BCOS Monero Village - Scaling and Economic Implications of the Adaptive Blocksize in Monero

Video thumbnail (Frame 0) Video thumbnail (Frame 4169) Video thumbnail (Frame 5635) Video thumbnail (Frame 6820) Video thumbnail (Frame 16122) Video thumbnail (Frame 18690) Video thumbnail (Frame 24639) Video thumbnail (Frame 26360) Video thumbnail (Frame 28367) Video thumbnail (Frame 36602) Video thumbnail (Frame 38145) Video thumbnail (Frame 46421) Video thumbnail (Frame 56957) Video thumbnail (Frame 68701) Video thumbnail (Frame 80445) Video thumbnail (Frame 92189)
Video in TIB AV-Portal: BCOS Monero Village - Scaling and Economic Implications of the Adaptive Blocksize in Monero

Formal Metadata

Title
BCOS Monero Village - Scaling and Economic Implications of the Adaptive Blocksize in Monero
Title of Series
Author
License
CC Attribution 3.0 Unported:
You are free to use, adapt and copy, distribute and transmit the work or content in adapted or unchanged form for any legal purpose as long as the work is attributed to the author in the manner specified by the author or licensor.
Identifiers
Publisher
Release Date
2018
Language
English

Content Metadata

Subject Area
Graphics processing unit Scaling (geometry) Block (periodic table) Adaptive behavior Multiplication sign Adaptive behavior Projective plane Interactive television Mereology Data mining Type theory Crash (computing) Arithmetic mean Process (computing) Different (Kate Ryan album) Core dump Block (periodic table) Metropolitan area network God
Ocean current Slide rule Scaling (geometry) Scaling (geometry) Modal logic Data storage device Basis <Mathematik> Digital signal Data storage device Term (mathematics) Mereology Perspective (visual) Band matrix Moore's law Graphical user interface Internetworking Term (mathematics) Band matrix Business model Endliche Modelltheorie Communications protocol Communications protocol
Digital electronics Differential (mechanical device) Multiplication sign View (database) Range (statistics) Set (mathematics) Neuroinformatik Mathematics Bit rate Core dump Pairwise comparison Information security Fiber (mathematics) Mainframe computer Area Simulation Email Electric generator Menu (computing) Database transaction Price index Demoscene Connected space Band matrix Tablet computer Message passing Internetworking Internet service provider Analytic continuation Thermal conductivity Point (geometry) Slide rule Trail Functional (mathematics) Divisor Virtual machine Drop (liquid) Twitter 2 (number) Number Frequency Moore's law Centralizer and normalizer Internetworking Band matrix Operator (mathematics) Divisor Analytic continuation Summierbarkeit Operations research Pairwise comparison Information management Standard deviation Multiplication Scaling (geometry) Information Plastikkarte Line (geometry) Cartesian coordinate system Evolute Tube (container) Perspective (visual) Word Software Connectivity (graph theory) Table (information) Window
Ocean current Point (geometry) Trail Divisor State of matter Multiplication sign 1 (number) Bit rate Limit (category theory) Plastikkarte Database transaction Product (business) Element (mathematics) Data model Mathematics Internetworking Different (Kate Ryan album) Personal digital assistant Business model Divisor Communications protocol Workstation <Musikinstrument> Regulator gene Block (periodic table) Plastikkarte Database transaction Maxima and minima Markup language Limit (category theory) Measurement Perspective (visual) Radical (chemistry) Internetworking Frequency Personal digital assistant Connectivity (graph theory) Right angle Block (periodic table) Marginal distribution Communications protocol Sinc function
Scaling (geometry) Multiplication sign Median Database transaction Mathematics Bit rate Arrow of time Multiplication Physical system Proof theory Curve Theory of relativity Block (periodic table) Sound effect Median Database transaction Maxima and minima Bit Term (mathematics) 10 (number) Derivation (linguistics) Type theory Process (computing) Order (biology) Normal (geometry) Block (periodic table) Resultant Electric current Spacetime Point (geometry) Empennage Functional (mathematics) Divisor Robot Maxima and minima Element (mathematics) Inclusion map Term (mathematics) Well-formed formula Reduction of order Communications protocol output Normal (geometry) Implementation Form (programming) Default (computer science) Scale (map) Default (computer science) Multiplication Scaling (geometry) Key (cryptography) Cryptography Personal digital assistant Function (mathematics)
Logical constant Group action Scaling (geometry) Weight Multiplication sign Function (mathematics) Client (computing) Database transaction Perspective (visual) Graphical user interface Spherical cap Formal verification Arrow of time Circle Smoothing Endliche Modelltheorie Position operator Exception handling Proof theory Block (periodic table) Sound effect Infinity Maxima and minima Database transaction Median Bit Staff (military) Element (mathematics) Data mining Proof theory Process (computing) Internetworking Vector space Coefficient of determination Right angle Quicksort Block (periodic table) Freeware Resultant Point (geometry) Rational number Empennage Addition Adaptive behavior Maxima and minima Valuation (algebra) Limit (category theory) Number Permanent Term (mathematics) Band matrix Communications protocol Linear map Scale (map) Scaling (geometry) Weight Mathematical analysis Planning Total S.A. Limit (category theory) System call Vector potential Similarity (geometry) Personal digital assistant Calculation Formal verification
Scaling (geometry) Price index Information privacy Exponential function Perspective (visual) Velocity Different (Kate Ryan album) Finitary relation Information security Physical system Proof theory Zirkulation <Strömungsmechanik> Block (periodic table) Multitier architecture Sound effect Maxima and minima Database transaction Message passing Arithmetic mean Frequency Angle Chain Order (biology) Nichtlineares Gleichungssystem Quicksort Point (geometry) Empennage Maxima and minima Average Scalability Number Element (mathematics) Frequency Chain Goodness of fit Term (mathematics) Energy level Divisor Nichtlineares Gleichungssystem Communications protocol Metropolitan area network Distribution (mathematics) Scaling (geometry) Weight Chemical equation Physical law Line (geometry) Limit (category theory) Sixty-six (card game) Software Personal digital assistant Function (mathematics) Game theory Zirkulation <Strömungsmechanik> Weight View (database) Multiplication sign Combinational logic Insertion loss Function (mathematics) Parameter (computer programming) Database transaction Total S.A. Nominal number Mathematics Cryptography Bit rate Formal verification Cuboid Endliche Modelltheorie Data conversion Gotcha <Informatik> Service (economics) Wrapper (data mining) Data storage device Price index Term (mathematics) Digital object identifier Band matrix Type theory Proof theory Velocity Block code Block (periodic table) Information security Fundamental theorem of algebra Service (economics) Vapor barrier Adaptive behavior Theory 2 (number) Natural number Ideal (ethics) Angular resolution Execution unit Mathematical analysis Software maintenance Information privacy Number Communications protocol
