<?xml version="1.0" encoding="UTF-8"?><rss xmlns:dc="http://purl.org/dc/elements/1.1/" xmlns:content="http://purl.org/rss/1.0/modules/content/" xmlns:atom="http://www.w3.org/2005/Atom" version="2.0" xmlns:cc="http://cyber.law.harvard.edu/rss/creativeCommonsRssModule.html">
    <channel>
        <title><![CDATA[cyber•Drop - Medium]]></title>
        <description><![CDATA[Distribute Your tokens to any number of addresses in the Ethereum Blockchain - Medium]]></description>
        <link>https://medium.com/cyber-drop?source=rss----fb4b117b4346---4</link>
        <image>
            <url>https://cdn-images-1.medium.com/proxy/1*TGH72Nnw24QL3iV9IOm4VA.png</url>
            <title>cyber•Drop - Medium</title>
            <link>https://medium.com/cyber-drop?source=rss----fb4b117b4346---4</link>
        </image>
        <generator>Medium</generator>
        <lastBuildDate>Thu, 19 Nov 2026 23:17:57 GMT</lastBuildDate>
        <atom:link href="https://medium.com/feed/cyber-drop" rel="self" type="application/rss+xml"/>
        <webMaster><![CDATA[yourfriends@medium.com]]></webMaster>
        <atom:link href="http://medium.superfeedr.com/" rel="hub"/>
        <item>
            <title><![CDATA[Ethereum Address Ranking]]></title>
            <link>https://medium.com/cyber-drop/ethereum-address-ranking-f2b0329867ac?source=rss----fb4b117b4346---4</link>
            <guid isPermaLink="false">https://medium.com/p/f2b0329867ac</guid>
            <category><![CDATA[cryptocurrency]]></category>
            <category><![CDATA[ethereum]]></category>
            <category><![CDATA[blockchain]]></category>
            <category><![CDATA[data-science]]></category>
            <dc:creator><![CDATA[cyber•Drop]]></dc:creator>
            <pubDate>Tue, 12 Nov 2019 08:35:36 GMT</pubDate>
            <atom:updated>2019-11-11T15:04:33.587Z</atom:updated>
            <content:encoded><![CDATA[<p><a href="https://github.com/belya"><em>Anatoli Belchikov</em></a><em>, </em><a href="https://github.com/Snedashkovsky"><em>Serge Nedashkovsky</em></a><em>, </em><a href="https://github.com/xhipster"><em>Dima Starodubcev</em></a></p><h4>For us, ranking addresses by significance in the Ethereum ecosystem has become an interesting and fascinating experiment. By the time our customer’s network was launched, it became clear that the method we had chosen was far from perfect, and a different distribution method was chosen.</h4><h3><strong>Task description</strong></h3><p>We were tasked with identifying “influencers” of the Ethereum network. “Influencers” are addresses that most often interact with a group of tightly coupled addresses. In mathematical terms: addresses associated with the largest number of addresses in their own community (subgraph with dense internal connections and sparser connections between other subgraphs).</p><p>The address ranking algorithm for identifying “influencers” should have the following properties:</p><ul><li><strong>Speed</strong><br>Ranks of vertices need to be calculated for a large number of them, at the time of 01.03.2019 in the Ethereum network there were about 931M transactions and more than 66M addresses. Such amounts of data limit the use of many graph algorithms.</li><li><strong>Repeatability</strong><br>When the algorithm is executed for the same data, the same result will be obtained.</li><li><strong>Predictability</strong><br>Based on ranks, it can be assumed that there are connections between vertices which connections were not known.</li></ul><p>SpringRank was proposed as such an algorithm because it has all of the above properties. This makes SpringRank an interesting tool for finding “influencers” on the Ethereum network, as will be described later.</p><p>SpringRank algorithm is described in Appendix 1.</p><h3>Search for “influencers”</h3><p>The results of ranking by the SpringRank algorithm have two properties:</p><ul><li>Between vertices with similar ranks are likely to have greater number of connections.</li><li>It is more likely to have connections in the pair coming from the vertex with a higher rank.</li></ul><p>Because “influencers” addresses have the highest number of transactions, their rank will be close to the ranks of other related addresses. Our method of searching for “influencers” involves searching for clusters of ranks and selecting vertices that are in the centers of these clusters.</p><p>The method of density clustering adapted to a significant number of clusters (more than 100K) was used to search for clusters of addresses. From each cluster, a group of addresses was selected which ranks are as close as possible to each other, and its size is proportional to the number of addresses of the corresponding cluster.</p><p>For example, the process of processing the address <a href="https://bloxy.info/address/0xab5801a7d398351b8be11c439e05c5b3259aec9b">0xab5801a7d398351b8be11c439e05c5b3259aec9b</a>, allegedly owned by Vitaly Buterin, is shown. The figure below shows this address and its corresponding cluster.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/0*Tf_94BZeIGlZIE4i" /></figure><p>Below is a graph of transactions among some addresses whose ranks are located in the center of this cluster. You can see that addresses often interact with each other, i.e. form a community. You can also see that the address in question sends and receives transactions from other addresses, being the “influencer” in this community.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/649/1*GvthiH7eklS1PF28rAY3kQ.png" /></figure><p>Community leaders, which are dominated by outbound transactions, are the addresses sought. This condition was chosen because clusters with a predominance of outgoing transactions have a higher rank than clusters with a predominance of incoming transactions.</p><p>The dependence of the share of outgoing transactions on the rank of the vertex is shown in the figure below. Similarly, we can imagine the dependence of the share of outgoing transactions on the rank of the cluster center.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/0*5RzsM93bjub65TpW" /></figure><p>The share of distributed reputation for the address was calculated by the formula:</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/136/1*7ETg95YTeH0Q0n18dXKy6g.png" /></figure><p>Where r — the rank of the vertex,</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/121/1*5BnUxsqfZqJ6yxBMBAJskA.png" /></figure><p>maximum and minimum ranks among all addresses.</p><p><strong>Implementation</strong></p><p>We decided that the final list should contain no more than 1M addresses of “influencers”.</p><p>From 66M Ethereum addresses, we selected addresses that meet the conditions:</p><ul><li>Non-zero volume of ETH sent during the last year;</li><li>Balance more than 0.001 ETH;</li><li>It is a wallet. <br>We made the assumption that only wallet owners will be able to take advantage of this reputation.</li></ul><p>We got a list of 4M addresses.</p><p>Next, the ranks were calculated by the SpringRank algorithm for all Ethereum addresses. The existing implementation was accelerated on the GPU using the AMGX library, which allowed us to calculate the ranks for all Ethereum addresses in less than 20 minutes of the algorithm. The implementation can be found in the repositories:</p><ul><li><a href="https://github.com/cybercongress/cyberd/tree/master/research/ethereum">github.com/cybercongress/cyberd/tree/master/research/ethereum </a>— CPU;</li><li><a href="https://github.com/belya/cyberd/blob/gpu/research/ethereum/common/gpu.py">github.com/belya/cyberd/blob/gpu/research/ethereum/common/gpu.py</a> — GPU.</li></ul><p>A GPU implementation of the KDE algorithm was used to find rank clusters. The density estimation results were interpolated to make it easier to find local minima and local maxima by sliding window. Implemented in this way clustering turned out to be orders of magnitude faster than the existing algorithm MeanShift, which allowed to find hundreds of thousands of clusters in an acceptable time. Implementations of KDE and clustering algorithms can be found in repositories:</p><ul><li><a href="https://github.com/dchao34/bbrcit_kde">github.com/dchao34/bbrcit_kde</a> — KDE for CUDA;</li><li><a href="https://github.com/belya/ethereum-rank-airdrop/blob/master/ranks-filtering-process.ipynb">github.com/belya/ethereum-rank-airdrop/blob/master/ranks-filtering-process.ipynb</a> — KDE-based clustering.