6th HLF – Hot Topic: Blockchain and distributed ledgers
This is a modal window.
The media could not be loaded, either because the server or network failed or because the format is not supported.
Formal Metadata
| Title |
| |
| Subtitle |
| |
| Title of Series | ||
| Number of Parts | 37 | |
| Author | ||
| License | No Open Access License: German copyright law applies. This film may be used for your own use but it may not be distributed via the internet or passed on to external parties. | |
| Identifiers | 10.5446/40154 (DOI) | |
| Publisher | ||
| Release Date | ||
| Language |
Content Metadata
| Subject Area | ||
| Genre | ||
| Abstract |
|
00:00
Internet forumMusical ensembleControl flowQuicksortType theoryCASE <Informatik>Machine visionPoint (geometry)Data analysisLevel (video gaming)Perspective (visual)BitFood energyPhysical lawCentralizer and normalizerView (database)Medical imagingDefault (computer science)Partial derivativeAlpha (investment)SoftwarePhysical systemPower (physics)Multiplication signAverageCartesian coordinate systemMechanism designGame controllerOpen sourceNumberEndliche ModelltheorieInternetworkingData managementCommunications protocolBookmark (World Wide Web)XMLComputer animationMeeting/Interview
05:08
Internet forumBit rateBusiness modelComputer scienceElliptic curvePublic-key cryptographyEndliche ModelltheoriePhysical systemFlow separationProduct (business)Right angleMereologyIdentity managementPoint (geometry)Multiplication signMassCartesian coordinate systemInternetworkingExterior algebraProjective planeSoftware testingIterationParallel portAnalytic continuationProcess (computing)Real-time operating systemComputer configurationContext awarenessOcean currentWeb pageWeb 2.0TelecommunicationMoment (mathematics)Surface of revolutionQuicksortCASE <Informatik>Error messageVideoconferencingSoftwareProgramming paradigmSmartphoneLaptopTheorySlide ruleDefault (computer science)Standard deviationDataflowBit2 (number)Social classComputer fileMessage passingAndroid (robot)10 (number)AdditionGame controllerMathematicsCryptographyComa BerenicesFile systemMeeting/Interview
12:26
Internet forumBit ratePhysical lawSoftwareEquivalence relationSupercomputerCombinational logicState of matterComputer-assisted translation1 (number)ResultantAlgorithmType theoryPhysical systemQuicksortOpen setMechanism designPosition operatorEndliche ModelltheorieRepresentation (politics)X-ray computed tomographyCore dumpParallel portProjective planeReading (process)Category of beingDifferent (Kate Ryan album)Medical imagingNegative numberBitFacebookPredictabilityLevel (video gaming)Programming paradigmComputer fileComputer architectureIdentity managementData storage deviceComputing platformMathematicsRow (database)HypermediaCartesian coordinate systemInformation privacyMobile appInformationUltrasoundRight angleInheritance (object-oriented programming)Execution unitTexture mappingNP-hardNumberPoint (geometry)Sound effectMeeting/Interview
19:43
Internet forumLine (geometry)CryptographyPresentation of a groupBitMereologyMultiplication signDifferent (Kate Ryan album)Term (mathematics)Frame problemCASE <Informatik>Right anglePoint (geometry)FeedbackCross-correlationCorrelation and dependenceMusical ensembleMathematical analysisParameter (computer programming)Basis <Mathematik>Endliche ModelltheorieCentralizer and normalizerSoftwareSet (mathematics)Physical systemCore dumpSocial classUniqueness quantificationAtomic numberFraction (mathematics)Profil (magazine)Token ringElement (mathematics)CausalityVolume (thermodynamics)TwitterStudent's t-testInterpreter (computing)Form (programming)Lecture/ConferenceMeeting/Interview
27:00
Internet forumEndliche ModelltheorieOvalDifferent (Kate Ryan album)Token ringFeedbackSoftware developerRight angleCommunications protocolDirection (geometry)AuthorizationPhysical systemMultiplicationIntegrated development environmentTerm (mathematics)IterationExtension (kinesiology)Point (geometry)CASE <Informatik>WordParameter (computer programming)Lattice (order)Presentation of a groupTheoryProjective planeShared memoryInformation securityPrototypeOSI modelMeeting/Interview
31:38
Internet forumDifferent (Kate Ryan album)Natural numberUltrasoundBlock (periodic table)ComputerCASE <Informatik>System administratorWhiteboardComputer programmingStudent's t-testDesign by contractPhysical systemCodeNatural languageSpacetimeInstance (computer science)Group actionValidity (statistics)Cartesian coordinate systemObservational studyInformationPoint (geometry)Information privacyKey (cryptography)Digitizing
