could compete since they have all have more than enough funding
imagine chainlink is amazon, but they only have their AWS service, which is fantastic because their AWS service is huge
now think of maker, think about it from their perspective
are they going to not compete and give up a huge market by allowing "AWS" to run their "cloud computing"
nope
they're going to continue developing their own and compete
i hold link, have since just after ICO, and still have 75% of my stack
but i'm not retarded, competition is good and i bought into all native defi tokens as well as all tokens on defi protocols developing their own oracle services, and even band when it was $2 just months ago
so fuck off with this retarded shit. stop being a loyalist and step back and realize you can make money by not being attached to something just because you've been brainwashed into thinking their market share can never be eaten into
i mean, just the fact that you idiots think chainlink will have 100% of market share boggles my mind
no wonder you're all up 24/7 watching charts wondering where the top is, all your money is tied up in link and any thought of competition sounds absolutely ridiculous and impossible to you because your brain can't accept the fact that you might not ever realize any substantial gains

Attached: 1597629525371.jpg (756x606, 978.06K)

>they have a couple of young engineers who go oh yeah easy we can easily figure out the oracle problem by ourselves
newfag programmers always reinventing the wheel while not delivering a product. many such cases

>Oracles are a winner take all
damn, you guys must've put every bit of money you had into link to deny logic and say shit like this to not have a panic attack any time a competitor steps in

>their own oracle service
we've heard it all before
witnet, mobius, doracles, bandicles, flare
they are all pepsi to Sergeys coke

Most places will use multiple data sources to stop any kind error