</li></ul><h3><strong>Distribution of addresses by significance</strong></h3><p>Among the addresses, the first 20k addresses received more than 80% of the distributed significance.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/476/1*wC0wR3OhcYoOLSM7ZISUuw.png" /></figure><p>The table below shows the top 10 addresses in the Ethereum network by distributed significance.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/554/1*eIOdDO4jBamuZ3DFpLQNBA.png" /></figure><h3>Possible tasks</h3><p>What can be done to rank addresses by significance.</p><p><strong>Other methods using SpringRank:</strong></p><ul><li>Strengthening the predictive ability of the SpringRank method;</li><li>Implementation of regularization (L2) to obtain a more stable result;</li><li>Implementation of hybrid algorithms for identifying influencers (SpringRank &amp; Graph analysis algorithms).</li></ul><p><strong>Address ranking based on identifying behavior patterns:</strong></p><ul><li>investor;</li><li>developer;</li><li>experimenter;</li><li>financial center;</li><li>trader;</li><li>stock exchange;</li><li>and others.</li></ul><p><strong>Measurement of activity addresses:</strong></p><ul><li>creation and calls of smart contracts;</li><li>total investment;</li><li>turnover in ETH and ERC20 tokens;</li><li>activity when receiving airdrop;</li><li>and others.</li></ul><h3><strong>Appendix 1. Description of SpringRank algorithm</strong></h3><p><a href="https://arxiv.org/pdf/1709.09002.pdf">The SpringRank algorithm</a> solves the problem of ranking the vertices of oriented multigraph. The algorithm assumes finding of ranks through energy minimization for the system of springs with identical length. Each spring corresponds to an arc and is stretched between points whose only coordinate corresponds to the rank of the beginning and end of the arc. The spring coefficients are proportional to the arc weights between the respective vertices. The description above can be represented by the total energy of all springs. :</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/331/0*2j3GLGqaH4Cf79N5" /></figure><p>The minima of this expression are at the points where the derivative is zero, i.e. they are solutions of the following system of equations:</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/333/0*8317McEYXwuwl-Df" /></figure><p>Where</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/70/0*3HUytB27STIfk5yC" /></figure><p>- the diagonal matrices of the in- and outdegree of the vertices, respectively, 1 — a vector that consists of ones, and A — the adjacency matrix of the graph.</p><p>It is worth noting that, based on the definition, the matrix</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/192/0*b1IfMaqWFl_ZVyM7" /></figure><p>has the rank</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/74/0*CjVu69U3vUupC6Kt" /></figure><p>(the inequality is strict if the graph is not connected), and there is a line on which all solutions of the system of equations lie (1). The article proposes to choose the solution in which one of the ranks</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/50/0*1MF3N0NkjT4VOakd" /></figure><p>Thus, the algorithm has one local minimum.</p><p>It should be noted that the generative model described in the article about Spring Rank allows us to use rank values to calculate the number and direction of arcs between vertices by formulas:</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/221/0*Gk184R-7JXc7MKmi" /></figure><figure><img alt="" src="https://cdn-images-1.medium.com/max/208/0*1h0h0sGtXuD8AwYA" /></figure><p>I.e. the number of arcs between vertices</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/23/0*kmHz6NzceNcr_lmI" /></figure><p>increases as their rank difference decreases, and the number of arcs from a vertex with a lower rank is less than the number of arcs in the opposite direction (these values are equal to</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/44/0*NEeIaCIp8ha5t8hS" /></figure><p>and</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/89/0*BBvSYhWyPxZnF7IQ" /></figure><p>).</p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=f2b0329867ac" width="1" height="1" alt=""><hr><p><a href="https://medium.com/cyber-drop/ethereum-address-ranking-f2b0329867ac">Ethereum Address Ranking</a> was originally published in <a href="https://medium.com/cyber-drop">cyber•Drop</a> on Medium, where people are continuing the conversation by highlighting and responding to this story.</p>]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[Huge Ethereum Mixer]]></title>
            <link>https://medium.com/cyber-drop/huge-ethereum-mixer-68d08bc6af0a?source=rss----fb4b117b4346---4</link>
            <guid isPermaLink="false">https://medium.com/p/68d08bc6af0a</guid>
            <category><![CDATA[blockchain]]></category>
            <category><![CDATA[trading]]></category>
            <category><![CDATA[data-science]]></category>
            <category><![CDATA[ethereum]]></category>
            <category><![CDATA[cryptocurrency]]></category>
            <dc:creator><![CDATA[cyber•Drop]]></dc:creator>
            <pubDate>Tue, 12 Nov 2019 08:35:07 GMT</pubDate>
            <atom:updated>2019-11-01T08:59:26.814Z</atom:updated>
            <content:encoded><![CDATA[<p><em>The article was removed from the </em><a href="https://medium.com/u/9fcb6f8cf209"><em>cyber • Fund</em></a> <em>blog and was republished on this blog.</em></p><p><a href="https://github.com/Snedashkovsky"><em>Serge Nedashkovsky</em></a></p><p><strong>68% of total Ethereum transaction value controlled by one system</strong></p><p><strong>UPDATE: We have </strong><a href="https://github.com/cyberFund/DataScience"><strong>released</strong></a><strong> the Jupiter notebook on Github so that anyone can repeat the analysis done.</strong></p><p>Analyzing the Ethereum transactions, <a href="https://cyber.fund/"><strong>cyber•Fund</strong></a> has come across a finding that struck us so much that we rushed to dig deeper into this issue. We now wish to share our findings with a community hoping to come up with an explanation together.</p><h3>What we found</h3><figure><img alt="" src="https://cdn-images-1.medium.com/max/843/1*cP_Sh1lIOhFcKdCX8FY_pw.png" /></figure><p>The clusterization of all Ethereum addresses from the inception of Ethereum until 15.09.2017 revealed a class of addresses that we will call in this paper temporary. These are addresses where funds come and leave within a short time interval — no longer than 1 hour, after which the addresses are never used again. The temporary addresses constituted 46% of all active addresses and processed 65% of total transaction value during the analysed period. Analyzing transactions where these addresses were involved, we managed to collect piece by piece a full picture of what was going on:</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/916/1*n4us6JDygm1aiVNYpt1cZA.png" /></figure><p>In the centre of the picture above, one can see a core of the Mixer which consists of more than 95% of temporary addresses. This core interacts with a group of shell addresses, or a shell layer, which includes both permanent and temporary addresses. The shell layer in its turn receives ETH from what we will call input addresses and sends ETH to output addresses displayed left- and right-hand on the scheme respectively. We looked up the names of the owners of these addresses in Etherscan. Only a few of these addresses had a name of the owner attached to them on <a href="https://etherscan.io/">Etherscan</a>. The other names displayed on the picture above are the names mentioned in users’ commentaries on Etherscan, so we can only assume that they are the real owners of these addresses. In the end, it turned out that the total amount transferred into and out of the core is 4 times higher than the total that entered and left the shell and the core taken together. This made us think of a mixer mechanism (further referred to as Mixer).