Axiom of choice Group action Scaling (geometry) Multiplication sign View (database) Workstation <Musikinstrument> Combinational logic 1 (number) Parameter (computer programming) Disk read-and-write head Food energy Dressing (medical) Mathematics Cryptography Graphical user interface Different (Kate Ryan album) Cloning Cuboid Endliche Modelltheorie Information security Proof theory Email Algorithm Block (periodic table) Software developer Sound effect Database transaction Maxima and minima Bit Digital object identifier Degree (graph theory) Band matrix Data mining Proof theory Type theory Befehlsprozessor Moving average Quicksort Block (periodic table) Information security Arithmetic progression Thomas Bayes Point (geometry) Empennage Divisor Adaptive behavior Maxima and minima Drop (liquid) Rule of inference Power (physics) Element (mathematics) Revision control Chain Crash (computing) Term (mathematics) Well-formed formula Data structure Communications protocol Pairwise comparison Dependent and independent variables Scaling (geometry) Information Key (cryptography) Copyright infringement Analytic set Planning Total S.A. Limit (category theory) System call Information privacy General relativity Word Personal digital assistant Codec Musical ensemble Communications protocol Active contour model
Empennage Scaling (geometry) Multiplication sign Maxima and minima Inclined plane Information privacy CAN bus Chain Cryptography Right angle Block (periodic table) Communications protocol Information security Game theory Proof theory
so up next we've got Arctic man he's gonna be talking about scaling and economic implications of the adaptive block size in Manero I just read that from there but basically he's gonna be talking about Meniere's adaptive block size what does that mean what does that mean for the economies of scale all that different type of stuff he's a smart guy this is a Manero core team member you're in the presence of a literal God so I'm just kidding so that's everyone give a give a big snap to Arctic mine well thank you for the excellent interaction first a little of my own background I got involved in Manero in 2014 after being involved with the Bitcoin since 2011 I originally started as an investor and also as a miner when you can mine Bitcoin with GPUs that was the day I saw the crash of just after the crash of in 2011 were becoming one for 32 to 2 that was the race I got involved with bacon for 2012 and 2013 and one of the things that I did a say as is I asked the question how can it fail I asked the question where a particular problem so I spend a lot of time studying the scaling issue and the blocks size debate in Bitcoin it is in this process and only 2014 when I actually came across a post in Bitcoin talk I talked about mineiro did not mention scaling at all but it actually got me interested in Manero and the first thing I did is I spent three weeks analyzing how many are scales and how is it different from Bitcoin I started getting more involved and interested in more narrow learning about Manero and in particular spending a lot of time looking at scaling in Manero in early 2016 I was given the honor of being part of the core team of monaro which I accepted and was originally research that I did primarily as due diligence as an investor has become now part of my contribution to a significant part of my contribution to the Monaro project so what am I talking about is scaling I'm gonna break my talk into two parts two
aspects of scaling now when people talk about scaling in a cryptocurrency these are totally commonly collated so they talk about these two aspects of scaling as one is you have fundamentally different the first part is technological support for scaling and the question there is is can the technology actually support scaling today in the near future the second question is how does the social covenant and the business model and a protocol support for long term Kaling so what is a social covenant basis model and protocol long term support for scaling and now it's fundamentally different from the first question can we transmit the blockchain through the internet for example with current technology so the
question is first question one asks is is it possible with our current and future technology to provide the necessary computing power bandwidth and digital storage needed to support blockchain growth and we're talking over decades so the question is can we really continue to scale and what is going to happen with technology this is a technological side of scale the second question is what are the lessons from history that we can learn when it comes to scaling and I'll present my next slide in my next slide I it's actually quite romantic most people are familiar with
scaling from the point of view of say if you're a decade I'm talking here going to all the way back to the 1850s this by the way is a shot from the cost of a million standardized operations per second so we can think of a standard set of computer operations and then you price it as a function of time and in particular one looks at it this is of course a large scale well looks like the changes and trends that have occurred in basically the last 160 years and these are profound implications of how we do finance to how we do banking to how we keep track of our money and to a lot of technologies we use today in in centralized alleges and it's quite a significant and very way so if you look before about 1940 IU in the Second World War basically it was a flat history we were talking about the early tabulating machines so they came on the 1890s by IBM some ideas in the United Kingdom a British Empire at the time of the Babbage engines various engines that were created this area here is primarily tabulating tabulating machines that were improved by the harvest company with later became IBM and going to the 1930s and again we saw this application of tableting machines with the Second World War of course there was a major demand for computing technology primarily on the side of the Allies in the sense of how do we crack the german enigma and this there led to Fermanagh research in the United Kingdom they used telephone relays large numbers of 12 on relays and build some of the most primitive computer see on the computers we then get into things such as vacuum tube copiers in this range