35:11
Internet forumBit rateDatabaseRight anglePoint (geometry)FacebookMultiplication signMassMusical ensembleQuicksortContext awarenessWebsite2 (number)Physical systemBackdoor (computing)NumberDivisorBlack boxData managementOpen sourceAlgorithmPay televisionGame controllerDataflowInformation privacyInformationCartesian coordinate systemSoftwareField (computer science)Scaling (geometry)Standard deviationCentralizer and normalizerCASE <Informatik>MereologySpectrum (functional analysis)Term (mathematics)Structural loadAuthenticationChainDatabase transactionFormal verificationBlock (periodic table)Address spaceObservational studyNatural numberEnterprise architectureSoftware developerNP-hardMaxima and minimaHybrid computerUsabilityOcean current
40:34
Internet forumXMLComputer animation
Transcript: English(auto-generated)
00:24
Everybody, there's still a little bit of energy for our last panel today, it was a hard marathon without the coffee break. I'm happy that the room is still full and so many people are still interested even without coffee in our topic.
00:42
A short introduction of the new panel, which is the session, the application orientated session. To my left we have Mihai Alisi, he's founder of Akasha and founder of Bitcoin magazine and co-founder of Ethereum.
01:02
Akasha is a new type of social network based on blockchain technology and his vision is a free and really decentralized internet. In the middle we have Demelza Hayes, she's researching at the University of Liechtenstein about the role of cryptocurrency in asset management and I think she's one of the persons who has a Bitcoin wallet.
01:23
So we can maybe ask a question again from the last panel. And on the right side from your point of view is Dexter Headley from University of California. He has a medical background, he's working with AI and cancer imaging and he has big hopes as well on blockchain technology
01:45
for helping to manage the data, the private cancer imaging data he needs for AI users.
02:01
As I asked the other panelists starting around what is the use case you are most excited about, I think maybe the answer is clear. So I would like to add to that question what is the use case you are most excited about and why is blockchain really a perfect solution for your use case?
02:25
Let's start with Dexter. Sure, thanks. Clearly I think health has the most, at least in my world, to benefit from a distributed decentralized sort of system. In America, probably more so than in Europe, we have more socialized health systems and
02:45
a capitalist health system is very easy for bad players to take control of patients' data, which is just incredibly wrong on a number of levels, but from a scientific perspective it stifles innovation.
03:05
So whereas I see in business, AI is sort of making a big deal in search and advertising all over the place, in medicine, by law that is sort of not the case, because technologies like distributed ledgers, they don't exist yet, sort of, or they haven't existed yet.
03:27
So I see blockchain, distributed ledgers, this idea of democratized data as the main benefit to healthcare. Tim Alpha, what's your favorite use case application?
03:43
Great, okay, well definitely I think the open source aspect of public blockchains basically allows each cryptocurrency to be its own experiment in monetary policy. So for the past 800 years or so, we've had tons of centralized currencies and basically two professors from the US, Reinhardt and Rogoff,
04:05
did a data analysis of these currencies that have existed and in almost every case it ended in hyperinflation or default or partial inflation, where basically the purchasing power of the currency was debased over time, so this means that the average people in society lost their savings.
04:24
And I think that with cryptocurrencies we can see right away that the cryptocurrencies that had inflationary mechanisms did not gain networks of people who used them. So for example you had a Frei coin, which was a coin that started in Germany early on,
04:40
it did not gain a very strong network because it had dimmerage embedded in the protocol, where basically the currency just lost its value over time. And I think that what's going to happen over time is not that, going back to Mohan's point, it's not that people want to get rid of our current existing system, it's simply that we want to add competition to it.