</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/865/1*W3-YEaDsQVL_o7ykH67mfg.png" /></figure><figure><img alt="" src="https://cdn-images-1.medium.com/max/784/1*i6Eh-41mwU_L1opYVE3BOQ.png" /></figure><p>Of all transactions executed on the Ethereum blockchain during the analysed period, addresses with the incoming amounts of ~500, ~1000, ~2000, ~3000, ~5000 and ~10,000 ETH constitute <strong>68.5% (2,601,041,693.6 from 3,791,195,132.0 ETH) in money terms and 10.7% (6,216,314 from 58,035,623) in terms of numbers</strong>. The further analysis shows that these addresses are linked with each other and might be controlled by a single entity.</p><p>This is how the share of the Mixer in all Ethereum transactions has changed over time:</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/965/1*NEfrrb8_qXOAl393B4HZBA.png" /></figure><figure><img alt="" src="https://cdn-images-1.medium.com/max/974/1*c6hZRGTw7Z8S_s0SPV1Bdg.png" /></figure><p>The system seems to be first tested in 2016, and since the start of 2017 it came into active use. This might be explained by the increasing capitalization and liquidity of Ethereum. Most interesting is that an overall growth pattern of Ethereum transactions looks very differently when the Mixer share is excluded. If the Mixer transactions are left out of the analysis, it becomes evident that they contributed to most of the overall Ethereum transaction volume growth.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/964/1*CngsrwwueumLuKIliMc4FA.png" /></figure><h3>Analysis that was done</h3><figure><img alt="" src="https://cdn-images-1.medium.com/max/836/1*9cTXxc4ADvfSG5svs0KQXQ.png" /></figure><p>In terms of the incoming transaction volume, these addresses are distributed as follows:</p><p><strong>Distribution of Addresses by Transaction Volume</strong></p><figure><img alt="" src="https://cdn-images-1.medium.com/max/967/1*BtUY1Oy_sv-Qmbc4wULPNQ.png" /></figure><p>Out of the total of 6,282,858 addresses involved in all transactions executed on the Ethereum blockchain since its inception till 15 September 2017 the following sets of addresses gained our attention:</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/757/1*YFG19X_iSWOU3QKY-YxrQQ.png" /></figure><p>These addresses account for 67.5% of all transferred ETH and constitute 8.5% of the total number of transactions on Ethereum during the analysed period. So why do we think these addresses are linked?</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/843/1*6jPvuhYpefRRyblhNogMlA.png" /></figure><p>The graph below displays how these sets of addresses replace each other almost one by one. Take one set of addresses, e.g. addresses with incoming amounts of around 1000 ETH. After being active for some time, this set of addresses becomes inactive and this is when another set steps in, e.g. that with 3000 incoming ETH per transaction. Thus, addresses “act” as if orchestrated following one another over time which makes us think there is a certain system managing these activities. These addresses constitute the core of the scheme.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/973/1*5DALtunHXfhlkz72IyXTXA.png" /></figure><h3>Findings</h3><p>Analyzing the system further, we identified temporary and permanent addresses that surround the core and are linked with it. The calculations for the core and the linked addresses for the period starting from the inception of Ethereum until 15.09.2017 bring about the following results:</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/860/1*ntq4y4x5ZWWSe9uW8mChXg.png" /></figure><h3>Hypotheses</h3><p>These are possible explanations for the detected activities we could come up with:</p><ol><li>The protection offered to clients by crypto-exchanges: all clients’ funds are mixed so that the funds’ sources cannot be tracked and those holding clean money cannot be unjustifiably accused of any illegal activity</li><li>A mechanism set in place to protect U.S. residents who wish to avoid control from U.S. regulatory bodies</li><li>A mechanism used by a large private exchange to preserve the privacy of its clients; this exchange might be operating with fiat money</li><li>A mechanism used to securely transfer crypto-assets between crypto-exchanges</li><li>Any kind of Ether-laundering scheme</li></ol><p>These are only hypotheses which we would like to discuss with anyone interested in our findings. If you have any other suggestions or explanations, please do not hesitate to share them with us. You can find more details in the Appendix.</p><p>You can always contact us at: <strong>datascience@cyber.fund</strong>, <strong>analytics@cyber.fund</strong></p><h3>Appendix</h3><p><strong>Note: If you wish to look up the addresses below on </strong><a href="https://etherscan.io/"><strong>Etherscan</strong></a><strong> by yourself, use the list in this</strong><a href="https://github.com/cyber-drop/DataScience/blob/master/transaction_analysis_huge_ethereum_mixer_171125.ipynb"><strong> Jupyter Notebook</strong></a><strong>.</strong></p><h4>Top 20 input addresses (ETH transferred into the Mixer)</h4><figure><img alt="" src="https://cdn-images-1.medium.com/max/753/1*ZBrSABPyCh-KbsryCyRsUA.png" /></figure><h4>Top 20 output addresses (ETH transferred out of the Mix)</h4><figure><img alt="" src="https://cdn-images-1.medium.com/max/746/1*xnZoUWlvtg8oiZIBvUcUYw.png" /></figure><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=68d08bc6af0a" width="1" height="1" alt=""><hr><p><a href="https://medium.com/cyber-drop/huge-ethereum-mixer-68d08bc6af0a">Huge Ethereum Mixer</a> was originally published in <a href="https://medium.com/cyber-drop">cyber•Drop</a> on Medium, where people are continuing the conversation by highlighting and responding to this story.</p>]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[New Metrics on SANgraphs: ETH Locked Over Time]]></title>
            <link>https://medium.com/cyber-drop/new-metrics-on-sangraphs-eth-locked-over-time-1261af6a27bb?source=rss----fb4b117b4346---4</link>
            <guid isPermaLink="false">https://medium.com/p/1261af6a27bb</guid>
            <category><![CDATA[dapps]]></category>
            <category><![CDATA[blockchain]]></category>
            <category><![CDATA[santiment]]></category>
            <category><![CDATA[ethereum]]></category>
            <category><![CDATA[analytics]]></category>
            <dc:creator><![CDATA[cyber•Drop]]></dc:creator>
            <pubDate>Sat, 25 May 2019 11:22:21 GMT</pubDate>
            <atom:updated>2019-05-25T11:21:19.057Z</atom:updated>
            <content:encoded><![CDATA[<p><em>Authors and metrics creators:</em> <a href="https://github.com/ninjascant"><em>Maxim Odegov</em></a><em> and </em><a href="https://github.com/Snedashkovsky"><em>Serge Nedashkovsky.</em></a></p><p>Over the past few years, a number of DApps have emerged that require users to lock a certain amount of Ethereum in their smart contracts in order to access their services. The aim of our new metrics, now live on <a href="https://santiment.net/dashboards">SANgraphs</a>, is to measure the amount of ETH locked away in these contracts.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/0*K2PZcSEPOHP3h2Eu.png" /><figcaption>Eth Locked Metrics, now live on <a href="https://santiment.net/dashboards">SANgraphs</a></figcaption></figure><p>Our new charts showcase full historical and real-time data of ETH locked in 5 of the most popular DApps of the abovementioned variety:</p><p>Here’s an overview of each system and the amount of ETH they currently store.</p><h3>Maker DAO</h3><p>At the moment, Maker DAO is the single biggest DApp in terms of locked ETH, and by a wide margin. In return for locking Ethereum it issues DAI, a USD-pegged stablecoin. Any user of the platform is able to issue DAI tokens by locking ETH or burn DAI to get the collateralized Ether back.</p><h4>Maker DAO’s Working Principles</h4><p>The central element in the Maker system is the contract called SaiProxyCreateAndExecute (0x526af336D614adE5cc252A407062B8861aF998F5). The process for issuing DAI tokens is as follows:</p><ol><li>To issue DAI, you have to call the SaiProxyCreateAndExecute contract with the createOpenLockAndDraw() method. The transaction also includes the desired amount of Ether to be locked.