Alternative Uses for Blockchain - Computerphile
ฝัง
- เผยแพร่เมื่อ 12 ก.ย. 2022
- Blockchain has a controversial reputation, linked as it is to cryptocurrency but Professor Peter McBurney of Kings College London thinks it's an important an invention as the web itself.
This video was initially titled "Blockchain Benefits" to reflect Prof McBurney's talk
/ computerphile
/ computer_phile
This video was filmed and edited by Sean Riley.
Computer Science at the University of Nottingham: bit.ly/nottscomputer
Computerphile is a sister project to Brady Haran's Numberphile. More at www.bradyharan.com
![TREASURE - 'KING KONG' COMEBACK SPECIAL [THE TREASURE SHOW]](http://i.ytimg.com/vi/Jx1wyMaQzf0/mqdefault.jpg)
I love the fact he talked about that long island tea company and didn't mention that the SEC forced them to be delisted for fraud and several people were arrested for insider trading because of the name change
Blockchain has one use case and one use case only: decentralized trust... there are problems that could be solved by that solution, namely cryptocurrency... yet every single one of several hundred elevator pitches for blockchain startups I have been forced to listen to, at some point inevitably involved that particular misconception of an oxymoronic 'private', or otherwise 'controlled' blockchain, since non of the business cases could possibly ever work, when trust would actually be decentralized.
The whole blockchain discussion proves to me, that giant amounts of private and public fortune are controlled by a wildly incompetent financial industry that can't be bothered to do even the most superficial effort to understand the smallest possible set of necessary facts before they invest in some bullshit. I also have a strong feeling, that problem contributes a lot to why we as a society can't have nice things.
If you have a private Blockchain under the control of a single party you just have a centralized service using an extremely inefficient technology.
I failed to recognize how any use case described in the video benefits from the application of a blockchain.
Obviously interviews have to be edited down for time but I'm (not) astounded that given ten minutes of screen time, Professor McBurney didn't give a single compelling example of a use case for any blockchains.
"There's also all sorts of genuine business development." [Citation needed]
"Companies want their data distributed because of antitrust laws--" what? How does that even make sense? What kinds of data does he think he's talking about? If data is distributed on a public trustless blockchain it's observable to the other users of the chain, which companies generally don't want.
"Who is going to get the benefit from monetizing this data?" You're aware that when companies give their data to third parties they are allowed to keep copies, right? Like, oil companies aren't giving their business ledgers to an accounting firm and not keeping a backup, and I cannot imagine a universe in which "monetization" of ledgers and accounts, the kind of stuff that can actually be distributed on a blockchain, would be helpful for ANYONE, much less consumers.
Questions of access and security are completely orthogonal to the questions of blockchain technology.
I hope that all the engagement and comments are worth the unsubscriptions.
I don't find the proposed benefits compelling.
What Prof. McBurney describes here as the benefits of "tamperproofness" are just normal cryptographic signatures, at best the improvement here is just using a chain of hashes (merkle tree) to prove an ordering.
The insurance example is weak. A blockchain cannot _prove_ a fire insurance contract, as the house that's burnt down exists in the real world, rather than being contained within the blockchain. Someone has to update the blockchain to inform it that the house has burned down. And as such, the insurance company still has to be trusted to honour the claim; If you claim an accidental fire but the insurance company claims arson, a blockchain cannot settle that dispute.
The data protection example is even worse. Blockchains have nothing to do with access to data (and existing blockchain applications generally make all data public to all parties). Any data protection, encryption, or access control have to be built on top of the blockchain just like it could be built on top of any other conventional database.
Would love to see the prof or somebody from Computerphile pin this comment and provide a clarification to your comment ! Really well made points here
You have the right questions, and there are answers. See innovations in blockchain “oracles” to satisfy the problem of introducing external information to a blockchain.
Smart contracts can cryptographically obscure specific data from the public by default and allow only authorized access as designed within the contract. These layers of granular access controls specified by open and immutable standards is what makes blockchain a profound data security innovation.
I ask: How are you going to enforce a contract or promise in the real world?
If the answer is: Well, you’re going to court.
Then what’s the benefit of the blockchain?!
"Someone has to update the blockchain to inform it that the house has burned down"
The answer to this is decentralized oracle networks. They're already updating subjective price data used in decentralized futures/options markets such as synthetix. Sure, it will be some time before a decentralized oracle network can tackle insurance data such as if a specific house has burned but with enough different IOT sensors to feed the data, nodes in the network to put it on the blockchain and keep each other in check, it is definitely a possibility.
@@K1RTB The smart contract can itself be an escrow service.
"Blockchain Benefits" - the video was 12:03 longer than I thought it would be.
"Guarantee that it won't be altered"
maybe guarantee is a bit of a strong word, more like "be highly confident that it won't be altered"
How could you ever guarantee that something won't be altered?
