[ 3 / biz / cgl / ck / diy / fa / ic / jp / lit / sci / vr / vt ] [ index / top / reports ] [ become a patron ] [ status ]

/biz/ - Business & Finance


View post   

File: 34 KB, 800x600, 78787878.jpg [View same] [iqdb] [saucenao] [google]
10153540 No.10153540 [Reply] [Original]

The off chain data problem cannot be solved.
Take at look at this joke: https://www.reddit.com/r/ethtrader/comments/6w5wcn/important_update_mayweathermcgregor_smart_contract/

Ethereum hasn't improved in this aspect for over 3 years. Vitalik continues to bury his head in the sand to work on TPS all while the elephant in the room is that blockchains cannot handle non mathematical consensus.
This has many implications. Any API used for data is also not meant to be used for smart contracts with money on the line. these websites submitting gold value, stock market information, movie prices, train times, etc are not anticipating it. their intern submitting the data doesn't give a fuck.

Then we have the chainlink and Angur mess. endless second layer "solutions" with new coins bloating the network in a never ending spiral. In emergency situations, this can never solve things in a timely manner. it wont work. and AI wont solve it either becase the "truth" is malleable as we have learnt, and turn smart contracts into a philosophical dead end.
Imagine my house burns down, and i have a mortgage smart contract on a financial blockchain. who determines to what degree the damages are? who decides that beyond this point its irreparable? who assesses this damage and comes to an agreement? if it's the risk taker who provided the loan, then wahts the fucking point of the smart contract?

Smart contracts will always need a centralized appeal to authority or government. So the question begs, why not use a database or a private blockchain at best?

>> No.10153575

>>10153540
>wall of text

What's wrong with the contract on reddit?
contracts are supposed to be immutable
go buy EOS RAM or something pajeet

>> No.10153587

You're sort of all over the place there sport

>> No.10153602

>>10153587

actually my thoughts are a bit. i dint want to flesh this out to much because it would be a fucking huge read and take me an hour to write it.

basically the smart contract or offchain data problem cannot be eliminated. we cant have a decentrilized solution. even the presented ones are not worthy as in emergencies they take too long to reach consensus. and they bloat the network with more coins.

>> No.10153606

>>10153540
>Imagine my house burns down, and i have a mortgage smart contract on a financial blockchain. who determines to what degree the damages are?

An independent contractor with enough reputation and licensing to perform the inspection. All of this information will be on some of the linked chains for other smart contracts to access and verify.

It's pretty much like it is now and there will always be incentives to play fair or basically ruin your reputation on a never-forgetting web 3.0 forever.

>> No.10153609

>>10153540
Hmmm so in short, more power to the people? Sounds good to me.

>> No.10153614

>>10153575

the cotnract was ruined because the website it took data from changed from lower case letter to capital letters. that was it.
think of the implications of this. thats some basic as fuck shit it cant even overcome with a centralized solution(data from a bettign website).

>> No.10153642

>>10153606

In THOERY. we have no solid presentation of this at all yet. The onus lies on future developers to "figure it out".

How many fucking chians are going to connect to each other, all requiring a coin and gas fees? what if there is some sort of immediate financial problem, where a stock or such plummets? the decentralized solution cant get consensus that fast, so we need a centralized one. One that anticipates being used. the infrastructure for such things is not even here yet.

>> No.10153658

>>10153540
> implying that it's impossible to get reliable information on the internet

> those websites submitting stock market information

People and institutions place multimillion dollar stock and derivatives orders that are executed conditionally based on data...if there is enough demand for reliable data for something (boxing fight, e.g.) for smart contract use, a reliable data source will be created

it's called economics you fucktard

>> No.10153669

>>10153587
lel

>> No.10153689

>>10153658

Aye, but its not decentralized. thats waht everyone says. It will have to be centralized at the point of off chain data and taht is just reality.

>> No.10153702

>>10153540
>>10153642
A blockchain is a closed system, in which transactions are proofs, and where new proofs are derived exclusively from existing proofs. A previous transaction commits to a carefully-crafted number, and a future transaction reveals an associated number that can only be produced from the knowledge of a secret hinted at in the commitment.

This mechanism -- this axiomatic system -- has turned out to be eminently suitable for monetary systems, i.e. cryptocurrencies.

However, it proves claims only about itself.

Cryptocurrencies demonstrate nicely that the essence of money is purely an abstraction, i.e. money lives in its own abstract world.

However, a blockchain will never be suitable to prove claims about the real world, if only, because it is impossible to prove anything about the real world.

Billions of dollars are being wasted right now on blockchain projects that attempt to prove claims about the real world. By adding just one real-world data element into a blockchain, it will no longer prove anything anymore.