here then you move to transistors and then we start looking at integrated circuits all the way the core into integrated circuits Moore's law it's primarily this range down here and he of course we're talking about huge big centralized computers we think of mainframes that would be owned by major governments or major corporations it was at this point in time banking was primarily and is not done through centralized Ledger's of any kind it was no ledger it was simply too expensive to keep track of transactions if you have to manually enter every transaction if you didn't have to transmit them through telegraph lines where it could cost in today's dollars 50 even 100 dollars a would think of an email where you spend 100 a word or 50 a word to send it or a text message this is prohibitively expensive so we were dealing with a cash society and with a bearer instrument society securities for example were gilded pieces of paper which were then stored and and whoever hold the instrument owned that security owned that funds so it's kind of the history and 1950 saw scene the first credit cards come out and then we started seeing the evolution of the centralized Ledger's starting primarily with the strength of the mainframe computer in the 1970s and 1980s till today credit cards for example or so really conceived in the 1950s but really then it start to take couple about the 1970s debit is more technology of the 1990s so this is kind of important to sort of understand the history what's significant is that between 1940 in 2000 which is the shot here we see a factor of 10 to the 13 in the drop of price of computing and it won't stop here for a minute and a sixth question this is 10 to the 13 is a 10 trillion and by comparison need to ask the question if we move from a decentralized layer to a decentralized hfx mokona Mineiro and let's say you have a hundred thousand notes what you're actually are doing is you're making a hundred thousand copies of that information and then synchronizing them all so your main poster cost of 10 to the 5 I mean even ten to the six but it's still when you look at the history it becomes extremely affordable and the indication would be that in history of of human of the world the we went from bearings from is to be a short period of time of centralized ledgers and then we are actually moving to decentralized lectures so my next slide I'm gonna show a simulation more recent but we're talking about again the bandwidth and this is bandwidth at high in the United States but it's got indicator can from 1983 I'm talking about a log scale to 2018 and we see about an increase in Internet connectivity 1.5 million times in in the amount of bandwidth even more interestingly in the last eight years alone we see an increase in conductivity of approximately twenty six times and we don't think about this only we get an upgrade on our internet service and it's actually more pronounced if we look at developing countries where they may actually jump two or three technologies when they put in fiber and they don't even bother with things such as copper so what we're seeing is trends in technology that show this trend of the price dropping and what is very expensive in the past becomes incredibly affordable so it come a couple of questions is it possible
well given time yes it is and we all need a continuation of these trends even at a slower rate and then we were we would see decentralized ledge essentially being competitive in the future with today's cost of a centralized lecture so then money aerobic or any kind of centralized decentralized taking place the cost differential is roughly about 20 to 30 years so the cost of say today having a decentralized ledger could be comparable the cost of centralized so I just say a generation ago that's a rough indication there's a whole bunch of things in the pipeline 3d ships if you have nan if you have in your phones and your tablets on underneath and steel so those are three - 3d ships the SSDs that means instead of laying one layer you lay multiple layers on top of each other carbon nanotubes is being considered as a potential semiconductor 5g that's coming down the pipeline on the we're talking about gigabit lens even within copper we're seeing doses three point X which is going to be capable kick up it over even existing cable networks fiber-to-the-home I'm knowing in both in in Vancouver they're laying the fiber to the home right now sure initially they're gonna say 150 make up a symmetrical but that has capabilities in the gigabits so we're looking at a growth in technology and then we need to ask the question so it is possible let's see what we have learned from history well do a couple of
cases and the first case very interesting is what we can learn from the credit card industry now that credit card industry will love to criticize the banks but sometimes we're gonna learn from them we can learn from their mistakes in 1950 the diners club was introduced for 7% merchant fee now why would you create and design a payment called the 7% merchant fee well because it costs a lot of money the technology of the day to keep track of the transactions so you develop a business model around that and the business model basically is you say ok Europe you're a merchant we're gonna provide this service and we're going to incentivize the sale we're gonna make it attractive for a customer to come to the establishment I should give us 7% but we'll get yourself and that makes a lot since if the merchant has maybe a 50% margin on the product it does not make sense for low margin measures but here's the interesting thing I'm a Canadian I live in Canada and I would have purchased gas with the Canadian credit card in the United States my cost would be roughly 7% today why because I would pay 3% of the gas station every gas station basically charges a three percent surcharge for using a credit card and in an additional four percent on the spread between the Canadian US dollars then it's just charged by the bank and the credit card companies so my effective cost in 2018 is the same as a woman in the effective cost back for Dinah scrap in 1950 so listen here is really significant it's the cost of technology can fall by a factor of 10 2013 10 to the 14 something that can arrange but the business model that was developed in 1950 based on the technologies of the 1950s and the cost associated with those technologies is still in use today and it's starting to fail another example right in this in this community are you take a taxi for a short ride so 3 dollar surcharge were using are critical well I would say about 38% if you have a 7 or 8 dollars actually right maybe you spend 100 or 3% but it's still quite high and does not reflect in any way the cost of the technology my second example is from the cryptocurrency and