@@petargolubovic5300 I don't know, ask the presenter, he said that
@@Embassy_of_Jupiter it can't be altered...
... Which is why we're going to fork the entire thing and ask people to use the new fork in which we had to go back to undo the previous entries since our largest wallet got broken into Oceans 12 style 🤣
@@petargolubovic5300 ironically we can already do this and it has nothing to do with blockchain: It's called content addressing, basically just referring to a piece of data by a cryptographic hash.
Wait till this man discovers Git. Gonna blow his mind.
computerphiles account must have been taken over... what about signatures?
That last part was a weird way of saying "blockchain can't do this right now and we're not sure if it ever can" but making it sound like "blockchain can do this".
To be clear, I'm all for computer scientists and engineers doing their science and engineering, and being thorough before jumping to conclusions. But the fair skeptics have been saying, it's been a decade. Software innovation moves fast. Where are the fruits of its promises?
You have to admit, him describing the web and blockchain as the only two major important technologies puts a lot of overexcited bias smell on him that we've had over the past few years.
Didn't think Git was revolutionary? Torrents? CDNs? 3D Graphics? Neural Networks? Things that are already used for a multitude of practical things and/or advancing rapidly.
Also insurance as I know it doesn't need "interlocking" promises. I don't promise to pay the insurance company money, I actually pay them money. Only after they get the money do they start to promise to pay me in the event of an accident.
Agree - Git and 3D graphics in particular have both been *actually* revolutionary. One fundamentally changed the way software was written, the other is a technology that has become as standard inside every single PC and most smartphones as flash storage and LCD displays
@@user-zu1ix3yq2w it certainly has failed to produce a viable cryptocurrency, just look at how unstable the values of the currencies have been
As someone who's pretty cynical about all blockchain technology (I made several POC apps for IBM on their blockchain platform and was able to watch in real time as the interest and work dried up) I only made myself watch this video because I wanted to know what alternative uses were being proposed. I'm extremely disappointed; none of the proposed alternative uses this guy mentions are particularly interesting or novel, and I would go so far as to say some of them (like the insurance example) are outright a bad fit for blockchain. I'm glad he mentions the challenges involved with the tech, at least.
It's highly debatable as to whether or not Blockchain technology solves the problems you've described in a way that other technologies can't but you're mostly leaving out the glaring pitfalls and other issues too. Even if I agree with your points, and I do not, the blockchain doesn't scale anywhere near as well as other technologies already do. That's not an easy thing to fix either because it's a core facet of how it functions, the decentralized ledger reaching consensus is what scales poorly and it's also the only redeemable aspect of the technology.
The first thirty seconds of this video both concern and confuse me. How could anyone think there have only been two revolutionary IT technologies in the past few years?
In contrast, here are some of the revolutionary IT technologies I have seen in my adult life: The miniaturization of semiconductors, the explosion in global data storage capacity, the invention of smartphones and the expansion of global communication, the internet (www), and so many more. However, I am not sure I would put blockchain on my list.
What would you put on your list?
Blockchain: the ideal solution for most of the problems that arise from the invention of blockchain.
So an **incredibly** inefficient cryptographic signature (by many orders of magnitude) with the singular advantage of requiring you not to trust a single large entity (like a bank, government) but to trust that 50% of the nodes are not controlled by a single large entity, or a collaborating group equivalent to it (which would still be free to alter history).
To adapt Upton Sinclair's old adage: it is difficult to get a man to talk objectively about something, when his investment portfolio depends on you not understanding it.
The title feels like a punchline to one of those "shortest book ever" jokes
Can the professor name a few of these companies making great real-world use of these public blockchains, please? Preferably aimed at consumers (as opposed to private "investors")
Still waiting on the revolution. Aaaaaaany day now.
Blockchain entries' USP aren't that they're encrypted, it's that they can't be removed. A catastrophic nightmare for privacy.
they're as secure as any other PKE transaction or record. security is always a tradeoff. Obviously, this wouldn't be a great solution for having secure conversations, but for proving provenance or voting or sharing medical records, this is the perfect (indeed only) solution, since it enables ZK proofs to be implemented.
"Blockchain" is taking its rightful place along terms such as "snake oil"
we will see how smart you sound in a few years . all the worlds biggest colpanies are already using blockchain and more plan on starting . the entire logistics of the world trade organisation is going to be blockchain based . you have never spoken to a blockchain dev have you
I still have yet to hear a good use case for blockchain, and these twelve minutes have not changed that.
zero-knowledge proofs
The only thing blockchain can truly protect are things that are created and live within the blockchain itself. This is why crypto, from a technical standpoint, works.
Things that are "mapped" onto a blockchain are still subject to all of the, duplication. modification, theft etc because the "real" asset exists outside the blockchain.