05:05
So now what we're going to do is we're going to allow this technology to experiment with other models that might not be the current models, and what that's going to do is if those models have some beneficial features, the current system can adapt. And overall I think we're going to have a higher quality in currency and in stock markets and several other financial applications.
05:28
Mihai, you're… Hey, does this work? So I'm going to echo some of the points made by Silvio on the previous channel, on the previous panel. As in self-governance is a big one, especially if you think of this setup
05:44
where for the first time in human history we have over half a planet connected, interconnected, although we have these intermediaries in the middle that can easily manipulate the masses. And we saw in the early days the gentle push to buy a product and then a gentle push to vote left or right and so on.
06:06
So it's a big part of our society and how it can influence our future and the identity. And when you put the two as in self-sovereignty, owning your identity or data. And when you combine these two there is like this, I don't know, utopia in a sense.
06:24
But at the same time it feels that for the first time we have the right pieces of the puzzle to enable us to experiment towards finding what works and what doesn't. And if you look at the bigger picture for the first time in human history we are able to test and iterate super fast many parallel governance, identity and economic systems
06:43
that will or might become very relevant in the near future, who knows. But if you look at the current economic system and how everything is hinging on, you know, is this year going to crash or is it the next? It's not a matter of when things will go bad, it's rather a matter of when.
07:04
And having an alternative and an option to the current system doesn't seem like a bad thing on the contrary. It's in our own interest as a society to have an alternative in case things go bad. And, you know, you had a clear example with Cyprus and other economies where people,
07:21
which they thought they had their money and so on, were effectively cut from their wealth. You know, you had revolutions and everything. So that's, for example, an application where I'm excited about as in offering people the alternative when the moment or occasion arises. And hopefully when that moment happens we will be able to transcend that in a more peaceful fashion
07:46
than, you know, in the past where it was always revolution blood on the street and so on. In this case, hopefully, by having something there that could, you know, slide into the daily flow of the people that might, you know, offer a better way forward.
08:01
Maybe tell us a little bit more about your idea around Akasha and why it maybe helps to decentralize the Internet as well. Or why is the Internet centralized at all? Wasn't that the idea of the Internet to be decentralized and what went wrong? Well, it depends on who you ask, I suppose. But ultimately with Akasha and this odyssey of discovery, so to say, in the
08:24
past few years we started from like how can we make a blockchain application work. And we realized that there is this, again, complementary technology. So when you say blockchain, it's not necessarily just that. It's like you're having horse glasses and looking at just one piece of the puzzle.
08:43
You also have, for example, the interplanetary file system which can be combined. So, you know, one of the previous panelists highlighted how the interplanetary, well actually people were storing messages on the Bitcoin blockchain. Which is like super expensive and very inefficient.
09:02
So in this context you can merge the two and come up with all sorts of applications. And in our case it was a social network, right? And in this context where you have like self-sovereign identities and like this data spiraling around the user and user being in control of it and choosing with whom I want to share this data. You have like this emergent social network between the identities which are those interactions that in some cases
09:25
require trust, in some cases require semi, you know, some kind of trusted node connecting the two entities. And that's also where I think something that is exciting is especially this room and
09:41
the work done by these bright minds when it comes to mathematics and computer science. In some cases we are looking at decades old mathematics, like for example the elliptic curve cryptography which is at the heart of many blockchain projects. And in this context of blockchain or new, you have the opportunity of actually applying
10:03
some of these hidden gems lost in the footnote in history or on some Wikipedia page. And actually people without even understanding when they send that, you know, Bitcoin or whatever cryptocurrency from one side to another, they're making use of public private key. And it's I think, as Sylvia was saying, a matter of iterating and exploring what can be done with this.
10:25
It's almost like we are in the early 90s of the web. We kind of see the black text on the white background. That's like the digital money. And we cannot really imagine what can be done with this moving forward. And just like we had with the web, we had the continuous process of iteration and exploration in which, okay, we had the dot com bubble.
10:46
Yes, we had also the bust. But after that, we do have things that are completely changing the way we interact with each other. You have social networks, you have real time video communication, you have all sorts of things that maybe would not have been possible without that failure.