we're talking about Bitcoin now Bitcoin initially when it was released in 2009 and conceived in 2008 did not have a minimum one a maximum one megabyte block size limit this was added in 2010 approximately 8 years ago shortly thereafter there was supposed to be a con talk with a proposal to actually get rid of so we could scale like paper that was the birth of the Bitcoin 1 megabyte block size debate 8 years ago the debate is still going on today to this day but in the meantime technology's not lying still the cost of transmitting those blocks over the Internet and the speed is a factor of 26 anybody who's familiar with the current Manero block sizes it's basically the difference between the current Monera block size and a typical mono transaction sorry in a typical Bitcoin transaction so we're talking about the existing four are not bulletproof so the this twenty six percent decline in in the cost it's not factored into the debate at all we're still stuck in it in the existing business model I'm gonna basically say that we are going to keep debating this of course the technology does not change it doesn't wait so what is really a solution
well the real issue in both of these cases is that the protocol the business model the social covenant does not support radical changes in technology and and this is the key element the in the case of the credit card state the the technology the protocol was designed for the technology of the 1950s it becomes totally irrelevant today and they have having bigger and bigger problems litigation regulation in the European Union and of course these fees in the case of Bitcoin again the protocol didn't support the scaling technology and we still start in the old protocol changes so the question is what is needed and so we need to have a social covenant that actually they can take advantage of future improvements in technology even those future improves that technology are as dramatic as the ones in the past and at this point in time is when we take a look at manera so
what makes Manero support the scaling so what we have a more narrow is the crypto node penalty function now if we look at Berry's Inc nations of it essentially what it says is you have a penalty this is the block reward so everything is proportional to a block reward this would be the median transaction block size over the last 100 blocks or 300 thousand bytes whichever is greater and this is the proposed block size that you can of mind and essentially you put up impose a penalty on the minor that then has to be made up with by transaction fee and this is basic definitions the couple key elements of others are interesting the first thing is that the transaction block size if there's a maximum of two of twice so you can only make it twice as big and it's only apply when the proposed block size is greater than this effective medium so those are the key elements in it in the penalty the basics now in order to really understand this better I'm going to redefine the cripton lock penalty and in order it introduced
this term called B which is basically the percentage increase in the block size so if you have a 300,000 bot sizing a 30,000 3,000 by block size and you add thirty thousand bytes that's a 10% increase because in terms of this we can then understand kind of how the whole system behaves so basically the penalty can then be reduced to the block reward times this percentage increase squared this is the basic reduced Krypton oden penalty form there's a really interesting point about this if B is less than zero ie if you don't there's no refund now anyone care to guess why this is a burning hormone arrow and he stabs at this why not refunding the penalty yes that's true but that should know something else it actually creates a burning on one arrow and the reason is because it delays the tail emission so at any point in time in the future there's going to be a smaller amount of monaro in the economy was the penalty supplied and so as a result a side note about that okay so then we simply have the new reward this is a very interesting if you look at the rate of change of the penalty with how far the curve you are I mean how big of a scale you do then you get this to our base times B so depending on the point of B is essentially linear and that's what you would expect for a quadratic penalty so now here's the next question what happens if we add a transaction T with the size of empty so now we're a point in the penalty formula I'm going to add an additional transaction and this is the question the miner faces because the - to decide does this transaction generate the fee to justify its inclusion in the block so then you basically add the term you take out the the to be the B squared term because it's the a16 penalty in your left for these two terms depending on the relative size of b mb t bt being the additional space of the transaction you have two cases if B's much greater than VT you can basically eliminate the B squared term and then you get the essentially the infinitesimal transaction model or in the case where B is significant then it becomes dominant and you get a case where where we have a linear penalty we can then calculate the fee per byte in bytes and again our base is calculi dividing by the effective medium block size now here's the key element as that block size goes down the price of a legitimate transaction of Monera false in terms of Manero sorry as it increases so basically if you go from say 300 thousand bytes to three million bytes yep your fee per byte has fallen by a factor of ten at the same time the spammer wanting to spam a certain percentage has to pay the same amount of money in terms of Manero so that's a very interesting element so the ham or the legitimate transaction falls in price because of transaction size stays constant so there's more transactions sharing the same amount of penalty so this allows us to then define two types of transactions so we need to find a reference transaction where a reference transaction is is a typical more narrow transaction that is applied at the penalty of the minimal 300,000 by tens level so what we're saying is we've got a three hundred thousand bike bot block size we're gonna add a typical transaction and from that we can calculate what the fee has to be and basically that is the default fee per byte that you pay money because the idea is we pay enough of a fee in that transaction so that transaction alone can support that scaling of itself otherwise the miner will mine in the Brock's ashwin scale that's how the default fee is set and an avoids this issue of it being stuck like what can happen is nobody will create it and so you have a massive growth in the TX pool but no it's actually using the the scaling effects not paying enough of a fee that's how the defaults feel Manero is set it is determined by the ratio of a typical transaction to the minimum effective process this minimum amount of 300 thousand bytes the loan - a simple multiple of the normal fee so that that's where penalty effectively determines the fee per byte and effectively determines the fee that you pay