One place where it might work is for something like music copyright but would require the original owner to actually insert the encrypted recording into the blockchain itself for later proof if the need arises. But I'm sure even that has holes in it
It's interesting how this video singularly fails to produce a compelling argument for the premise of the video: that blockchain is a revolutionary technology. Especially because it doesn't grapple with a core issue that plagues all the projects using blockchain in a non-cryptocurrency context, that the one benefit of a blockchain - that it doesn't require trust between parties - completely evaporates in smaller implementations, so called "private blockchains", because subverting the validation mechanism becomes trivial - just write calculate a new, mathematically valid, longer blockchain on a faster computer and claim that yours is the true one. From a definitionally pure perspective, you'd even be correct.
Anyone making claims about the revolutionary nature of blockchain for technology other than cryptocurrency should be required to confront that issue. Or at the very least come up with a clear cut use case for this revolutionary technology, which isn't entirely inferior to a distributed ledger without a blockchain attached. The revolution can't be that you've enabled certain companies to more conveniently negotiate certain complex contracts. You could fix that with regulation or a change in the way regulatory oversight is done very easily, no revolution there. You certainly can't put that on the level of the web, and anyone comparing the two should be able to (and required to) make a vastly better case for it than Professor McBurney managed here.
He does completely ignore the most important innovation , proof of work, and the reason why blocks in a blockchain exist and the most important application of a blockchain= money (bitcoin) and the implications of a money that can be owned by code instead of humans and something that is fungible and not easily censored.
cool, so there is this blockchain, onto which each of the six oil companies wrote the promise "we will not talk about prices in our meeting". a promise that they already give via many other channels. but, you know... what if they talk about prices anyway? nobody will notice, at least not directly. but i'm sure that with blockchain, things will be different. somehow...
98% of the use-cases of a blockchain already have one or more central trusted authority, whose public reputation keep them moderately honest. They could use an ordinary distributed trust algorithm between servers to agree on each new root hash of a merkle tree (say, each server has its own sub-tree of changes, and they all agree on a node containing the hash of each and the previous root), then use classic digital signing to all prove that the new root is authentic since everyone has signed it, and make all of *their* transactions available via torrent so that third-party archivists can keep copies. Then no company needs to hold on to their competitors' transactions, but the ledger is still immutable, and the full history available to the public. And all you have to do is discard a handful of unnecessary parts of a blockchain to get to its core technology, and mix in *better* technologies, more suited to the problem at hand.
With regards to his last point: we do have these access control systems for distributed ledgers that have "smart contracts". Many "decentralized finance" programs have programmable access control that allows infinitely complex access control schemes. Most common is admin assigning things like "can write to this data section", "can read this data section", etc.
Source: I have written many systems that do this!
All these promises and I've yet to see it solve a real-world problem better than the solutions we already have.
Also, biggest thing after Web? Deep neural networks? Virtual Reality? Very, very strange claims.
Yeah, still not convinced. Blockchain is the classic solution in search of a problem. Yes you could use it to replace a database under the control of a single entity, but permanence and need for devolved storage is a curse as much as a blessing. With ever growing storage needs and a need for volunteers to host copies of the data store, you are really going to have to store stuff that there is a need to "live forever". Will anyone really want to commit to hosting and maintaining multiple copies of high volume and quickly outdated data like insurance contracts? Seems sus to me.
I'm on the programming side rather than the IT admin side, so maybe i'm wrong, but i don't see how the traditional role model can't work. You create a permanent role for "External Surveyors", you allow members of this role fo add further members to the role, everyone that has those permissions to access that data are all under a single role and have a script set to run to purge the role of all userd on a given date.
Again, i'm not the admin side so i don't know if the feature exists in any of the current solutions for members to add other members, but as far as the model goes, even if it doesn't currently exist, it could be added. I don't see how this isn't a usable solution.
Of course highly subjective but I can think of revolutionary technologies in IT with a LOT more impact than blockchain:
- personal computers
- public key crypto
- neural networks
- unicode
- version control
- iot
- databases
- search engines
This sounded like finding the "solution" first based on fashion and then searching for problems that don't really fit very well, (and for which existing dedicated solutions exist.)
I think we should be updating our antitrust laws
Blockchains don't add any value in the scenarios covered. They can be a useful internal data structure as a form of database for obscure distributed transactional problems like bank transfers, but the public scenarios described can all be implemented without blockchains, require off-chain infrastructure (e.g. contract law) to function and adding blockchains doesn't improve them.
This is an unusually low quality video for this channel and I'm astonished they let someone geek out over technical solutions to legal problems who doesn't seem to understand the legal problems themselves and doesn't consider the ethical and social implications.
The tradeoff for blockchain is efficiency. Most transactions value efficiency over decentralization. That is the easiest way I could explain why we don't use it for many things to someone without a technical background.