</li><li>The contract changes Ether to WrappedEther (WETH) and sends it to the SaiTub contract (0x448a5065aebb8e423f0896e6c5d525c040f59af3). This contract stores all WETH that is locked in the system.</li><li>Next, SaiProxyCreateAndExecute contract generates PETH tokens equivalent to WETH.</li><li>SaiProxyCreateAndExecute issues DAI tokens and sends it to the user.</li><li>When being issued DAI, the user forms a debt position. Over time, a so-called Stability Fee is accumulating for the position. To unlock his/her Ether, the user has to burn the issued DAI tokens and pay the stability fee in MKR. In turn, PETH are burnt and WETH converted back to ETH and sent to the user.</li></ol><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/0*t6G1tjkTugnTP-ru.png" /><figcaption>Figure 1. Main Mechanism of Maker DAO</figcaption></figure><h4>WETH and PETH tokens</h4><p>Wrapped ETH is used within the Maker system and is the equivalent of ETH compatible with the ERC-20 standard. More than 85% of all WETH tokens in existence are locked in the Maker network (with the SaiTub contract)</p><p>Besides WETH, MakerDAO also utilizes Pooled ETH (PETH), another ERC-20 compatible token. It is used directly as a collateral asset to issue DAI. By issuing DAI, a user locks PETH and can get it back by burning DAI.</p><h4>ETH Locked Volume</h4><p>At the moment, the Maker system stores more Ether than all other mentioned systems combined.</p><p>The current balance of the SaiTab contract (the contract which stores all WETH) is 1.79M WETH. Here is the contract’s balance over time:</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/0*u-T214ifuZP9phJG.png" /><figcaption>Figure 2. ETH locked in MakerDAO</figcaption></figure><h3>Ethereum Name Service</h3><p>Ethereum Name Service (ENS) offers a secure way to address resources both on and off the blockchain using simple, human-readable names (like example-name.eth). The ENS system consists of two main parts:</p><ul><li>A registrar contract (0x6090a6e47849629b7245dfa1ca21d94cd15878ef), which stores all domains, its owners and resolvers domains.</li><li>A number of Resolvers — the contracts that transform domain names into contract addresses. This is achieved by hashing the name and calling the addr() method of the resolver contract. This contract provides the user with a smart contract address linked to the provided name.</li></ul><p>Another important part of the ENS system are the auction contracts, which help facilitate bids for domain names. Following a bidding process, the top bet is stored on the auction contract. To compute the Ether locked in the ENS network, we need to sum the balances of all auction contracts.</p><h4>Working Principles</h4><ol><li>To register the name in ENS, you have to start the auction by sending the transaction to a Registrar contract.</li><li>In turn, the ENS-Register creates a unique contract and the bidding begins.</li></ol><p>Following the auction, all users get their ETH back except for the winner, whose funds are stored on the contract in return for the desired name. ENS also takes a small commission fee from all the participants.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/0*hgGeG3HF1DxY_EC0.png" /><figcaption>Figure 3. Main Mechanism of ENS</figcaption></figure><h4>ETH Locked Volume</h4><p>There are 428,807 ENS auction contracts today. Their total balance is 161K ETH. At its peak on June 3rd, 2017, the balance of ETH locked in ENS reached 752k ETH.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/0*IM1w84H-CIPYjzOw.png" /><figcaption>Figure 4. ETH locked in ENS</figcaption></figure><h3>Compound</h3><p>Compound is a protocol that aims to easily create money markets — token pools with an algorithm for calculating interest rates based on a token’s supply and demand.</p><p>Users invest or borrow tokens into pools and earn or pay the interest. As a result, a single money market for an individual ERC-20 token is created. A money market is essentially a smart contract which stores a list of given market participants and their deposited or lent balances.</p><p>Compound’s discerning feature is that the process of investing and/or lending is fully automatized. Users don’t (have to) care about who gets their tokens or when the tokens are paid back.</p><h4>Working Principles</h4><p>The key element of Compound’s system is the contract called MoneyMarket (0xbd8e0def19aabbc3751b0b5ee0558cadffce759b). It allows users to lend or borrow tokens.</p><ol><li>To lend your tokens to the system you have to call the Supply() method of the MoneyMarket contract. After that the interest rate is paid to the user. Remember, the interest rate is calculated according to the supply and demand for a particular token.</li><li>A user is able to get his tokens back with the Withdraw() method.</li><li>There’s also an option to borrow tokens. The main stipulation is that the borrower’s balance must be higher than the borrowed sum. That said, the borrower’s tokens are not frozen and the interest rate is levied on the tokens.</li><li>To repay the loan, a repayBorrow() method is called.</li></ol><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/0*CrH_n_s4MfXNjfMy.png" /><figcaption>Figure 5. Main Mechanism of Compound</figcaption></figure><h4>ETH Locked Volume</h4><p>Currently, a total of 74.5k ETH is locked in Compound’s contract. It is a historical maximum for the Compound network and has been growing steadily in the past few months.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/0*y_Tzm5Bwx44rK88r.png" /><figcaption>Figure 6. ETH locked in Compund</figcaption></figure><h3>Uniswap</h3><p>Uniswap is a decentralized protocol for an automated token exchange on Ethereum. Its key feature is that trades happen without makers, made possible by special exchange contracts that essentially double as liquidity pools.</p><p>Any user is able to replenish a pool by sending ERC-20 tokens and an equivalent amount of Ether. In return, the user earns some share of the total commission fees for the contract’s trades. A user’s share is proportional to his funds in the pool.</p><h4>Working Principles</h4><ol><li>To exchange Ether for tokens, a user has to send ETH to an exchange contract.</li><li>The exchange contract gets a needed amount of tokens from the pool and sends it to the user.</li><li>AddLiquidity() method is used to replenish the liquidity pool. Once called, the user receives ERC-20 pool tokens proportional to his/her share of funds in the pool.</li></ol><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/0*KFyCrU6KmHC7Ts2s.png" /><figcaption>Figure 7. Main Mechanism of Uniswap</figcaption></figure><h4>ETH Locked Volume</h4><p>Uniswap contracts currently lock 29.3k ETH. Same as Compound, Uniswap-locked Ether is steadily growing and has just recently reached its all time high (31.7k ETH), before retracing slightly.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/0*VobEfbPIiNd-QUrj.png" /><figcaption>Figure 8. ETH locked in Uniswap</figcaption></figure><h3>Augur</h3><p>Augur is a decentralized prediction market protocol where users can bet on the outcome of predefined future events.</p><p>To be able to bet, a user has to send his Ether to a Delegator contract placed at 0xd5524179cB7AE012f5B642C1D6D700Bbaa76B96b.</p><h4>Working Principles</h4><p><strong>Creating a Prediction Market</strong></p><p>When creating a new market, the user must specify the following parameters:</p><ol><li>Expiration date — the date when a predefined event happens.</li><li>Resolution source — a trustworthy source that will signal the event’s outcome.</li><li>Reporter — a user that writes a report about the event’s outcome.</li></ol><p>There are two collaterals needed to create a market on Augur:</p><ol><li>A validity bond paid in ETH, which acts as a guarantee for the creation of a valid resolution source. A validity bond is paid back if and only if the outcome is valid.</li><li>A designated report ‘no-show’ bond paid in REP. It acts as a guarantee for choosing the right reporter. The reporter must sound off on the outcome in 3 days after the expiration date.