11:04
It's a trial and error, right? So with Akasha, what we're trying to do is basically an experiment driven project in which we are trying to figure out how to make things work using this technology by putting in the first place the human as in not being the product. And tapping to explore these alternative business models that otherwise, you know, just the
11:26
default internet standard is surveillance capitalism as in monetizing your data and so on. But in this paradigm, you also have the open, let's say, question of what can you do if you have these untapped resources around. So like if we take IPFS again as an example, we have these petabytes, like huge amount of data available in smartphones, laptops and so on.
11:49
That in theory, people could monetize by sharing it with the network. So in that case, you know, you already have a flow of wealth and redistribution in the society that even someone in Africa, you know, has on the
12:04
second hand phone, which is like if it's an S2 or S4 as in an Android phone, it's already quad core, might have a few tens of gigabytes. So I think where this is going is that people and entrepreneurs should be excited
12:21
about this to try and explore additional business models, alternative business models and also for scientists in this community to overlap with the, let's say, the more non-conventional blockchain crowd. Because it is in a way a very interesting overlap and like an application of some of the core ideas that were invented years ago, decades.
12:43
And by this overlap, hopefully we can turn some of those ideas in years into actually meaningful applications rather than, you know, decades. That sounds like it has many parallels to Dexter's idea or project with distributing health data or especially images which can help AI to classify cancer.
13:08
Maybe you can tell us a little bit more about how that works with blockchain technology. Yeah, definitely. So, whereas you're trying to put or Akasha is trying to put personal
13:20
information and, you know, baby pictures or whatever that you're free to put anywhere you want. Most people put on Facebook now. I mean, the idea would be put it on a decentralized storage platform. In medicine, you're not. In medicine, this idea of who owns your medical data, well, I don't. I can't change my medical record.
13:42
But I'm a custodian of my medical data, right? I can decide to give it to you and to you and not to you. So, there are a lot of parallels. IPFS, I'm putting, I want to put gigabyte files of data on IPFS. I don't know how much baby pictures are or whatever you would put on social media.
14:01
So, technologically, it's very similar. The architecture is probably exactly the same, but the application is very different than the need. There's a whole privacy law that has shut down innovation in medicine in America, and there's no way around it. I think blockchain, distributed ledgers, and even more so, I think as physicians, I mean,
14:23
can anybody name one killer app out of Apple's 1 billion, you know, apps for health? There's none. I mean, I feel like as physicians today, we have totally failed our digital patients. Why would they give us anything? So, I mean, my idea is kind of similar, very, very similar to yours. I
14:44
want to, at least, here's the mammogram you paid for, you might as well see it. It's yours. And if you can see it, then you can share it with me. I mean, the law says this can all go through today. It just doesn't because of bad actors and bad players. So, it's very similar. I totally agree.
15:00
And probably the same motives. Own your data. So, your idea is that patients can somehow keep their data or own their data and sell it to researchers or to Google? How many people in this room have had some scan done in medicine? Ultrasound, MRI, CAT scan? How many have seen it?
15:24
None of you. A few of you. The really hardcore ones. But the idea would be, well, you paid for that. Why haven't you seen it? Right? You pay for all kinds of stuff. You see it in medicine. There's, you know, you just, what happens to it? It disappears and somebody else profits from it. And that doesn't seem right. It just doesn't seem right.
15:44
Something maybe connected to this is also the idea of how Bitcoin beyond, you know, like the digital currency, you also have this concept where you have this big pot of money and some incentives that makes people move towards that thing, right? And the results almost a decade later from the simple idea with the algorithm distributing that unit, which, you know,
16:04
subjectively has some value, depending on who you ask, has led to this, like, equivalent of a supercomputer generated by these people that were basically just trying to get some of that, which is, like, more powerful than the top 500 supercomputers combined.
16:22
And it's not a super nation state. It's not one single entity that was able to generate that. It was a bunch of people. So, like, an idea here is that if we can use the same type of thinking to create the incentives and the right incentives to actually better humanity, right? So instead of just breaking shot to 56, maybe in this scenario, if there is some kind of incentive
16:44
where people, if they put their CAT scan and there is some value kind of, you know, pot of money, some of that value to be distributed towards researchers that would actually make use of that data and then feeding back into the whole system. So to sort of tie in to what you do, there's different values on the data, right?