Saying that blockchain is one of only two revolutionary inventions puts this guy in first class on the hype-train. Computing has completely transformed the world around us in the last 40 years. So far blockchain has managed to find application in a few new currencies that are nine parts scam one part waste of energy.
Blockchain is an interesting idea but nothing presented here couldn't be solved better in another way. The shared data application was perhaps the most compelling case for me but it would almost certainly still require a third party to be involved in the real world. The issue of the third party (or one of the other companies involved) exploiting the data would still have to be covered by contract and would be better solved with a change in the law.
and what would surprise many is that some L1s actually do store their data on centralized services like AWS anyways...the distributed ledger space is a disappointment
Up until now the main problem of every block chain idea is scale. You need a certain amount of users to get the tamper proofing, but the bigger the system gets the less effective it becomes. Its nice that Proff McB starts of claiming BC as one of two revolutionary inventions so that you instantly doubt anythingh he says.
None of this addresses basic errors - either human or technological - which will, without question, be addressed by people.
If the blockchain, for whatever reason, mistakenly "promises" that the bank owes me 17 billion dollars, does anyone think the bankers are going to throw their hands up and say, "welp, a promise is a promise, pay the guy!" No. They will fix the error manually by whatever means they need to.
So that incontrovertible "promise" only means as much as the willingness of the stakeholders to uphold it.
Seems to me like all the benefits of distributed ledgers can also be gained from CA’s and git and Access Control the way described can be done today using sql Databases. Therefore I’m having a real hard time understanding why I need a distributed ledger to do all this in a much more bloated way.
SQL can do these things much much more efficiently, but it also has the downside of being centrally controlled. True decentralization is the major technological strength of blockchain. If implemented with careful design, a blockchain can defend against bad actors from any vector. This is a novel innovation.
Because whoever offers the database can alter it. Also, whoever attests keys/signatures can lie
@@dilldev but it can’t though. Like the problem with blockchain is that it provides more securities against man in the middle attacks and other “code” based attacks like that. But it provides less security for actual attack vectors that most scammers use in the real world. You know, like basic phishing scams that just grab your login and steal your shit.
With a centralized system, that attack vector can be compensated for. With a blockchain the ledger can’t be changed and reverted so easily. And the problem is those kinds of attacks are much much more common
Blockchain has been around since the Bitcoin white paper was published in 2008. That's 14 years we've had to find a practical non-planet incinerating ponzi scheme application and it still hasn't happened. I love the idea of digital currency but I'm not very optimistic about blockchain as the solution at this point.
@@BlazeMakesGames Even in a decentralized system, there can be a large consensus about reverting something and it could theoretically happen. Especially in the example where it's a blockchain specifically controlled between a few companies for their trustless data exchange. They could come together and hardfork if something like that happened. There is always some kind of solution.
How are any of these uses "alternative"? Banking, insurance, employee diploma tracking, stock trading, all tools of opression in service of the financial elites.
If im understanding right, that last section was basically "unfortunately everyone with the leger can read everything on it so every thing i said won't actually happen until that's fixed"
Really not mentioning Neural Networks as a revolutionary technology? If you haven't been asleep this year, you will have heard about GPT-3, Dalle-E 2, Imagen, Stable Diffusion, OpenAI Whisper,... all of which would have been unthinkable a few years ago. AI 'understanding' vision enough to generate any image you can think of, write perfect transcripts from audio, and solve math questions etc. by just being trained on a giant amount of text.
Maybe a few years ago, I would have seen the technology more favourably, but over time, it's been used for too many scams, and when it's not used in a scam, it's merely used as the hot new technology just for the sake of it, or for the sake of investors who are excited about buzzwords. On a technical level, those applications, they would be better served by non-blockchain solutions. Cryptography is fine, and sure, given enough time it can be cracked. Blockchain is fine, but not when the network is controlled by a singular entity (e.g. government funded / controlled botnets). The former is more energy efficient, so why use something else? If it ain't broke, don't fix it.
Amazing how smart people can find alternative solutions to problems that don't need fixing.
look up zero knowledge proofs.
... Central Database with extra steps? The communication to that database can be secured, the database can be secured (read encrypted). The data (read sql file with extra steps) can be corrupted the same way the encrypted database is. By corrupted computer/program that "miss reads the database/sql file"
Are the companies using the blockchain to prove that they are not a cartel? If they are a cartel, wouldn't they just lie in unison and then sign that?
All of the legal uses suggested would still involve traditional contracts and lawyers with the possible addiiotn of legal expertise regarding distributed ledgers. And since blocks can't typically hold any large amounts of data, the blockchain isn't a way to store it and instead you would most likely just store a link to a centralised server anyway. That link can be shared without adding anything to the chain and now we can't prove anything. All the blockchain can do is show that falliable humans sometimes did things.