at the other end we have the opposite situation in the opposite situation what we have is we have a transaction infinitesimal size will be pushing the penalty to the maximum so right at the maximum in this case basically your fee per bite is twice the block reward divided by again the effective they have the median effective block size and in this scenario the minor would actually make four times the block reward in fees if someone were to push the penalty to the maximum now in all this analysis were assuming a rational miner that mines to the best economic interest and therefore all those transactions in the block for our own perspective so basically what the miner does is she would include in the block transactions the highest paying first until you get the lowest paying transaction that actually pays the penalty so it's always the lowest paying transaction that pays the penalty so this is kind of the the basic ulema hafiz' set okay but here's
what gets interesting what happens if the base reward goes to zero hey I don't want any cash but I'm still fees the fees go to zero so if there's no base reward there are no fees so it is impossible in this model to replace the base reward with a fee market you can have a fee market but a free market for something entirely different if your base or one is zero you have no fees and you have an insecure coin this is precisely a situation with corruption or reference by car nominally in my narrow is a classic example you say okay we're going to stop the tail emission at some point in the future okay how do you how do you actually incentivize the - you don't have any incentive because basically there's no penalty they just increase the bra the block size ad infinitum and you get no money for fees so your client becomes insecure same thing with by coin by coin right now is below Mon arrow style emission in fees so is it really a problem the concept of a fee market I was proposed by Satoshi it's in fact putting is a serious question in this case I'm going to cover another example here and this is the issue of often blocks and peter recent paper this has been presented by some in some circles as a replacement or a fee market yes it's a fee market but it's a fee market addressing the problem of often blocks created by pushing the technology too soon it's a it's a penalty staff also present in monroe and again if it's checked with one looks a reference the result is proportional to the block reward in the absence of the broad road the recent penalty files it is a very valuable research very important research but it does not provide an answer to replacing the block reward with fees and this is actually a very sort of difficult question one of the conclusions are one can make is that if you have another cryptocurrency and they impose a penalty in terms of the block reward that is weaker than that in one arrow then it's going to run in the same problem so one can understand for example the position taken by bitcoin call they said well we have we may have a problem here we can't take the risk we won't allow the block size to increase effectively imposing an infinite penalty and that's a reasonable course of action it's better to have a small coin than insecure coin the other interesting thing is this issue is present in all major proof-of-work currencies with very few exceptions and exceptions are Manero I am dodge coin not familiar with that does have a permanent emission doesn't have an adaptive process limit and the really old but not well known fry coin now Frank coin does have a maximum number of coins what they've done is it been created emerge and basically in that situation the demurrage generates the block reward so the coin can stay secure yes it's worth about three hundred thousand in total market cap sitting right at the bottom my market cap but it does address this question in a different way so those are the exceptions I think has it's a if it has a yes I think are some there were folk from one arrow that keep the Taylor mission so there will be fine yeah so if they've got a Taylor mission that they would work if for example they have demerged they were also girls with made to work but if you have a fixed amount of coins and you don't have to merge I have some real constants and that would be my my answer to that including some very high valuation cryptocurrencies okay so the next question here and this is another a bit of an aside is money honest of course planning to put both brookson now bullet pros have very interesting characteristic and that is that it has a verification time that scales with a total number of outputs and and by up was I'm also including the padded outputs that will be put in and then these the size of the bulletproof scales as the log base 2 of the number of outputs well becomes fairly obvious then if you allow this thing to if you allow very large bullet proofs with large numbers of outputs then what happens is one has a potential attack vector which was quite a bit of a concern and so we have to deal with that attack vector and how the solution to deal with it is to replace the block size and all those penalty calculations are short before with a block weight and the block weight takes into account the verification time and this is was an idea that was actually initially proposed by Andy Toshi and smooth also discuss it I cannot refined it and we came up with the following and basically
what we do is you look at the bullet proof and there is you take if it's got one or two outputs you leave it as is and if you've got more than two outputs you say okay if I had a bullet-proof with two outputs and then multiply that by the number of outputs I get a size I take the difference between that size and the size that you actually get and I claw it back by clawing back I'm saying you're not really counting on it I won't gonna give you about 20% so this creates a term which is linear in K I can show you which is B P base times time scare down here and that effectively deals with this attack back to the other element of it of course is that the maximum number of outputs in a bullet-proof is limited to 16 so you can't really prevent any kind of attack from this angle so this is kind of a tweak but it's kind of shows the kind of things that it is I mean the purpose of fees in a cryptocurrency is to deter spam attacks this is where Satoshi put the one megabyte limit in pick on the first place and in this case again we have to be careful if we're going to change the privacy Rinku financial