</li></ol><p><strong>Bet Creation</strong></p><p>Bets in Augur are created as follows:</p><ol><li>To bet on a certain market, a user sends ETH and triggers a publicTradeWithLimit() method of Augur’s Trade contract.</li><li>The trade contract then calls the depositEtherFor() method of the Delegator contract (and sends the user’s Ether) to update a user’s balance.</li><li>The trade contract calls the createOrder() method of the createOrder contract.</li></ol><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/0*I7f2vjfTvvJMU7PC.png" /><figcaption>Figure 9. Main Mechanism of Augur</figcaption></figure><h4>ETH Locked Volume</h4><p>At the moment, 4k ETH is locked in the Augur system. The historical maximum of 14.67k ETH was reached on February 2nd, 2019.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/0*SN_ccouAZ9z0I9BO.png" /><figcaption>Figure 10. ETH locked in Augur</figcaption></figure><h3>Сonclusion</h3><p>At the moment, all described systems combined lock a grand total of 2.058M ETH. The overwhelming majority — more than 1.7M Ether — is locked away in the MakerDAO system. Here are the 5 systems ranked by the amount of ETH stored:</p><ol><li>Maker DAO (1.79M ETH)</li><li>ENS (161k ETH)</li><li>Compound (74.5k ETH)</li><li>Uniswap (29.3k ETH)</li><li>Augur (4k ETH)</li></ol><p>And here’s the combined volume of ETH locked in all 5 DApps over time:</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/0*dhwFjSrewL0IuPO7.png" /><figcaption>Figure 11. ETH Locked in all 5 DApps over Time</figcaption></figure><p>The total amount of Ether locked in these 5 projects constitutes 2.23% of Ethereum’s total supply. Interestingly, this cumulative balance exceeds the current balance of all miner pools and miners on the Ethereum network combined (2.05M ETH locked vs 1.2M ETH owned by miner pools and miners).</p><p>You can now access and analyze historical and real-time data for ETH locked in all 5 systems (as well as their combined balance) on <a href="https://santiment.net/dashboards">SANgraphs.</a></p><p><em>Originally published at </em><a href="https://santiment.net/blog/new-metrics-eth-locked/"><em>https://santiment.net</em></a><em> on May 21, 2019.</em></p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=1261af6a27bb" width="1" height="1" alt=""><hr><p><a href="https://medium.com/cyber-drop/new-metrics-on-sangraphs-eth-locked-over-time-1261af6a27bb">New Metrics on SANgraphs: ETH Locked Over Time</a> was originally published in <a href="https://medium.com/cyber-drop">cyber•Drop</a> on Medium, where people are continuing the conversation by highlighting and responding to this story.</p>]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[How changeable are ERC-20 tokens in time? Behavioral analysis studies]]></title>
            <link>https://medium.com/cyber-drop/how-changeable-are-erc-20-tokens-in-time-behavioral-analysis-studies-163df851758f?source=rss----fb4b117b4346---4</link>
            <guid isPermaLink="false">https://medium.com/p/163df851758f</guid>
            <category><![CDATA[ethereum]]></category>
            <category><![CDATA[erc20]]></category>
            <category><![CDATA[data-science]]></category>
            <category><![CDATA[behavioral-economics]]></category>
            <category><![CDATA[blockchain]]></category>
            <dc:creator><![CDATA[cyber•Drop]]></dc:creator>
            <pubDate>Tue, 21 May 2019 18:53:31 GMT</pubDate>
            <atom:updated>2019-05-21T18:52:55.528Z</atom:updated>
            <content:encoded><![CDATA[<p><em>by </em><a href="https://github.com/Snedashkovsky"><em>Serge Nedashkovsky</em></a><em> and </em><a href="https://github.com/ninjascant"><em>Maxim Odegov</em></a></p><p>One of the key events of 2017 was the explosive growth in the popularity of ERC-20 tokens (hereinafter referred to as tokens). The growth in the volume of transactions with tokens led to the growing popularity of the Ethereum network and the cost of ETH.</p><p>In 2017 and 2018, the daily number of token transfers grew from 1K to 400K and peaked at over 1M. At the moment, the daily number of token transfers has stabilized and is more than 250K.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/916/0*dH3pNHwsfGfuNwub" /><figcaption>Figure 1. Token Transfers</figcaption></figure><p>Token transfers make a significant contribution to the use of gas of the network.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/945/0*uRTcvLi_49LplW7c" /><figcaption>Figure 2. Gas Used</figcaption></figure><h3>Classification of token transfers</h3><p>It is interesting to understand what types of transfers exist and what is the distribution between them.</p><p>We have identified three main classes of transfers:</p><ul><li><strong>Ordinary users transfers</strong>. <em>Individual transfers carried out by common users.</em>e number of transfers in this class is more than 100.<br>Other transfers of non-tradable tokens also belong to the class.</li><li><strong>Bounty &amp; Promo Airdrops, Crowdsales</strong>. <em>These transfers are made within the distribution, both monetized (Crowdsales) and free for the recipient (Bounty &amp; Promo Airdrops). These token distribution methods are characterized by a large number of transfers over a limited period of time.</em></li><li><strong>Traders &amp; Exchanges transfers</strong>. <em>Transactions aimed at the implementation of medium and high-frequency token trading. A significant number of transactions and a small number of unique recipients are characteristic features.</em></li></ul><p>Next, we consider the metrics on the basis of which we will divide the classes. Metrics are calculated for the address pairs of the sender of the transfer — token.</p><h3>Count of Token Transfers</h3><p>A key indicator for the division of classes is the number of token transfers per unit of time calculated for each sender. For clarity, we have divided the transfers into 4 classes:</p><ul><li><strong>Single transfers</strong> <em>(equals 1)</em>;</li><li><strong>Not a significant number of transfers</strong><em> (from 2 to 10)</em>;</li><li><strong>Significant number of transfers</strong> <em>(from 11 to 100)</em>;</li><li><strong>A large number of transfers </strong><em>(more than 100)</em>.</li></ul><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*I6g4d_2uMmlScutD1qee6Q.png" /><figcaption>Figure 3. Token Transfers by daily count</figcaption></figure><p>As can be seen, transfers with the number of sent transactions more than 100 for one sender per day dominate. Classes with significant and not a significant number of transfers we have combined due to their small size.</p><h3>The Uniqueness of Recipients of Transfers</h3><p>The second most important metric characterizing the token transfer is the ratio of the number of sent transfers to the number of unique recipients for the selected sender per day. This metric shows the frequency of interaction with token recipients. We have divided transfers into the following classes:</p><ul><li><strong>Single transfers</strong> (<em>For these transfers, this ratio is 1);</em></li><li><strong>Transfers to unique recipients (</strong><em>Less than 1.1);</em></li><li><strong>Transfers with a significant number of unique recipients (</strong><em>From 1.1 to 2);</em></li><li><strong>Transfers to a limited number of unique recipients</strong> (<em>Greater than 2).</em></li></ul><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*WN2-bdZqJEclMEpJ9OeEDA.png" /><figcaption>Figure 4. Token Transfers by uniqueness of recipients</figcaption></figure><p>Transfers to unique recipients dominate (45%) and then there are single transfers (19%) and transfers to a limited number of unique recipients (17%). Transfers with a significant number of unique recipients we have added to the class of transfers to unique recipients.</p><h3>Token Liquidity</h3><p>In order to assess the activity of ERC-20 tokens, we have calculated the number of transfers made with these tokens on decentralized exchanges.</p><p>We have divided transfers into the following classes:</p><ul><li><strong>Non-tradable tokens </strong><em>(No DEX trades).</em></li><li><strong>Low liquid tokens</strong> <em>(The number of DEX trades is less than 5,000).