Blockchain is not even remotely taper proof (51% attack). Should I keep listening after 30 seconds?
Every time I hear someone talking about blockchain, they never sound confident or are never clear with the benefit with good examples.
What can blockchain do? It can replicate SWIFT, and it can replicate insurance contracts.
I'm failing to see the big step forward here, especially since all the financial blockchains have proved themselves vulnerable to losses, glitches, irreversible errors and attacks by sufficiently motivated groups. Nothing presented here seems to mitigate those issues.
I was truly hoping to hear a good use-case for blockchain. I am once again disappointed.
If a group of companies wanted a third party to handle interchange data, they could easily form a separate entity composed of elements from all companies, vetted by all the companies or an agreed arbitrator/third party.
I didn't follow the data access rights part at the end, since it again seemed to be replicating systems that already exist.
If you can't understand the benefits of a publicly verifiable immutable ledger, you should re-examine your biases. DPLs allow for electronic voting, which was thought to be impossible before. Nearly every big tech company is working with DPLs with the notable exception of Google (but i suspect there are some x projects that are being kept private).
The point isn’t that blockchain can replicate specific services but that it can perform the tasks of those services and do it in an uncensorable, unmutable way. The smart contract executes without permission from a government or a CEO. SWIFT can be blocked externally.
You’re right that many blockchains have been hacked, but the weakness is always either centralization or poor coding. Some have proven their security by preserving a huge potential honeypot over many years.
So blockchains are a "revolutionary technology" that allows oil companies to avoid regulation?
(also, I'm not convinced blockchains actually solve the stated problems, nor are they the "only" way of solving them)
I guess it's nice to hear a blockchain enthusiast say the quiet part out loud for once: they're built to amass wealth and escape the law
@@squishy-tomato you can comply to the rules while still breaking the spirit of them.
As an example, a restaurant with a buffet usually will not have a written rule stating that you're not allowed to trash food. It is assumed that every person will create some food waste because they are going to pick up things they won't eat.
Yet, those same restaurants do hold the right to kick out patrons that they find who are abusing the system by denying access to said food to others.
You can compel a company to be forced to use a blockchain and add information into it. But how do you get them to confirm that event XYZ was actually added to that chain? As long as everything reconciles, then everything looks legitimate even if event XYZ is an oil spill.
It never gets added to the chain, and thus it "never happened" officially even though there was witnesses all around
@@Solinaru This is a tangent. The example given was clearly to adhere to regulations in a formal and verifiable manner.
Do we really want insurance information to be on a blockchain? Do people need to know what addresses have expensive, enumerated items and which do not? Gosh, I know if I was looking for people to rob, I'd definitely use this neat blockchain to keep track of where to hit.
This video is like 90% fluff and almost nothing substantiated.... Besides the points other comments already did make: Why are we suddenly talking about the salaries of "blockchain engineers" and how a company that had something with blockchain in their name had their shares skyrocket in a video about "Alternative Uses for Blockchain"?
I'll give the benefit of the doubt and guess, that this video was done without any script on the professor's part? I'm left with more questions than answers with this one, and not in a good, sciencey way...
"Blockchain solutions are often much worse than what they replace." - Bruce Schneier
"To me, the problem isn’t that blockchain systems can be made slightly less awful than they are today. The problem is that they don’t do anything their proponents claim they do." - Bruce Schneier
"By its very design, blockchain technology is poorly suited for just about every purpose currently touted as a present or potential source of public benefit." -- Bruce Schneier
"From its inception, this technology has been a solution in search of a problem ...." - Bruce Schneier
I don't get the example that he's making. What would be the data in this case? And why should everyone posess it, but noone should know the data of the competitors? The ledger should be shared after all
Dear Computerphile;
Thank you for being the best channel on TH-cam. It is because of this fact, and only this fact, that I won't be unsubscribing after hearing you innanely describe how blockchain could be used to help oil companies skirt regulations. Thank you.
The two most significant inventions in the transportation industry are the automobile and the wiper.
Blockchain is the unwanted residue of Bitcoin.
When people tell you that blockchain is a technology in itself they are scamming you.
Computerphile's quality is really, really slipping. I heard a lot of tech-marketing terms, but no actual explanation about how blockchain technology solves any problems that it didn't create itself.
This video is indeed poor quality as it suggests blockchains are inherently "tamperproof" which is a lie or extremely misleading. His suggestions of use cases is also absurd and reflects he doesn't understand the technology well. To answer your question as to what use cases Blockchains solve it provides censorship resistant consensus creating non fungible tokens(best for money like Bitcoin) . Furthermore blockchains without Proof of work is simply used for marketing or out of ignorance and blockchains are inefficient by design (for censorship resistance) and have very narrow use cases.(Not a panacea for all problems)
@@sanisidrocr And you are claiming you know more than a professor who worked on actual research projects for years while you mix up blockchain as a technology and cryptocurrencies? Maybe you watched "Line goes up" too much.