>> No.10153707

>>10153540
>reddit.com
Hang on...

>> No.10153710

lmao
>puts sunglasses back on, sips energy drink, revs up lawnmower again

>> No.10153731

>>10153642
>How many fucking chians are going to connect to each other, all requiring a coin and gas fees?
It's really hard to tell depending on how versatile they are, but I'd be surprised if there'll be more than there are economic zones right now with each having 2 or 3 chain solutions that work in tandem with each other.
This could be a combination of different types of chains for different type of usages and things like DAGs and Hashgraph are only the beginning of the inevitable evolution of this new wave of decentralized trust consensus computing.

>>10153702
You don't need absolute, mathematical proof about the real world up to quantum uncertainty, but just enough proof that it's better and cheaper than centralized systems, which isn't that far off if governments don't just outright ban blockchain etc. over their fear that they might lose their monopoly on currency.

>> No.10153756

>>10153702

EXACTLY. This is why i said BTC is the only use for blockchain. or Monero. maybe XRP.

Blockchians cannot handle non mathematical data and concencus. they can only abstract raw maths. that is easy for them to do, and very easy to code and make bullet proof.
off chain data? broken. smart contracts cant work. this is likely why BTC removed smart contract functionality, to protect the network. the original team foresaw this.

Its not oging to be fixable. it became a philosophical debate and chase over what the "truth" is in all its subjectivity which philosophers have never been able to answer in a practical way. as we have seen in reccent times, "fake news" is more prevalent, and we have seen certain "bodies of trust" like Facebook taking it upon themselves to decide waht that truth is.

Basically it can never fucking work, unless we accept a degree of centralization. but then, i argue, wahts the point of a public smart contrat platform over a database and the traditional means?

>> No.10153795

>>10153731
sports betting

decentralized nodes scour reaults from multiple bookies to come to an answer then sc is executed on consensus from nodes, bad actors are immediately blaxklisted by vote, not completely decntralised but better than having sportsbet.com kill your account forever on a whim. is that good eniugh or am i brainlet?

>> No.10153846

>>10153795

it removes the decentralized aspect. liek you said.
Additionally, these websites dont intend on smart contracts scraping information from them. that can cause issues legally.
betting is one of the few ways blockchain smartcontracts can work imo. and thats not good. complicated things like taking flight times for travel, exchange rates, commoditie prices vs a time value, mortgages, even a literal dead mans switch contract if someone dies... endless problems. about as endless as reality itself. you cannot code these things and all their infinite variables.

the only "solution" is a centralized authority.... which eliminates the need for a blockchain for smart contracts.

>> No.10154157

>>10153846
Fuck it's Astro, again... Been a while how have you been? Back at the ol' FUD game I see, well good luck. Kinda missed your ability to turn people into frothing lunatics.

>> No.10154217

This is how you know Link will be bigger than Eth

>> No.10154308

>>10153614
You're suppressed to account for this when making the contract poojet

>> No.10154401
File: 18 KB, 539x374, 539w.jpg [View same] [iqdb] [saucenao] [google]
[ERROR]

>>10153540
>>10153606
>>10153702
>>10153731
>>10153846
>>10154157
Ari here, I have a scanner that browses forum posts for smart contract related topics. This site seems to pop up frequently and is generally the one I actually go to because I don't have to sign in or create an account(and hence the delay in getting to this post)

You guys are completely missing the point. Chainlink and Town-crier exist only to ensure that the smart contract computation is secure and can't be tampered with. It says nothing about ensuring the actual source data, which is obviously impossible(although I can think of many ways to improve the reliability of oracle feeds)

Think of it like receiving a package on one end of a bridge . TC ensures that the bridge is structurally sound and protected from bandits during transit. That's all. I'm free to answer a question or two for a few minutes if anyone is still here!

>> No.10154439

>>10154401
When $1000 per link

>> No.10154726
File: 47 KB, 600x656, 1515339058592.jpg [View same] [iqdb] [saucenao] [google]
[ERROR]

>>10154401

you seem very flexible.

Can you web with the best of them?

>> No.10154816
File: 154 KB, 725x1024, 1503314504986.jpg [View same] [iqdb] [saucenao] [google]
[ERROR]

>>10153540
You can have a lot of moving pieces on this with different degrees of centralization. You can have public blockchains and private blockchains and use them at the same time for different things. I believe tokens have a role to play in all this, it's really hard for me to say what's the "real value" of the tokens before seeing how many "real contracts" end in failure or how much is at risk on average.