17:02
So a true positive exam where, okay, they say you don't have cancer and you really don't have cancer, there's a lot of this, right? But if somebody said you have cancer and you don't have cancer, I mean, there's a lot of consequence to that and pain and suffering and that data should be valued more.
17:21
It's going to improve the model that's currently representative of radiologist knowledge. And the other as well, if somebody, if you're a patient who does not have cancer that somebody claimed that you had cancer, those kinds of data are inherently more valuable than the true positive and true negative value.
17:43
Who is going to value how much that data costs? It's sort of like a stock or a bond, how it performs is what its value is. And to me, a distributed trustless system is how we can value it and then have an exchange where this is open. It's private, yes, I guess I'm in the open blockchain camp here, but there can be an active marketplace.
18:07
I work with at least three or four companies that want access to my data set. And the way it works is you pay me money, I get to do whatever questions and answer whatever questions you have and give you the results. But I can never give you the data, right?
18:21
If there was a mechanism to have patients to directly be vendors of their own data, that would solve a lot of problems. And they're hard numbers, there's 40 million mammograms run a year. I just want five of them, just five million, just total. That's totally doable, but not by any one healthcare institution. That's not possible.
18:41
Just one short question for understanding, do they have more value, these false negatives and false positives because AI can learn more out of them or get more precise? Yeah, if you're learning what you already know, it's not as valuable as learning the wrong predictions, right? So yes, luckily for me, there's a lot of wrong predictions in breast cancer, but there are different levels.
19:00
The woman who lost her breast because of a false reading that she didn't need to lose is probably the highest category. She should be paid the most for her mammogram versus somebody who had a true negative reading, right? So there's different values, and I'm not technically enough to pretend to understand how it works, but I think of it like a stock. There's different performance and different data points, and they should be valued based on what people are willing to pay for them.
19:24
Something else maybe to add, since we're moving towards a future, maybe not so distant, is like when you have these nodes connected, like a network effect, be it the people that actually contribute to the images, the researchers that might actually use the fund to do the research, and maybe you'll also have this concept of crowdfunding, which is really easy in this paradigm because every identity is also a wallet.
19:45
You can have bottom-up built laboratories and maybe even hospitals in the long term, so that token or that can also represent the membership that can benefit the… Well, here's the problem. I work for a federal institution, and I am terrified to have an ICA, even though I have a legitimate cause.
20:04
Who knows if it's legal or not? I mean, I should probably talk to you after, but it's a thin line. You don't want to sell your reputation by just having another ICO in a big pot of shady ICOs. I'm not necessarily advocating to have an ICO, but I'm trying to
20:20
explore how the right incentives in an ecosystem can create interesting feedback loops that can actually bet the people that participate in get, at the other end, better off. In terms of health, in this case, not necessarily as in gains, monetary. Last question before I open to questions from the auditorium.
20:42
Demelza, we had this more or less philosophical discussion about value. I hear this argument quite often the last days from people here that cryptocurrencies don't have a real value. What is your comment on that? Okay, well, it depends on how you understand value, but in 1871, there was the Principles of Economics written by Carl Menger.
21:06
He basically wrote that value is subjective, so the way that you value an orange is going to be different than the way I value an orange. And actually, it's the fact that we have different values of that orange that is the basis for why we trade the orange.
21:22
So if we valued the orange equally, no trade would occur. So basically, with cryptocurrencies, it's basically a subjective value, and you can determine that value for yourself. And anyone that says it has no value, well, that's clearly empirically not true. So empirically, it has value.
21:41
And if it's going to have long-term value, that's very difficult to forecast. But I think, in general, the concept of having currency that is global and permissionless, I think it's hard to argue that that concept doesn't have value. In your talk, you made an interesting point that there was no correlations between stocks and cryptocurrencies.
22:04
Can you talk a little bit about that, why that happens, and what's your interpretation of that? Sure. I mean, one thing I didn't mention in my presentation is that if you do timeframe analysis, so if you change the dates that you pull data from, you can see that the correlation is increasing.
