Measuring Decentralization of Blockchain Networks by Mally Anderson, Everett Muzzy (Devcon 5)

[Applause] everyone my name is everybody I'm a researcher and a product market consensus builder being here I'm Ally a person of my age so over the last couple months we've been trying to figure out ways to measure and define the decentralization generally in the most popular comparative metric between for the tallest is Rupa most of us know that work decentralized in spirit watching as the lower the throughput at least at this stage of the text evolution so transactions percent in there not a very effective measurement so what we actually talk about when we talk about decentralization how can we objectively measure its extent and to monitor its evolution over time obviously these undulation of decentralization is that the binary is or is the condition but a very complex and emergent process that will change this never grace what David could me man objectively originally we set out to define a cushion or some standard measurement that replaced throughput as compared to data plan club we quickly realized that we get into apples oranges pretty fast when we try to compare most of these metrics across protocols so we never focus a bit to say what can we effectively measure on tea room right now and watch Jam over time what's actually happened on main net and what does it tell us about the progress we're making or not making so here the questions we set up answer this etherium actually getting where these part of their metrics so the net worth getting more centralized does the data reveal areas we need to focus on addressing our changing give that some of the trends were observing can we make any meaningful predictions about the future and which of these metrics can be very possible so our approach to measuring the entity civilization over time began with determining which elements of appearance architecture both on and off chain most significantly impacted space optimization so let's run off with at this stage of the research we did by nineteen key subsystems spread across four different categories to investigate we attempted to anchor our conclusions in as much launching and objective data as possible which can get tricky and it's also important to note that there are hard data points that we have not covered that we wish we could because there's an important but not necessarily on chain and there they can be very difficult to quantify those include concepts like the strength and distribution of power grids on which notes run and legal jurisdictions and relative stability of countries on which a large thumbs and notes are posted so whereas many of these data points as possible we track the revolution the corner reporter as far back as possible essentially from the earliest days of theory through gradual adoption rampant speculation hacks cryptokeys bubble and sort of the subsequent course correction see the path yourself so we magically charted about 20 or so different metrics obviously no time to run through all those dead so we chose a few of the spread across these different categories that we're gonna talk about a bit so looking an account growth we looked at start off looking at total come from over time compared to active count growth over time so the x-axis is quarter a reporter across the past four years will be on most of the charts that we see and then the blue line is to either grow over time and the red line is active addresses which we define as a versus a transact at least once and that spike in q2 is the Shanghai PAC that come to so we see a conference increasing steadily over time but we see active addresses more or less starting to flatten after the bubble so as it wasn't used to know what what this is the growing Delta between the two lines over time which likely is due to the increase in smart monarchic to be on the network rather than just simple p2p value transfers here's another account growth graph this one shows account growth over time by total addresses and gray versus the dresses with a nonzero peak balance and orange so what we see is a fairly steady and not server addresses on the network with no major bumps even during the dramatic price fluctuations we can't necessarily say this music spent a steady increase in the number of individuals who leave because the addresses are sooo not synonymous but what we should consider again is the bring bells between the new addresses and the new non-zero addresses that is getting suddenly wider that's worse than our contracts emerge smart contract addresses have to know any balance we're seeing a pretty clear increase in both nonzero addresses bands from our contract addresses over time the aleutian could indicate a network is supporting more diverse investment or decentralized Texas business logic [Applause] this following graph illustrates each ownership of the top ten 101,000 addresses compared to total supply of your time so the bottom in red the top ten and then yellow our top 11 200 and then top 101 to a thousand and all the way up to hundred percent so the story that is tells it's fairly obvious we see the top 10 and 100 addresses on the 8th your network don't any steadily lower percentage of the total of holding that could just be like you do the as a passive result of increase in supply over time we do see some of these larger accounts possibly be pushed into down into lower tiers of ownership while others took their place which also can account for the recent uptick we see and the percentage of each of a 1000 addresses in the past few port when we're looking at the trend of ownership since 2015 we see that the ownership is becoming more diverse across practices we can't assume that more new addresses means more new unique individuals participating the network however we can see that the number of nonzero addresses are increasing and the concentration of the top 10 and 100 are decreasing alongside each other over time