> Imagine my house burns down...
Some contracts WILL end in disaster. That may be a way to discover the price of the tokens. We can have third party auditors on the resolution of smart contracts. What I mean in the end is that if the data that was used wasn't reliable, you can still sue, courts aren't going to disappear even if this shit actually ever gets adopted. Of course, for the inconvenience of having your house burned down and the insurer trying to be a cheeky cunt, some extra compensation has to be expected. Since we're on this, some court resolutions are unfair. Could this shit bring even a worse hell? Maybe. But there's a lot of upside on "I'll own this today once 3 people say everything's good to go" vs "I'll own this in some months after a lot of paperwork between 93 different people and they finally decide everything's good to go".

We're not taking down the state or getting rid of the banks. At best, we may gain fairer negotiation terms and if that happens it's by accident. Also yes, ETH is messy as fuck for a lot of different reasons but it's still one of the most serious leaders we have participating in building infraestructure.

>> No.10154863
File: 959 KB, 1092x856, 1518113158184.jpg [View same] [iqdb] [saucenao] [google]
[ERROR]

>>10153540
>Basically it can never fucking work, unless we accept a degree of centralization. but then, i argue, wahts the point of a public smart contrat platform over a database and the traditional means?
>>10153602
>>10153702
>>10153756
>>10153846

Look OP, I know your Jewish masters told you all of this but it just isn't working. Blockchain is the future whether you like it or not. No, you can't run a fucking server on the blockchain but it's enough to move your own coin and ecosystem and economy without your rothschild bosses printing money for themselves.

>> No.10155492

>>10154816
This. People were saying alot of crazy claim about how the dot com companies were gonna change the world and while it did change the world to an extent it was completely different than anyone thought. We are not getting rid of banks or governments for sure. All tgat blickchain will do is make the handling of certain financial agreements more secure , efficient, cheaper, transparent, more trustless, and potentially anonymous in certain situations.

>> No.10155503

>Bitcoin is literally the only use for blockchain
Stopped reading there. Already agreed

>> No.10155548

>>10153795
most of the worlds money is used for betting purposes, just because the smart contract cannot automatically win the law suit against the dog food manufacturer when he dies does not mean there would not be vast high level effiiencies

>> No.10155565

>>10154401
>You guys are completely missing the point. Chainlink and Town-crier exist only to ensure that the smart contract computation is secure and can't be tampered with. It says nothing about ensuring the actual source data, which is obviously impossible
How is this the only mention of oracles in this thread and yet it completely misses the mark

I guess the smart anons going long on LINK really did leave this shit board that's now overrun with brainlets and pajeets

>> No.10155608

>>10155565
Those anons sold in january after convincing you to fomo in on a failed project.

>> No.10155641

>>10155565
I've noticed this too. It's impossible to have an intelligent conversation on /biz/ about the future of crypto anymore. All that's left here ate newfags and poos.

>> No.10155655

>>10155608
Just bought 100k Mobius thank you sir

>> No.10155708

>>10155565
I think we've gone over that enough times already.

> Why the ideal oracle (ORACLE) is hard to achieve. There is, of course, no perfectly trustworthy data source Src. Data may be benignly or maliciously corrupted due to faulty web sites, cheating service providers, or honest mistakes. [...] ChainLink reasons about its security protocols in terms of this ideal functionality ORACLE. Our goal in ChainLink is to achieve a real world system with properties as close as possible to those of ORACLE under realistic trust assumptions. We now explain how.

I don't think there's much point arguing about it when the mentality is still
> But what if they just build their own oracles guys?

I plan to hold my Linkies to 100x or 0 (unless something really special happens), but I'm not buying more until the fat fuck speaks.

>> No.10156561
File: 56 KB, 1200x449, DIGWPI9XkAAk75X.jpg [View same] [iqdb] [saucenao] [google]
[ERROR]

''blockchain is literally the only use for blockchain'' ....that statement is immediately made WRONG by for example GUTS tickets using the GET token protocol....which already has a working product that prevents third party ticket resales for ridiculous prices and is selling over a million tickets on the blockchain in the next year (including a jazz concert, the most successful French singer of all time)....so yes there are amazing use cases for blockchain

>> No.10156572

zazz * rather than jazz lol

>> No.10156655

>>10153540
Man, why are you guys pooping your pants that a NEW EMERGING technology dosent work properly or isn't easy to operate.

Give this shit 10 years, I can operate computers but most likely won't be able to set up contracts by then.
15-20 years, there's gonna be 500 easy ways for a Norman like myself to set up contracts.

This shit is only a couple years old and right now all of this is testing.
Calm your fucking dicks, just keep buying with fiat 5-10% of your wages.

Jesuz

>> No.10156660

>>10156561
This is just BTC with a different name.

>> No.10156732

>>10153540
that's funny