</em></li><li><strong>High liquid tokens</strong> (<em>The number of DEX trades is more than 5,000).</em></li></ul><p>Among all the ERC-20 tokens with which at least one transfer was made, 12% (5433) of the tokens are traded on decentralized exchanges. At the same time, most of the transfers (in total, 65.5%) occurred with traded tokens.</p><p>A similar ratio is observed for the number of addresses that sent tokens:</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/780/1*Mf1216IqSMXbXIkAZhvXGQ.png" /><figcaption>Figure 5. Token and sending address count by token liquidity</figcaption></figure><p>However, as can be seen from the following chart, the share of token transfers not traded on decentralized exchanges increases over time:</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*gKwxZXv_zkDxn7f8f0x8fQ.png" /><figcaption>Figure 6. Token Transfers by liquidity</figcaption></figure><h4>Merging Metrics. Count and Uniqueness of Recipients of Transfers</h4><p>Using the first two metrics: the number of transfers and the uniqueness of the recipients, we divided the token transfers into classes as follows:</p><ul><li><strong>Ordinary users transfers. </strong><em>The number of transfers for the sender-token pair is 1.</em></li><li><strong>Bounty &amp; Promo Airdrops, Crowdsales. </strong><em>Transfers with a unique recipient and a significant number of unique recipients (the ratio of the number of transfers to the number of unique recipients is less than 2). The number of transfers in this class is more than 100.</em></li><li><strong>Traders &amp; Exchanges transfers. </strong><em>Other transfers</em></li></ul><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*pZJP9p9J3kiW1gkKR560Rw.png" /><figcaption>Figure 7. Token Transfers by Classes</figcaption></figure><h3>Merging Metrics. Add Liquidity</h3><p>We have defined how transfers of the above classes are distributed by token liquidity into groups: non-tradable tokens, low and high liquid tokens.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*xaMCKrtZtcwYuVgN0mVCHg.png" /><figcaption>Figure 8. Token Transfers by liquidity. Bounty &amp; Promo Airdrops, Crowdsales</figcaption></figure><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*TLQp8vBpdV_uNFwjAkP3sw.png" /><figcaption>Figure 9. Token Transfers by liquidity. Traders &amp; Exchanges</figcaption></figure><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*bVAPcTqvGJJHRemxtDbbYg.png" /><figcaption>Figure 10. Token Transfers by liquidity. Ordinary users</figcaption></figure><p>We see that there is a significant amount of non-tradable tokens in Traders &amp; Exchanges transfers. We assume that these transfers should be classified as Bounty &amp; Promo Airdrops, Crowdsales.</p><p>Therefore, the rules for assigning to classes will look like this:</p><ul><li><strong>Ordinary users transfers</strong></li></ul><blockquote>The number of transfers for the sender-token pair is 1.</blockquote><ul><li><strong>Bounty &amp; Promo Airdrops, Crowdsales</strong></li></ul><blockquote>Transfers with a unique recipient and a significant number of unique recipients (the ratio of the number of transfers to the number of unique recipients is less than 2). The number of transfers in this class is more than 100.</blockquote><blockquote>Other transfers of non-tradable tokens also belong to the class.</blockquote><ul><li><strong>Traders &amp; Exchanges transfers</strong></li></ul><blockquote>Other transfers of tradable tokens.</blockquote><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*eZmA3ieOPSPmlwjBDrUHWg.png" /><figcaption>Figure 11. Token Transfers by Classes</figcaption></figure><p>As can be seen in fig.11 the <strong>last 6 months</strong> the number of token transfers by classes is stable and amounted to:</p><ul><li><strong>Ordinary users</strong> are <strong>55K</strong> transfers per day;</li><li><strong>Bounty &amp; Promo Airdrops, Crowdsales</strong> are <strong>190K</strong> transfers per day;</li><li><strong>Traders &amp; Exchanges</strong> are <strong>60K</strong> transfers per day.</li></ul><p>The use of ERC-20 by the major types of token senders is stable, despite the depreciation of the ether since August of 2018. This once again underlines the viability and convenience of the ERC-20 token.</p><p><em>Source: The calculations were made using </em><a href="https://github.com/cyber-drop/ethereum_analytical_db/"><em>Ethereum Analytical DB</em></a></p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=163df851758f" width="1" height="1" alt=""><hr><p><a href="https://medium.com/cyber-drop/how-changeable-are-erc-20-tokens-in-time-behavioral-analysis-studies-163df851758f">How changeable are ERC-20 tokens in time? Behavioral analysis studies</a> was originally published in <a href="https://medium.com/cyber-drop">cyber•Drop</a> on Medium, where people are continuing the conversation by highlighting and responding to this story.</p>]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[What decentralized exchanges are and how they work]]></title>
            <link>https://medium.com/cyber-drop/what-decentralized-exchanges-are-and-how-they-work-def21b8b4706?source=rss----fb4b117b4346---4</link>
            <guid isPermaLink="false">https://medium.com/p/def21b8b4706</guid>
            <category><![CDATA[trading]]></category>
            <category><![CDATA[decentralized-exchange]]></category>
            <category><![CDATA[blockchain]]></category>
            <category><![CDATA[cryptocurrency]]></category>
            <category><![CDATA[ethereum]]></category>
            <dc:creator><![CDATA[cyber•Drop]]></dc:creator>
            <pubDate>Sat, 02 Mar 2019 04:12:23 GMT</pubDate>
            <atom:updated>2019-02-28T02:35:27.000Z</atom:updated>
            <content:encoded><![CDATA[<p><a href="https://github.com/ninjascant"><em>by Maxim Odegov</em></a><em> and </em><a href="https://github.com/Snedashkovsky"><em>Serge Nedashkovsky</em></a></p><p>In Ethereum network there are two main DEX operating principles: off-chain and on-chain ones. In the former case, orders are kept on the exchange servers and the interaction with blockchain happens at the moment when a customer wants to close an order. In the latter case, orders are kept in the exchange smart contracts.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*nhq4oslUXqgwpxxwRCVn9A.png" /></figure><h3>Off-chain</h3><p>Most existing exchanges use this operating principle. In general terms, it can be described in the following way:</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/602/0*ScRqXWu9HHQ6SvBP" /><figcaption><em>Figure 1. Off-chain trading principle</em></figcaption></figure><ol><li>A seller (maker) makes an order, signs it and sends it to the exchange server.</li><li>A buyer (taker) selects the order and receives its details from the server.</li><li>The buyer sends a transaction to the exchange contract with a trade method call</li><li>The contract checks the order authenticity and if it goes well, the contract changes the participants’ balances and completes the trade.</li></ol><iframe src="" width="0" height="0" frameborder="0" scrolling="no"><a href="https://medium.com/media/5f0f69c404114aaf9ba1c4d9d4062467/href">https://medium.com/media/5f0f69c404114aaf9ba1c4d9d4062467/href</a></iframe><p>The example of the exchanges using an off-chain principle are IDEX, Bancor, as well as the exchanges based on the protocol 0x: DDEX, Starbit.</p><h3>On-chain</h3><p>The main difference of the on-chain principle is that the order signature is kept in the contract.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/602/0*tPPtXfbzi2juPfaL" /><figcaption><em>Figure 2. On-chain trading principle</em></figcaption></figure><p>The example of the exchange using an on-chain principle is Ethex. Another example is Etherdelta exchange. Despite the fact most Etherdelta trades use the off-chain principle, its smart contract has the method “order” which helps make orders and keep them in the contract memory. Using the example of Etherdelta exchange we can show on-chain trades in the following way:</p><ol><li>A seller calls the order method which creates the order hash and saves it into the contract storage (<a href="https://gist.github.com/ninjascant/23fb77a0341a238698677a2ccce4f854">code</a>)</li><li>A buyer can receive the order details directly from blockchain using an Order event log.