this could possibly suggest that contrary to popular area bicker the bubble was not overwhelmingly followed by Wales and holders buying back crypto at those lows just to make a buck what would be the eventual market uptick rather this negative correlation could suggest that more and more new people began accumulating at a steady rate after the bubble which alongside grown in circulation could have been reducing the percentage of concentration of these top 10 and 100 well holders so the ecosystem being as young as it is the unequal concentration of wealth isn't necessarily a huge red flag for decentralization in the long term even with the still large concentration of wealth by Wales etherium is still pretty far ahead compared to other watching protocols at least on this metric looking ahead of our birth concentration in the hands of a few becomes more of a concern when we shift to proof of state and influence on the network becomes more closely correlated with inertia as the beacon chain gets emotional and SPOs replaces proof of work it will be important to watch out for the speaking power concentrating in the hands of you so this graph shows the total circulating vitamin e is compared to the circulating volume of select PRC 20 tokens which are shown as to evaluate an equal so this green line shows the total amount of circulating ie moving between addresses for a reporter it's essentially correlated with the price of beef with the spike and circulating beef aligning with the price high and late 2017 in her name to maintain and then the bar chart shows the value of a select few eat RC 20 tokens circulating quarter reporter we looked at the top 10 by market cap plus a few interesting months we want to look at like diox Matic Ansley the purpose of this graph is to see effectivity Americas they were diverse from both the utility and the speculation perspective what we think it shows that displayed a relatively stagnant Heath price recently the diet and circulating tokens is increasing dramatically and not only is the circulating value increasing of the diversity and market share of tokens are increasing to suggesting that users are using more AR c22 events and these two graphs one the top one log shows the growing concentration of mind polls over time as measured by percentage of total block production on the top and the percentage of the total minors already paid out on the bottom and in each of these graphs each color corresponds to the same mining pool so for instance the green that's on the bottom of each rep it is either mine and so on the rest so over time we see that four pools have really started to dominated by a pool landscape those are either my knife to pool our pool and animal collectively they edged up pasts previous competitors like buying a pool tub and and indoor pool which are can see their sincerity see their their importance in that's three quarters so those four rules or maybe pools now account for over 72% of quarterly block production again the top graph and in a up to about 82 percent of the minors across mining pools McGinnis that's the bottom graph so in particular received complete concern but possibly concerned dominance and block reduction between ether line and start pool which today account for just under 50 percent of lots produced reporter and together we see keep in mind an annual payout of nearly seventy percent of the miners on chain so the concentration of influence among a few mining pools is definitely not ideal but it's not necessarily a major concern by itself miners are people agnostic they will migrate to whatever the pool offers the best incentives and if we assume rational behavior by miners as a single order to reach a half day close to 50 percent or start a basically with other rules to mount a 51% attack it's not outside of the possibility to presume that minors would abandon those fools to protect their income and switch to other rules however when we look at the number of buying goals of - over time as shown by this graph with mining pools in the orange and miners in in red we see the steep decline and both over the past year in short what this means that we had fewer minors that are active on fewer mining pools and fewer mining pools being responsible for network maintenance that's a quick note it's important to say that the number of - red is not entirely accurate we based the number of miners my pool space off on chain mail addresses but some but my impulse pay out there might have off changes through more traditional bank deposits so these numbers will be that although we can't tell those off to numbers so you know overall mining pools could be an area of increasing centralization on the theory we have lower each prices reduce block rewards and a fairly stagnant - great that could be the reason / - not joining the mining pools recently and also we just have a loss of efficiency concentrating the power of mining pools in fewer and fewer mining pools so with an eye towards the future the eventual switch to groov mistake will redefine the symmetric of centralization but until then it's fairly prudent that we keep an eye how this concentration of mining pools and miners in the relative network evolved over time so next one look at at no diversity so this data is from no tracker on your stance and shows that they'll account by country starting 2018 which is unfortunately only as far back as the data goes it's event so the warm colors are the highest that counts and the cooler color the lower note counts and in general we definitely see some an unfortunate bear areas that stay pretty consistent over time such as the African continent but we do see pretty uniform fluctuations rather than random seven spikes and in different geographical areas that are on the Wolfram sort of other countries and it's also helpful to see that the network is physically quite constricted into centralized across the globe and across a variety of legal and political