transaction technique we are faced with having to tweak the whole weight system to give us a weight that actually takes into account the verification cost so in future we will hear about block weights rather than block size that really only makes a difference we are more than to the outputs in a transaction but that's the reason why we will see that I'm going to now briefly talk about something else and this is called the equation of exchange in economics now the equation of exchange in economics it's actually a sort of a tautology and basically what you have is you have the money supply which in a cryptocurrency is actually set by protocol V is the velocity no people familiar with the velocity of money is it essentially is how many how fast the same piece of money circulates so you get say in the u.s. dollar example you receive say sorry and then say ninety percent is spent within the nine ninety day period that then how long before that same dollar comes back to you that is actually the velocity of money B is the actual price of goods and services in - in terms of the currency and that is can be thought of as if P is high that means you have a lot of inflation B is low we have deflation in this model here Q has been shown in a simplified manner and and by that I mean that typically in an economy you have a very large distribution of goods and services so if we have for example cups of coffee a million dollar watches in the same distribution you can have different kills and so you look at the total distribution of the goods and services so here's the really interesting question that comes in the first interesting point that comes out of this is in a proof-of-work cryptocurrency if we assume the economy doesn't change we have the same types of goods and services but we simply make it bigger thank you is actually proportional to the block we decide so this is basically saying I have some real implications in the case of Manero then if we double the transactional activity we have to double effectively the price of the currency because M doesn't change is set by protocol V hasn't changed because you haven't changed the nature of the economy is making economy larger and in fact that is actually from a theoretical point of view if you're looking for the perspective sell speculate or an investor you ask the question okay I want to buy this because I expect this the future increase in ink you in the future and therefore a decrease in P the other implication is if we come back to the actual how fees are calculated who notice that fees are now calculated with the vining by the block size well that has a real implication because now what happens is if the economy doesn't change fees actually should remain constant once you get tail emission in terms of actual purchasing power because if you double the size of the economy thereby doubling the price the feast in terms of my narrow fall by half so yeah we have a very elegant excuse me solution where the fees are more narrow will stay constant in real terms now there's a lot of KBS for this the first game yet of course is and this asset has recently happened in Bitcoin it's what happens if you change the velocity so for example you decide we're no longer a transactional currency we're now gonna become a store wealth and people are not going to move these things around and therefore your velocity goes drastically down that caused the speed will go down that's one possibility the other possibility is you change the nature of the goods and services so for example instead of buying cups of coffee would buy million-dollar watches well then in our example then of course you get a similar situation but it is an indicator the Monaro really is very much in in harmony with basic laws of economics and how the fees have said because essentially we're saying in more narrow we have a block size economy etc and sfe that basically come is to set it and forget it approach so if in the sense of our feast work and it isn't very much in harmony with the score the equation of exchange in economics now most people are familiar
with Manero I see fungibility privacy in anonymity coin and I spent virtually no time on this subject I guess one of the the messages is that there's a lot more to more narrow than fungibility privacy and anonymity if I'm Manero has the most successful support at the social covenant level for scaling of any major cryptocurrencies and I do not say that lightly largely as it's due to the combination of the adaptive block size limit the krepner policy with the minimum tail emission which gives you the security and the minimum 10 emission is absolutely critical to maintain proof of work security of course chains one is case the money are fundamentally different for the proof of work or occurrences in Bitcoin well again what happens in Bitcoin is that people compete against each other for fees and in effects block size and no matter what the fees go up you can't increase the days of competition well I am sure Manero is you're actually competing against the penalty so we're pushing against this penalty which is actually what you're you're competing against the next question and I really I know this is a difficult question for a lot of people especially if you're holding a lot of currencies that have this issue they're really fundamental unresolved questions with respect to the viability fee market as a means of replacing fees and if there's a message I think some quite real hard questions need to be asked about this it's a phrase in the original become white paper unlike the rest of the work which literally is a work of genius this phrase is provide with no justification whatsoever and nobody afterwards has tried to provide any kind of justification there is however a significant amount of literature already and questions been asked about its viability either in the president reflected fakes box size so that is a question that in my opinion needs to be really looked at much more seriously unfortunately virtually every other major proof-of-work wrapper currency has copied Bitcoin in the sense and I'm talking about like going Bitcoin cash I think such as C cash they've all done the same thing and - in fact that case you have support not just a - and also the masternodes and the community and the community about so then this is a question it's pushing Manero quite aside from privacy fungibility and anonymity hasn't rolled or most of the competition it is non a well-known aspect of Manero it is one that I hold very dear to my own heart because that's what I work on but it's it enables the kind of ideals that are being discussed here like we want to help people in the developing world we want to send money to poor countries in small amounts because that currency is plunging and value this is the example of