</li><li>Similar to the previous method, the buyer sends a transaction containing the “trade” method call to the exchange contract.</li><li>The contract checks whether one of the following conditions is true: the received order hash is contained in the orders object; the order is signed by a seller. The first check is required for on-chain based orders, the second one is for the off-chain orders.</li></ol><iframe src="" width="0" height="0" frameborder="0" scrolling="no"><a href="https://medium.com/media/04365cba77361ddfc16cca84d6feb058/href">https://medium.com/media/04365cba77361ddfc16cca84d6feb058/href</a></iframe><p>If the conditions are met, the contract changes the participants’ balances and completes the trade.</p><h3>Balance change</h3><p>In the process of trading user balances are changed in different ways.</p><p>Some exchanges keep finances on the smart contracts. To make or close an order on such an exchange a user needs to transfer tokens from his address to the exchange smart contract using the deposit or depositToken methods. In this case user balances are kept in the inner contract objects. When the trade is completed successfully the contract rewrites the balances kept in its storage, and tokens stay on the contract balance.</p><iframe src="" width="0" height="0" frameborder="0" scrolling="no"><a href="https://medium.com/media/182fa5c70a881934a28d63d641e6faf0/href">https://medium.com/media/182fa5c70a881934a28d63d641e6faf0/href</a></iframe><p>User balances change actually when the methods withdraw/withdrawToken are called.</p><p>Other exchanges work differently and change user balances each time when the order is closed. The example of exchanges using this principle are the exchanges working with the protocol 0x. in this case, before the trade is made users send a transaction with the approve method which allows the exchange proxy contract to send a certain number of tokens in the process of trading. When an order closes the exchange contract calls the proxy contract which makes a transfer between user addresses:</p><iframe src="" width="0" height="0" frameborder="0" scrolling="no"><a href="https://medium.com/media/a12fbacbff04c5d1d2eca9cf6622d626/href">https://medium.com/media/a12fbacbff04c5d1d2eca9cf6622d626/href</a></iframe><h3>Protocols</h3><p>Developers try to create exchange protocols which help make decentralized exchanges follow the uniform rules.</p><p><strong>0x protocol</strong></p><p>0x protocol is an off-chain exchange protocol. One of its main features is the usage of the liquidity pool common for all exchanges working with this protocol.</p><p>0x protocol exchanges use standard API due to which every exchange can get the access to other exchanges’ order books and display their orders in the interface. So, the contract specifies the address of the related exchange for every created order. Therefore, it doesn’t matter which interface was used to close the order,</p><p>only the first exchange will get a commission for it.</p><p>Generally, a flowchart with the liquidity pool can be described as following:</p><ol><li>A seller makes an order through the interface of exchange 1.</li><li>Exchange 1 through the standard API shares the order data with other exchanges.</li><li>A buyer receives the order information through the interface of exchange 2.</li><li>The seller sends a transaction with the trade method call to DEX smart contract</li><li>The smart contract checks the order validity</li><li>If everything goes well, DEX contract calls proxy contract which exchanges tokens between a buyer and a seller.</li></ol><figure><img alt="" src="https://cdn-images-1.medium.com/max/602/0*dQOSsJU63-cxlMi0" /><figcaption><em>Figure 3. 0x protocol trading principle</em></figcaption></figure><p>Another feature of the protocol is that user tokens are not stored in the exchange contract, they are moved between buyer and seller’s addresses through proxy contract.</p><p><strong>Bancor</strong></p><p>Bancor works like an exchange where users can buy and sell tokens, but the principles it uses are completely different.</p><p>The main feature of this protocol is that it uses smart tokens, it means, these tokens’ contracts can be directly exchanged with other tokens without any sellers. In this regard, every smart token has a connector — a contract which keeps the balances of other tokens. The process of buying using the connector looks like that:</p><ol><li>Buying a smart token, a user sends an attached token to the connector balance.</li><li>The connector calls the issue method</li><li>The smart token method issues a certain number of new tokens which are sent to the user.</li><li>Apart from sending new tokens to a user, the issue method also increases the total supply.</li></ol><iframe src="" width="0" height="0" frameborder="0" scrolling="no"><a href="https://medium.com/media/542d338913ce590f88d7e599d26728e1/href">https://medium.com/media/542d338913ce590f88d7e599d26728e1/href</a></iframe><figure><img alt="" src="https://cdn-images-1.medium.com/max/602/0*ZN6grpqLlMaDTyYg" /><figcaption><em>Figure 4. Bancor Network trading principle</em></figcaption></figure><p>And vice versa, when a smart token is sold, it is removed from the user balance with the destroy method call and also this amount is subtracted from the total supply.</p><iframe src="" width="0" height="0" frameborder="0" scrolling="no"><a href="https://medium.com/media/0a19fcb095790fdfcbdba752e457aa02/href">https://medium.com/media/0a19fcb095790fdfcbdba752e457aa02/href</a></iframe><p>Dynamically changing total supply lets a token be always available for buying.</p><p>Smart token technology can be used with any ERC20 tokens, for that Bancor Network provides Relay Tokens — smart contracts that store virtual user balances.</p><p>A full list of DEX protocols is available here<a href="https://github.com/evbots/dex-protocols"> github.com/evbots/dex-protocols</a>.</p><p>At the moment we are finishing the work on indexing trading of the biggest DEXs. These data and analytics to them will be available at <a href="https://santiment.net/">santiment.net</a>.</p><p>In the next article we are going to describe the history and the current state of trading on DEXs.</p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=def21b8b4706" width="1" height="1" alt=""><hr><p><a href="https://medium.com/cyber-drop/what-decentralized-exchanges-are-and-how-they-work-def21b8b4706">What decentralized exchanges are and how they work</a> was originally published in <a href="https://medium.com/cyber-drop">cyber•Drop</a> on Medium, where people are continuing the conversation by highlighting and responding to this story.</p>]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[Ethereum Miners Analysis over Time]]></title>
            <link>https://medium.com/cyber-drop/ethereum-miners-analysis-over-time-7dbe32b328fe?source=rss----fb4b117b4346---4</link>
            <guid isPermaLink="false">https://medium.com/p/7dbe32b328fe</guid>
            <category><![CDATA[data-science]]></category>
            <category><![CDATA[data-analysis]]></category>
            <category><![CDATA[mining]]></category>
            <category><![CDATA[ethereum]]></category>
            <category><![CDATA[blockchain]]></category>
            <dc:creator><![CDATA[cyber•Drop]]></dc:creator>
            <pubDate>Fri, 30 Nov 2018 05:59:14 GMT</pubDate>
            <atom:updated>2018-12-01T04:39:21.257Z</atom:updated>