systems so these two graphs look at the known science vs. pricing there's been a lot of fluctuation and the total council notes running if you're in network right now for example I think the number is about half of what this time last year so I'm surprised that looks a lot like civilization there are fewer notes overall and presumably fewer people running notes there are a lot of reasons why that could be you but these graphs show two factors we don't have to see if there's a correlation so on the Left we're comparing the number of active nodes to the average keith kreis for no reporter actually this is a week over week so maybe even the price is high it's a common belief that more people think it's worth it to run those that doesn't actually seem to be the case at least over the person 2019 which is again all the data we are able to get when we look at them to account in the same graph in blue as the egress in green and it looks like there's actually a medial correlation the new account was one of the slowest plants in June when the price is the highest and the count was quite high during the price tip at the end of the summer so even if this correlation might have been true historically and that's why people think it exists it doesn't seem to be true right now so then on the right we checked with no account against the total size of a default note of my death which is the red line and a parody which is the orange line about 97% of all the things on the network arrangement of those two clients and about 95 percent of the notes on the network or to fall notes and to excellence which recipe much fair game for them obviously the average you know that size is more or less increased over time as boxer mind it seems reasonable to assume that is the default note size gets bigger it gets much spensive and takes more energy to keep snow running and keep it synced so maybe sure people are doing so and that looks like it's the case when we look at the graph on the right or at least we see a clear relationship than there is with the price of these as we move toward Kos and charting the network the notes I Ferdinand won't have to be as much an issue so maybe this patrician will continue on the same trend into next year there are also some interesting experiments happening around the ecosystem to find ways to make bringing a know easier and cheaper so we want to keep tracking so we set up to start measuring decentralization we of course envision this your fabulous graph or tracker that would compare pretty consistent decentralization metrics across protocols that very quickly as balanced as the beginning sort of evolved into comparing apples and oranges and we of course recognize that based off the consensus algorithm you can't compare the same metrics across protocols and so we took that original nineteen subsystems that we showed we said hey what once will be possibly a competently compare across at least 50 protocols and via for instance have a proof of authority or a tune of a Buick law change very different than it would for a report of state law change and so the next step would be this first third conclusion we're not a large appointment not so not necessarily trying to make a decision about if these is doing a good or bad job decent rising theory where you can make any value judgment it's obviously activity on your theory is getting more diverse that develop their minds for action growing the Romanian steady progress on security and on the introduction abuse cases like defy have objected a lot of factors of into destabilization and activities that we've done necessarily have time to analyze yet so it's safe to argue that aetherium is far ahead most of the protocols in terms of sort of this you know network an ecosystem of decent utilization fourth division is that over the next over the lifetime of the network we've seen a much greater complexity and more layers of activity happening off chain which we can see here that will just get worse over time so in some sense this doesn't make that a perfect track but that also makes the security and decentralization of the base settlement layer that much more important to watch of the map we also wanted to point out that it's really important to have this historical data of the network so we can all watch these metrics continue to evolve during this it's much better at tracking Gator than any other critical but it can still be surprisingly hard to find let alone make sense of this very soon you know big picture and stuff so ether scam shared some of their non-public historical didn't record data with us which we really appreciate them we also want to say thanks to our colleagues they like you especially Phenix we meant no more rocket for helping us follow this data increase efficiency so as you mentioned we didn't have time to go over a lot of the other graphs that we pulled the name and analyze for sort of the study of D civilization so we have about a dozen or so more than that time to look but we're going to build a page on Alessio which is a data analytics platform where anyone interested can view those graphs and then hopefully in the near future we'll have a lot of tracker of these metrics over time this will have to constantly I'll take them and reanalyze them so this is just a few of the other graphs we have time to analyze including function calls on the top left that smaller one does include transfer on the bottom right just remove transfer it get it clear review the other function calls looked at each trading volume over Texas and then gas fees and then just a different way of looking at the top the top address concentration so we really hope this is something of interest to the network that people continue to ask some questions about it continue exploring it suggest some additional metrics or ways to quantify those subsystems over time and if you have any questions please definitely reach out to us and enjoy the rest of death [Applause]