thing of Venezuela the hood well you know things crashing so fast you want to send like really small amounts so the thing actually become whole the money in that currency so these kind of things so this type of social impact on a cryptocurrency can only really happen if you have scalability the next question what about second tier solutions lining Network great technology but a lightning network technology depends on being able to scale the underlying chain you have to close the shelves and if the other lines chain cannot support closing the channels you cannot have an effective Lightning Network if I might understand these or what is being done by globally to actually put this kind of technology on the mother of rock chain well so that's a real risk of a lightning Network so again the real questions there because if you can't close the channels if you read align them with paper they're talking about a hundred and sixty-six mega byte blocks in Bitcoin tonight as a minimum so I illustrates the problem this is great technology it doesn't solve the problem just as the example of reasons work great work excellent paper but it does not address the issue of this creating this FEMA it's a tough thing to say I personally do not hold any Bitcoin that's why I will say that publicly and my other main reasons is because of this but I trick just munch in this this needs to be a conversation on this subject given the amount of money involved at this point in time I will know thank you and if there's any questions for me okay basically what I mentioned about offering race and this is basically the work of Peter recents papers I recommend you read that because that's what the analysis was done but essentially it is this if you increase often rate the minor will now see a law or because they kind of mind all these blocks are gonna win so effectively in order to increase the block size or the block weight the minor has to charge more money to compensate for the loss of the often rates that essentially the argument in the recent paper basically it's an exponential function so it's a fairly stiff penalty but I can proportional to the block code that actually takes in so it would basically be a check and balance I guess against the currency like Manero for example getting ahead of the technological all other technology so puts a barrier in there it's a natural barrier it's very hard for a blockchain to actually measure this type of problem and that's one of the few ways it can like going from say two minutes to say four minutes this be some discussion of that that one point is one minute which is still low the yes way too low and in fact is probably the most extreme example that I am familiar with this very room and they're about 15 seconds well the problem with that is that you're you're basically hitting the speed of light it's not as much a scalability issue as a latency issue ladies he doesn't change so if we're sending a telegraph single 1850 across the atlantic you're basically the same latency on that single the switching is really much slower but then you have today not the ladies in the bandwidth okay there big difference if you increase the man on bandwidth then that is essentially the issue of often brought do you improve that by increasing the the the time it's possible I mean it's an interesting discussion but I think that there may be an issue there is latency as opposed to bandwidth and in
that particular because they're very serious problem if you start going below a minute any further questions well I mean I mean it comes back to the fundamental question of how fees are calculated and more narrow if you if you go back to if you look to the fee formula the total fees per block are proportional block reward so the question then becomes as this do you you have two choices you either create a really stiff penalty which is what bitcoin has done that may work but what happens in that point the the the shame can scale so you basically have limited the growth of the blockchain peri the growth of the economy no matter what happens to bandwidth the CPU power to any other stuff you can have another ten to the thirteen drop in these costs you still have the same problem well it is being misrepresented because yes yes I with respect I'm Peter essence paper has been presented as a free market that's how it's presented but the conclusions are not a fee market to replace the Satoshi model it is a fee market to address open walks in the presence of a block reward that is the conclusions that it does so yes it's a fee market that deals with something else subpoena rustles paper does not solve this at oshimen at all it is just as present in money as it isn't even in Bitcoin you pick on casual - or a light coin that doesn't change it serves a very important problem than it dresses it but not the one that we're trying to solve but yeah I mean it comes down to a feast going to zero because there's no competition and people might say well what about if I create a protocol of minimum fees well it's gonna be out of band payments and so that will basically cause it to collapse anyway any further questions yes full yes absolutely because basically what happens if you're a minor you're putting all this energy into bit in there into kinda mining or some box so you're not getting the rocks but it but there's still a cost to mind the fail of the failed all from rock and that's the cost of the miner has to seek compensation for and presumably in terms of fees question sorry okay that's the latency issue I mean that that's an example of a strong latency issue it's not about whether issue it is a latency issue because essentially what you have and it's no different in many respects all it's bigger than me mining a block in Prince George British Columbia that was my on top of say mine that the ricotta made in Western Cape in South Africa if you increase your latency then then obviously either you really increase the block type which is or you're gonna have to split the chain because of the time it takes light to get from Earth to Mars and back that's the speed of light problem and that's fundamentally your latency bomb and again I would argue in a scenario that bandwidth can be increased but not latency unless someone finds the way around special in general relativity sorry any further questions interestingly a combination say yes and no hey sorry bunny okay so if we actually what happened with Manero is there was secret Asics there was a fork was basically obsoleted the Asics and it was done by tweaking the proof-of-work enough so the ASIC couldn't actually efficiently mine it that led to the creation of a coin call ex-mo xmc which has gone nowhere but if a lot of money has be thrown at it I will say this this limited basic resistance and the combination of the response