22:25
So the first couple of years, the correlation basically didn't exist, but if you just take the last year of data or if you just take the past couple of months of data, the correlation is slightly increasing over time. And I think right now it's increasing over time because there's actually more institutional investors that are joining the cryptocurrency market, and they're basically speculating on cryptocurrencies
22:44
the same way they would speculate on the stock market. However, I think that originally it was basically uncorrelated because it has a different governance model. So, let's say stocks and everything, stocks and bonds, they basically exist within a regulated financial market,
23:02
and they have central clearing houses that clear transactions, and they have insurance companies and different intermediaries that reduce counterparty risk, and it's a very different governance model. With cryptocurrencies, basically there's nobody backing it, and there's also no centralized system of clearing,
23:28
and so basically you just have a different set of risks. And so I think that originally the correlations were different because of the different governance. I mean, one aspect of cryptocurrencies that makes it extremely unique as a financial asset class
23:43
is that it cannot be confiscated. So, stocks, I mean, if you look at the DTCC in the United States, this is a central clearing house, and every single year they get sued by companies for allowing naked shorts, which is basically where the settlement time takes more than an instant,
24:04
so with atomic swaps on a blockchain network you can do instant settlement, but with the DTCC it's a centralized clearing house, and you have stocks where the investor will go and buy all of his stocks off of the stock market, and then the next day his stock is still trading.
24:24
And you ask, how is that possible? How is it possible? Well, it's because of the way the settlement works. It's a T plus three settlement date. So you have three days to give back the stocks that you shorted, and sometimes these shorters don't actually give back their stocks,
24:40
which is when it becomes a naked short. So these centralized clearing houses have been attacked for losing stocks for years, and basically what it actually is is it's fractional reserve holdings of stocks. So it's the same way that the banks work with fractional reserves. It's the same concept.
25:01
And of course these brokerages earn money for lending out stocks. Okay, so I think that definitely cryptocurrencies offer a very unique return and risk profile compared to other asset classes, and that's why it's uncorrelated.
25:21
Let's go. Just since we're on this point of the correlation between crypto and equities, is there a correlation between fiat currency and equity and stocks? We're both uncorrelated. This currency as a whole is just not correlated. It's very low correlation. It's very low correlation. In one peer reviewed paper done about a year ago,
25:41
they showed that there was some positive correlation between Bitcoin and emerging market fiat currencies. So I think that could be because a lot of the trading going on, a lot of the volume for Bitcoin is actually done in emerging markets. So I think we have at least three questions from the audience to start.
26:03
Yeah, hi. So to be as my, I run the HLF Twitter channel, and I have a question to the males and the others too. It just picks up on what you just said. So when an ICO, so initial coin offering by a company occurs,
26:21
essentially you treat this currency as an equity. The same would be, say, a company issues stock. So with having stock comes certain rights for an owner. So that's the first part of the question, and the other part is, so if a cryptocurrency in the form of an ICO is treated as a stock,
26:45
how does that relate to a cryptocurrency such as Bitcoin or Ethereum, which are treated like currencies, like money that you can deal with? So can you elaborate a little bit on the differences of those use cases of cryptocurrencies?
27:00
Sure. Okay, where to begin? Well, I was at a Chatham House meeting at Wharton about two weeks ago, and this was one of the main discussions that came up. And basically, currently, the investors and ICOs do not have clear rights. I mean, basically, the white paper can claim, so the white paper that they release is kind of like a prospectus.
27:23
It's not a prospectus, but it's some kind of document that gives investors an idea of what they're investing in. That doesn't actually give rights to the investors, and I think it's a very tricky situation because if the ICO cheats you, and you go to the court and say, look, that startup company cheated me,
27:42
well, the court system would say, well, why didn't you go, why didn't you invest within our existing regulated financial system? So you want to skip all the costs of our existing regulated financial system, but then you still want all the benefits, right? So I think that would be a very hard argument to make, and I don't see why the courts would actually need to enforce the claims that are the rights of the investors in this situation.