@@BlueTJLP I am extremely knowledgeable being a developer working on Bitcoin for many years, but I can only judge his lack of knowledge based upon the statements in this video as I don't know him personally. Perhaps he was edited out of context
@@BlueTJLP and you are claiming that a professor is an omniscient god?
@@BlueTJLP there's a professor that claims that he can cure autism with his bone marrow and has done studies (ie, made up studies). Since I have never held a doctorate in anything, I have to believe this professor is correct. 😂
Well at least we've moved on to *alternative* uses for blockchain... Still not quite the final stage, but getting there.
Experience has already shown the claimed benefits like being tamper-proof are not durable, since major chains will fork in response to, for example, a large heist. Which fork prevails ultimately comes down to the kinds of influence and manipulation the technology claims to address, and worse, it's actually more vulnerable than existing mechanisms to those influences. After all this time I've yet to see anything blockchain does better than existing mechanisms other than cryptocurrency, and cryptocurrency isn't even really a currency, more like a speculative asset class with uncertain legal protections. As a currency it's mostly useful for things like money laundering, and I'm not going to argue with the legitimacy of someone trying to eg launder money out of mainland china, but it's not good for much outside of speculation and questionably legal transactions like that.
I expected this video to start and end with someone saying "none"
I always held computerphile in high regard. Until today. In fact, this might be the first time ever that I use the 'thumbs down' button on any youtube video.
This single video won't make me unsubscribe, but I will unsubscribe if we get more of this shit in the future.
So when are we getting the followup video of someone absolutely demolishing all of the points this guy attempted to make?
Like people need to face it, the Blockchain as a concept is a failed idea. The core problem people need to realize is that the Blockchain cannot explicitly do anything that can't already be done through other means. It is essentially just an alternative method of record keeping. Either you use a more traditional system where you Trust some specific centralized organization to keep track of your records, or you use a Blockchain technology to provide a solution of maintaining 'trust' between many parties without the need of a centralized organization.
No matter how you write the code or what features you try to slap on top of it, it all comes down to whether a Decentralized system is better or worse than a Centralized one, because that is the core of where its features and drawbacks will arise, and these are things that cannot possibly be solved. And Frankly I think it's objectively worse in virtually all use cases.
Like if you want to talk about data that is relation to any kind of organization, then why would that not be centralized with that organization? Like talking about insurance information, why would that not just be stored with the Insurer? There is literally 0 benefit from having that information be on a public distributed ledger instead. It's basically going to be centralized to that Insurer no matter what you do anyways. Movie Tickets is another popular example people use for some strange reason but again why would that not be simply just be controlled fully by the place you're buying tickets to? What benefit does the company gain from losing all control over who actually buys and trades their tickets around? This also rules out virtually all Gaming applications and the like.
So if we rule out literally all the applications that involve some kind of company or organization based application, that leaves us with not a lot of possible applications left. Obviously money is one but we already know how horribly that all works out so lets not do that. In theory you could argue Votes are another application that could maybe work as a way to do things like at-home online voting. But lets be real that opens up so many vulnerabilities that would make them about as secure as Trump *thinks* mail-in votes are. Another one I've seen brought up is Medical Records, but again, why not have hospitals take care of that and B) blockchains are sorta inherently public. So having a system where literally all of your personal medical information is out in the open for the public to see at will is kinda probably not what people want.
And one of the other core issues that does involve actually talking about the tech a little more directly is security and privacy.
The issue of privacy on the blockchain is weird, because it's one of those things where everything is perfectly private until it's not. The issue is that all the information on the ledger is inherently public. If anyone can store a full copy of the ledger on their computer, then that means that anyone can look at the contents. So you can't possibly hide the actual information itself. After all even if it's encoded, every user needs to be able to decode their own information, which means it's trivial to reverse engineer that to decode any other user's info. So the only thing that is unable to be deduced directly from the ledger is exactly what data belongs to what person. But you can probably immediately imagine how that probably won't last for long. If we're talking about things like medical info for example, and you want to dig up some guy's full medical history, then you could just simply search the ledger for certain pieces of info you already know, and then pretty quickly deduce by narrowing it down to find the person you want. And I mean big corporations will be able to exploit this the best. I mean they already have complex algorithms that are capable of crawling through mountains of data. It'll be trivial to use those to figure out people's identities on the blockchain based on existing data they already have and match everything together. And once your identity is leaked on the blockchain, it's gone forever and people can track not only everything you do from then on, but also everything you did in the past as well.