from the community this is the the fork receiver very high degree of support combined with development work in progress and a perfect example is what Howard was talking about combined with the basic principle that the threat is more effective in its execution if you're gonna try and pull that off and you wonder what this community is going to focus and leave us high and dry that's a deterrent effect so there's a whole bunch of factors that create a degree of ASIC resistance in Manero over but strictly speaking I mean it's been demonstrated that at this point in time if you don't dynamically change it or change the proof of work to address it there isn't that strong nation resistance is small the behavior of the community the responsible community and the threat of being left high and dry and these tweaks so that's why the ASIC resistors ammonia is coming from right now sure absolutely I mean but but it's what matters the question that I ask if he doesn't really matter today when you send an email that you might use decide to put that information it's into a Word document or an OpenOffice document and attach that multiplying the size of the email it's not because the cost is in there and I suspect that in the future that's what's going to happen the difference in cost it's not going to be significant if you're paying 100 a word to send an email you're gonna change your prose and in fact if one goes online there's such a thing called telegram style people wrote prose in such a way as to minimize these costs so I would argue that yes there could be a disadvantage but in the case of pretty well every coin the only calling it could potentially to compete with manera at this point would be that dodge coin because big coin doesn't allow this thing to scale anyway so we already head of it and it seems to all the major ones and that relies on implemented an adaptive block size limit could you create a more narrow clone with our privacy and make the transactions a fifth of the size yeah it's gonna be a huge competition I'm not so sure
well it shouldn't be on camera it's going to happen is that we're going to have bulletproof scum in that's going to drop the price in the very short term by a factor of five then on top of that you have the fact that the transactions are smaller so in a very short term you see a drop in fees were a factor of 25 so that's the answer of the short to medium term I'm more of a long term plan a person so I tend to focus I really believe you take care of the long term the short medium to take care of itself you bear patiently waiting personally I have Larkin sense of proof of stake and the reason I do is because even without looking at things like the not a snake problem and all those sort of traditional things is a very basic attack on proof of stake and what that is is you borrow the thing poof mistake is based on the concept that the owner of the coin is also the beneficial owner or the coin well that's not the case the minute that you have a third party that is holding the coins in trust of somebody else and maybe spend some of them and we've had great examples in Mongkok even early in bitcoin with pirate at 40 look him up in bay well what happens in these situations is the person who actually controls the coin therefore controls the stake may have an interest in crashing the coin roll others supporting it because they are not the beneficial owner they're essentially a creditor where they all know that currency so I I really have a real concern objectively and I think this lessons from fairness a lot of the problems that were caused in the 2008 crash were driven by the fact that you have people lending money there was known they were setting the the lender and the borrower up to fail and then they walk away with some fee it's the same problem so fundamentally I really don't believe that proof take a mistake actually itself we are even getting into the technical details so are we my sort of feeling about it I don't see it [Music] yes okay my argument is that the fundamental problem with a social covenant in Bitcoin is that it simply is fundament it's the fundamental flaw and the scaling algorithm so you have a flaw in a critical element of Bitcoin that leads to the conflict and then leads to the breakdown that that you're mentioning and it's very real the the key the biggest strength of Manero has is that we have a social covenant that can support long term technological change yes you have the social issues but if you don't have a fundamental flaw to deal with which I believe is the case in Bitcoin then I don't think the comparison is really their apt because basically the reason bitcoin is having this breakdown is because you have two very legitimate points of view one says it's a security issue we could not increase the block size the other says what about the businesses what about the second latias I mean you have a company that probably they're in in mineiro they're gonna build a structure on top of Manero when you try to do that in Bitcoin so you have a group of two legitimate points of view and in that situation it's very easy to trash the other side then they come with a constructive argument because you simply prove the other side is wrong and that's that's essentially a stage for the conflict absolutely absolutely by me you definitely mitigate the breakdown risk absolutely are there any more questions for we are there any more questions to comment any more questions comment absolutely excellent comment that's all I can say I mean it's essentially that 166 megabyte bar to support it so it's the same issue yeah absolutely one more question well in the United States that question was resolved actually a large degree by a ruling from Vinson in 2013 and it identified what is called a decentralized and centralized cryptocurrency interestingly enough monarus one of the lowest risk ones because we don't have a pre mine we don't have an IC o---- we don't have a post mine we don't have a funders reward all of which sets up the station in this country I think we're a very strong grounds there's some some concerns of moves in the in the European Union I think primarily they're too fixated on blockchain analytics so that could be a concern in that respect so asking are my myself feeling on that one I think the United States is probably one of the safest places for for for cryptocurrency how we doing for time well made
well I would argue that in this
particular example I would actually argue that the United States is gonna be a bastion on the war on cash it's reporting cash and all you have to do is take a taxi right in this city and ask the question how many people are gonna pay 3 on an innate policy right because
basically that is a fundamental battle in the world cash I think we're running out of time [Applause]
Feedback