28:04
However, I think that the way that the U.S. is going, I mean, the U.S. is basically going in a direction where they want every white paper to announce, basically, what a prospectus announces, okay? And they're going to have, basically, they've all come up with different models,
28:20
but there's one model where they basically have you come to a sponsored agent who is registered with the SEC. That sponsored agent is going to look at your white paper, make sure that it has all the requirements, and then basically announce that you can actually launch your ICO. But if you don't get that sponsored agent and the prospectus approved, you're basically committing financial crimes.
28:44
And then as far as the way different tokens are treated, in the U.S., you have all these different financial market authorities. So you have the SEC saying this is a security. You have the Commodities Futures Training Commission saying this is a commodity. You have the Federal Reserve saying this is a payment system.
29:08
So you have basically different ways to look at this technology. And I think that, again, this goes back to subjective value. How do you value it? How do you use it? Many people use it for multiple use cases, and I think that it's very hard to fit into any kind of existing regulatory environment
29:28
because it fits so many different policies. And I think the one final point to make on your question is that despite the extreme volatility, the well-known cases of fraud,
29:43
people still prefer using it. So what does this say about our current system? What does this say about our current financial system? Can I add something to it? Sure, I'll try to keep it short. So even the word ICO is like a hangover from IPO, right?
30:03
And if you look at it, stakeholder is like shareholder, again, a hangover. And we humans in the longer term seem to have this tendency of copying the previous iteration of what it was, and a very good example would be the automobile. So the early prototypes, let's say, of the automobile did not have a steering wheel.
30:22
They had a tiller, which was borrowed from what we had previously as boats. So I feel to some extent we are doing kind of the same thing, trying to borrow what worked in the previous iteration. And on the note, like in the presentation, I mentioned briefly the conversational DAOs, which should, in theory,
30:43
kind of alleviate, in a sense, this disconnect between the stakeholders and the actual foundation or whoever is actually developing the protocol because currently there's a complete disconnect between the two. Once you gave them the money, you sign on that. I agree that you have no rights. This is actually not to share all the stuff that you actually read
31:03
when you find out that you were scammed, but a system that would enable continuous feedback between the participants and the actual developer of the protocols should at least bring that shareholder ability to steer the course of the project, which should alleviate also the problem of this complete disconnect
31:23
between shares and tokens. So I think we are still very early, we are still learning, and this is us breaking eggs to make an omelet. Did you want to make a comment? Can I just say something, one last thing on that? There was a really nice paper written last year,
31:41
Coin Operated Capitalism by a professor at University of Pennsylvania, by the last name of Hoffman. It's an excellent paper. What he does is he took 50 ICOs from last year, he had 100 doctoral students in computer technology in the computing program look at the smart contracts of those ICOs,
32:02
and he found that there was a very small amount of the smart contracts that actually coded what the white paper said the coin did. The thing is that nobody actually reads the smart contracts before they invest. And he basically showed that these smart contracts can be completely modified,
32:20
so the amount of coins can be modified, the amount of committees that make validations can be modified, the whole entire system can be modified, and that's built into the code. So I definitely think that it's something to consider prior to investing, is actually understanding what you're investing in.
32:40
I'm sorry, but we have two more questions. It's interesting, I appreciate that, but we have two questions, and we are, again, over time, dinner is waiting. Is that okay that we skip the discussion here and let the next questions be asked? So I'll ask a question about value, but in a different sense. Many applications and consequences of distributed ledgers in society are controversial,
33:04
and that's beyond the scope of a scientist or an engineer to solve. So do you or did you work with an ethicist? How do you solve these problems? I think that's maybe something for Dexter.
33:21
I do not talk to the administration about blockchain technology at all. No, I did not work with an ethicist. Well, that's not true. So we work with an institutional review board, which is the ethical board to conduct any kind of research.
33:41
In my case, there's very little risk, even if we put your mammograms out in public, it's very hard to identify somebody from a mammogram with no identifying information. So there's very little risk in what I'm doing. Once you get into ultrasounds, the patient's name is on the ultrasound, there's ethical concerns of privacy.
34:01
But one of the reasons I picked digital mammography, one, because I know there's 40 million of them floating around America at any one year, and two, it's completely anonymous. So this can go on IPFS with very little ethical issues.