And then when it comes to security well that's the grand irony of blockchain technology. This is essentially a system invented by coders, who only think about things in terms of code, and not about how things actually play out in the real world. Blockchain tech is based around the idea of trying to maximize possible security from the kinds of attacks you'd expect coders to think about. It's basically impossible to interfere with by 'hacking' into someone's machine and trying to change some numbers around, or intercepting a transaction and scraping some off the top. But the fatal flaw is that they ended up making a system that is infinitely more vulnerable to PEBCAK issues.
When you actually look at how real world hacking works in about 95% of situations if not more, usually 0 code is actually involved. Why go through all the effort of trying to hack into someone's account when you can just call them up claiming to be from the password inspection agency and just have them tell you their login information. If you hack into someone's computer you don't need to bother trying to hack some numbers when you could just use their saved passwords on their browser to log into their account and then completely 'legally' (at least according to the blockchain) transfer all of their assets to yourself or modify them however you see fit.
Not to mention that simple mistakes like typos can be potentially irreversable. What happens when a doctor accidentally types that someone's prescription is 225 mg instead of 25 mg and it gets saved to the blockchain. If it's a situation where the doctor has the authority to fix that mistake, then why are we using a blockchain? And in turn why would we want to use any kind of system where a doctor *couldn't* fix that mistake?
Like lets face it, the idea of a distributed uneditable ledger sounds cool, until you realize that humans make mistakes, so being able to edit those mistakes is a generally good idea, and that kind of thing requires a centralized system of trust to manage it.
No actual novel use of the blockchain has materialized in the decade and a half it has been around, if crypto currency is not counted. And even that is having a hell of a time staying afloat these days. Revolutionary technology does not behave that way.
If adding “blockchain” to your company name raises the share price that dramatically, it’s not exactly convincing me that blockchain isn’t a scam.
And what happened after that? Wikipedia:“On February 22, 2021, the SEC delisted Long Blockchain Corp's shares, saying that the company had not filed financial reports since September 30, 2018, and that it never completed its planned transition to producing blockchain technology.” So definitely not a scam then, when the word blockchain is more valuable than the actual blockchain.
The only reason this video shouldn’t be deleted is because it’s useful context for the video you’ll have to make apologising for it
Not sure what your point here is related to the technology. A shady company used the hype in a bull market to pump up their share price from algorithmic trading bots. That has nothing to do with the technology. It's a scam by the company itself. Look at you getting all pissy and angry.
@@BlueTJLP if the only example of success is a scam, that doesn’t look good for the rest of it
Git was first released in 2005, bitcoin in 2009, so it was not the first implementation of a distributed ledger. Git is also my favorite example of the blockchain idea being used in an actually useful way, without all the hype and destruction of ressources.
Maybe blockchain should find a primary purpose first?
Commonly called a solution looking for a problem, the build up was interesting, but from about 9 minutes it was a ramble. Not been commissioned by a large petro chem to investigate by any chance?
If blockchain was useful, then actual boots-on-the-ground programmers would be using it to great effect. Over a decade since its creation, this still hasn't happened, because blockchain doesn't provide any benefits that can't be achieved better, faster, or more efficiently through other means. And shame on Computerphile for trying to blow this smoke up our exhaust fans.
One of the big problems, is how do you store and protect the 'ownership' of a persons data on a blockchain? Then, what happens when that gets lost. You either get consolodated trust on a platform that stores it on a persons behalf, so you loose the benefits of the distributed trust. Or, they loose the ownership, completly, potentially to another person is there is fould play. There is no way to retreave it. I havent seen a good solution to this yet.
I'm particularly confused why blockchain needs advances in access control more than any other form of information. Is it because blockchain necessitates that everyone be able to see the whole ledger or else the premise of decentralized trust falls apart? Sounds like blockchain just shouldn't be used in instances where access control is complicated.
So the only use case is to facilitate mega-corporations in circumventing antitrust laws? Cool.
And this is the 2nd best thing in computer science next to the web?! I guess he hasn't heard of neural networks and AI. I'd say AlphaFold for instance is much more significant than running git on a torrent network.
So, if you're developing a distributed ledger 'solution' without crypto, what are the incentives for running nodes and mining blocks? Making mining easy is dangerous because of hostile takeovers, so the wastefulness of mining is inherent to the system, right? You can't run a tiny blockchain with a couple of users...the safety of it comes with scale.
I'm disappointed to see this on an otherwise serious channel such as this.
"it wont be altered" congratulations you just invented checksums
This is definitely, BY FAR the worst video Computerphile has released in the 3+ years I've been watching the channel.
What you described sounds exactly like the baseline protocol
Expect more from Computerphile this is truly disappointing.
The truth is the world runs on mutable handshake deals built off trust. An audit trail is always good enough for tracking changes because there's a mutual trust. Blockchain provides no benefit and in reality, provides a mess of complications because of its immutable and computationally expensive nature.