34:21
What I wonder is, the first thing you think if you think about health data and blockchain is immutable and is stored forever, is that a problem in any case? So any IRB is going to mandate, even for genetics and genomics work, anonymized data is put into the public space and people work in the anonymized space.
34:43
I don't think blockchain is any different. I guess you can, I mean, the way to me it works is, at least for medical data, there is a steward of the data. For instance, I'm a steward of all these mammograms I'm going to collect. If you don't want to be in my study, I'll just stop sharing your,
35:01
you know, I will just stop being a steward for your data. To points that were raised last night at dinner and throughout the day today, if you already have a key to that data in IPFS, if you already have the address, well that's it. Then you have everything, right? Then you have everything, but prospect, in the future, I just don't give out that address if somebody does not want to be in the study anymore.
35:22
And of course, they agree to inform consent. All of this is explained and they have to agree, before I would accept your data. So, you know, in the clean world of academia, I think there are ways to still leverage this technology in spite of the shortcomings.
35:42
You know, it's no different to a regular database, to Mohan's point when he made it before. Short reaction to that? Regarding ethical implications, it makes you wonder, for example, Facebook and these companies that have the ability to influence in mass people
36:00
and they conducted actually experiments trying to see if they can make users either feel more happy or sadder depending on the content displayed on the feed. So there are tons of examples like these that make you question the ethical behavior of that company. And I think the open source and the transparency of blockchain, in some regards, can put in place a system that makes it hard
36:25
for people that want to exploit or do something unethical to be seen by others, similar to how you have the open source and if someone tries to plant a backdoor, you just count on the number of eyes to find that backdoor, right? So in a similar sense, if you have something similar to Facebook,
36:41
where the algorithm that actually decides what appears on your feed is a subscription to an address and you're just pulling the data from the blockchain where it is, you are at least in control of seeing the information flow. And when it comes to if we interacted with ethicists, we constantly have this soul searching into what we are doing.
37:04
Is this going to end up like some Black Mirror thing or is this actually going to be improving humanity? So I think this kind of question and deep questioning is something everyone should do in this field because it's something that potentially could influence
37:23
on scale in the current interconnected setup. To sum it up, open source and transparency, I think, could provide a solution to the lack of ethical standards because people could point to what's going on. Currently, you cannot. It's a black box.
37:42
Currently in medicine, privacy trumps ethics. The way data flows today is not ethical at all. Nobody would agree that we have an ethical system for data management. We have a private one. So last short question and short answer.
38:00
Yes, so actually I find the idea of an open transparent system very appealing but I was just wondering whether the discussion could include perhaps there being a spectrum of central versus decentralization when it comes to use cases. So do you envisage an ecosystem where all of that is possible and each one essentially answers a particular use case?
38:25
So I think that's spot on and in the future, there is this gap between fully decentralized, 10 years further down the road, who knows, maybe earlier, maybe later, and there will be this gap to bridge, which in terms of user experience when you have a completely decentralized system,
38:42
getting that information or that experience up to you just click and it loads, it's hard. So I think there will be an incremental use of just the required and where is the maximum impact, like strong authentication, verifiability, traceability, where these things trust, where these things can be fit in a, let's say, more centralized system
39:04
and gradually as the infrastructure matures and we have public blockchains being able to scale and the cost of transactions lower and so on, maybe in time we can move to a fully decentralized setup. But I think in the near future, the hybrid will be probably the approach that will win.
39:27
I think it depends on the application. I mean, by the nature of medicine, it's a private enterprise. Patients, doctors and developers are really the only people in my network that need to see this data.
39:41
Very different for a social network, very different for a financial network. So it totally depends on the application in sort of a context, I think. But for example, if someone would visit the website to upload that picture, if they would have to spend 10 seconds while the thing is rooted, they would probably end up not using it, right? So we have this example with PGP, which was great,
40:02
but people didn't use it because it was not user friendly. Even today it's hard to set it up. So I think we also have to think on how to make it usable so people actually use it, right? So I think everyone has to gain and to be practical where you use it because it's not there yet to be used in everything, right? Make it easy, I think that's a good point to wrap up with.
40:23
And I think this is what's coming. So thanks a lot. Thanks a lot for staying with us all the time without coffee. Thank you.