I’ve done some RBAC work before, but I’m not sure that his “dynamic, delegatable, revocable access control” is possible with a distributed ledger. Isn’t the whole point of the ledger that, in order to verify the current state of the ledger, you must have the full ledger? And if an entity has the full ledger to verify, how is access to that ledger revoked? Access to FUTURE STATE may be revoked, but any access given on a node of the chain would be impossible to revoke.
A lot of this stuff is solved by trusted third parties (like the French University example).
Would like to hear about tha Acess Control. Thank you.
No expression of "Intent" within blockchains,
Rules therefor will not serve intent, and there is nothing other than authority to maintain the rules, so there rules will not hold (unlike current gov rules that are based on intent).
use a different consensus algorithm challenge (impossible)
blockchain isn't a consensus algorithm.
"Its revolutionary!!" Here's a single very very specific example of use that has working examples we have already solved with existing technologies.
I think blockchain is interesting and once the scams and hype have run their cause it can have uses, but this talk that its as important as the web? And that is the best example he could think of?
Agreed. The best example of a blockchain is money like Bitcoin which is far from perfect but does solve multiple use cases. The most important innovation in a blockchain and why blocks in a blockchain exist is Proof of work which he fails to acknowledge or even understand.
@@sanisidrocr the only use of cryotocurrency is burning oil and gas
@@tomasg920 I agree that Bitcoin does provide a valuable service for the environment burning waste methane but most energy used for bitcoin these days is green(nuclear,hydro,wind, solar, thermal)
In my humble opinion: Few revolutionary technologies in the past had to be "explained" as to why it one day will be revolutionary. It just happens organically when a technology is truly revolutionary... from the very first cars, to the transistor, the television, the microcomputer, the internet, cellphones, GPS, etc., These technologies naturally saw explosions in use and development, up to a point where we can't see our lives without it. Lets take an example of why I think block-chain tech is over-hyped (again imho): While the internet is incredible technology, HTTP is just part of the technology stack, and so are the networks, databases and the servers running it. Block-chain is just part of a technology stack, it is about as revolutionary as a new type of database or network. Sure it may be extremely valuable within its domain, but to the average person it is meaningless. So it is not useless, but certainly isn't part of only two revolutionary technologies in his lifetime, that is just not true.
Its just not useful. Not yet, at least. Comparing it to the web is almost insane. Contracts, by almost definition, must be mutable. While the idea of a rigid contract is attractive towards people who work with computers, in actuality, that is a dangerous.
I think, ultimately, this is a philosophical discussion more than it was a technology one. And I suspect that prototype project with the bank has no relevancy whatsoever with this guy thinking this is a revolution and worthy of discussion on this channel.
Wow, seems like this dude completely missed out on the torrent technology or he is drunk on the blockchain rhetoric. 10 years before Bitcoin appeared, torrents were and still are doing peer to peer sharing of data tamperproof.
I think blockchain does have the ability to solve *some* real world problems, but the use cases are pretty *narrow* . The real problem is people are trying to shoehorn blockchain into their projects because it's cool, without realizing that it's not the best tool for the job. I respect all of the developers and scientists working on blockchain and smart contracts though.
I wonder if there will ever be an implementation of these ideas that don't involve planetary incineration.
Also, he framed the blockchain as tamper proof, which is not something one should say, knowing that it is a battle of computing power.
If quantum computing ever becomes feasable for a broad range of applications, bad actors with enough money (e.g. countries) will have a field day.
Wow. I always thought this was a legitimate, high quality channel. Guess not.
For once I feel an episode of computerphile is fluff which is unfortunate, block chain is a fad with no real commercial benefit, might be great for a file sharing torrent site but meh otherwise, just day dreams about process models.
Immutability isn't always a given. A few years Accenture was granted a patent for an editable blockchain and whenever quantum computing hits the commercial market place we may see that ability to be exploited against other blockchains...
Computerphile you're WAAAAY better than this. How did you go through making this without processing the fact that his message doesn't line up with how the technology is even being developed!
What are you talking about? I'm going to assume you're one of those lemmings who have no experience nor technical expertise with blockchains and crypto.
@@user-zu1ix3yq2w care to elaborate?
@@rafflezs The professor gave multiple examples of immutable ledgers being used in the real world lol
Cryptobros are the absolute worst cult. In the video we have a leading expert in this subject matter, and in the comments we have a clueless muppet claiming he doesn’t know what he’s talking about.
“From a technical perspective, many of the criticisms are valid”
Trying to manufacture trust between a group of people who inherently don’t trust one another is a fundamentally wrongheaded idea.
Good one .Next video he should discuss security vulnerabilities and challenges
Anyone who says the invention of blockchain or remotely on par with the invention of the internet isn't worth listening to.
You want us to unsubscribe, huh?