            <content:encoded><![CDATA[<p><a href="https://github.com/Snedashkovsky"><em>Serge Nedashkovsky</em></a><em>, </em><a href="https://github.com/belya"><em>Anatoli Belchikov</em></a><em>, </em><a href="https://github.com/ninjascant"><em>Maxim Odegov</em></a></p><p><em>Our aim is to see how Ethereum network mining has been changing over time, analyze the activity of mining pools and real miners and their influence on the market.</em></p><p>The analyzed period is up to block <strong>6,781,512</strong> (27 November 2018). All figures and conclusions are given for this period, if otherwise did not state.</p><p>A <strong>mining pool</strong> stands for an address, which gets a reward from the system.</p><p>A <strong>miner</strong> is a person who performs the mining and gets a reward from a mining pool.</p><p>We set ourselves the task of defining the change over time of incoming and outgoing transactions, as well as mining pools and miners’ balances.</p><h3>Mining pools</h3><p>Since the Ethereum network was launched, mining pools have got a reward of <strong>31M ETH</strong> and <strong>0.4M ETH</strong> of transaction fees. The number of addresses which have got a reward is <strong>4.7К</strong>. There are 4.5K block authors among them. The process, depicted these changes over time, is shown below.</p><h3>High reward inequality</h3><p>Reward distribution over mining pools goes with large-scale inequality. For instance, top 3 mining pools started to get more than half of the reward after a month of the launch.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*JOfHtdgUMCUVS5ExB53ihQ.png" /><figcaption>Figure 1. Distribution between Mining Pools, Change over Time</figcaption></figure><h3>Incoming and outgoing transactions</h3><p>Let’s have a look at the flow of mining pools’ funds.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*IMYLmwD7tUmCAYNUd9fSdA.png" /><figcaption>Figure 2. Mining pools Transactions Value, Change over Time</figcaption></figure><p>As you can see, the main incoming flow is formed due to mining rewards (31M ETH).Transaction fees (0.4M ETH) and other incoming transactions (4M ETH) are worth noticing as well. The great influence of other incoming transactions is caused by operations with mining pools’ cold wallets.</p><p>The amount of incoming transactions equals 35M ETH. Interestingly, outgoing transactions mirror the incoming ones.</p><h3>Balances change over time</h3><p>Mining pools’ balances have been decreasing over time and at present amount to 0.2M ETH.</p><p>The chart only shows the balances of reward-receiving addresses. Mining pools control a much larger amount of funds kept in cold and hot wallets.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*WNjD8qTbFxmOD7iNvgVXvw.png" /><figcaption>Figure 3. Mining pools Balances, Change over Time</figcaption></figure><p>Mining pools’ balances have been strongly influenced by ETH market value increase and by Ethereum network key events.</p><h3>Classification</h3><p>Having analysed mining pools’ transactions, we found out they are divided into two groups depending on the average value of an outgoing transaction.</p><ol><li>Less than <strong>10 ETH — 82%</strong> of total reward volume and <strong>91%</strong> in the last year;</li><li>More than <strong>100 ETH — 11%</strong> of total reward volume and <strong>8%</strong> in the last year</li></ol><p><strong>We made a decision to analyze the first group, so all the data are related to it.</strong></p><p>There are two typical types showing how mining pools work:</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*MdbFA2cxQkF5vy5dguOoog.png" /></figure><p>In our analysis we decided not to take into account the reward received by mining pool owners which amounts up to 3% of total reward.</p><h3>Miners</h3><p>In our work we defined the transactions common for miners and broke them down into three categories depending on the following attributes:</p><ul><li>A miner is not a mining pool</li><li>A miner doesn’t send funds to a mining pool</li><li>Mining pools’ incoming funds take 80% of the total volume of incoming miner’s transactions.</li></ul><h3>Incoming and outgoing transactions</h3><p>The amount of incoming miners’ transactions which belong to the first type equals 20M ETH, the amount of outgoing transactions equals 19M ETH.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*POfv4-V8zcBsXMIiFkoNIA.png" /><figcaption>Figure 4. Incoming and outgoing transactions of Miners addresses, Change over Time</figcaption></figure><p>Unlike the incoming transactions volume, the outgoing one tends to have strong peaks. The maximum outgoing transactions volume — <strong>125K ETH </strong>— was reached on <strong>19 July 2016</strong>.</p><h3>Balances change over time</h3><p>Current miners’ balances amount to 5% of the total mining reward and, unlike the mining pools’ balances, tend to grow.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*P91B_is5RHofY08PNyQafQ.png" /><figcaption>Figure 5. Miners Balances, Change over Time</figcaption></figure><p>We can see how strongly <strong>DAO hardfork</strong> (20 July 2016) and price changes in 2017 influenced miners’ balances.</p><p>We defined three groups of miners according to their contribution to the common balance:</p><ul><li>Miners whose current balance is less than 0.1 ETH</li><li>Miners whose incoming transactions volume is less than 0.1 ETH</li><li>Other miners</li></ul><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*tfMBpFhmPuBUeQcykJ7EKg.png" /><figcaption>Figure 6. Miners Balances, Change over Time</figcaption></figure><p>Interestingly, holders account for 2% of all miners’ reward and 40% of their balance.</p><h3>Conclusion</h3><p>Mining pools and miners’ contribution to Ethereum ecosystem can hardly be overestimated. Market movements are closely connected with mining pools and miners’ flows of funds.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*56ymgIPyEIuaONrORPORmA.png" /><figcaption>Figure 7. Mining pools and Miners balances, Change over Time</figcaption></figure><p>Based on the results of this research, new metrics will be integrated into <a href="https://santiment.net/">Santiment platform</a>.</p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=7dbe32b328fe" width="1" height="1" alt=""><hr><p><a href="https://medium.com/cyber-drop/ethereum-miners-analysis-over-time-7dbe32b328fe">Ethereum Miners Analysis over Time</a> was originally published in <a href="https://medium.com/cyber-drop">cyber•Drop</a> on Medium, where people are continuing the conversation by highlighting and responding to this story.</p>]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[How we searched for miners for Robonomics token airdrop]]></title>
            <link>https://medium.com/cyber-drop/how-we-searched-for-miners-for-robonomics-token-airdrop-9b64e5c70f68?source=rss----fb4b117b4346---4</link>
            <guid isPermaLink="false">https://medium.com/p/9b64e5c70f68</guid>
            <category><![CDATA[blockchain]]></category>
            <category><![CDATA[mining]]></category>
            <category><![CDATA[data-science]]></category>
            <category><![CDATA[ethereum]]></category>
            <category><![CDATA[data-analysis]]></category>
            <dc:creator><![CDATA[cyber•Drop]]></dc:creator>
            <pubDate>Fri, 30 Nov 2018 05:50:58 GMT</pubDate>
            <atom:updated>2018-11-29T18:06:26.209Z</atom:updated>
            <content:encoded><![CDATA[<figure><img alt="" src="https://cdn-images-1.medium.com/proxy/1*TkSikk1moDgB6hRBAnHe-A.jpeg" /></figure><p><strong><em>by </em></strong><a href="https://github.com/Snedashkovsky"><strong><em>Sergey Nedashkovsky</em></strong></a></p><p><em>“Since summer Airalab team has launched </em><a href="https://blog.aira.life/developers-diaries-june-july-2018-7dedfb915743"><em>Robonomics network in the Ethereum infrastructure</em></a><em>. By the end of 2018, our team will launch the economy of the XRT token in order to create an incentive to support Robonomics by independent providers. Since it will be necessary for providers to spend gas on the execution of the Robonomics program in the decentralized Ethereum computer (read more in </em><a href="https://robonomics.network/robonomics_white_paper_en.pdf"><em>Robonomics whitepaper</em></a><em>, chapter 6 “Robonomics Token, XRT”), we believe that the first people to offer to become providers of our network are mining pools and miners. But we do not intend to offer “knocking on their doors”, together with the </em><a href="http://cyberdrop.ai/"><strong><em>cyber•Drop</em></strong></a><em> team we decided to distribute up to 1% of XRT tokens, which can be obtained by automatically signing the message with your address, unless, of course, your address is found on the cyber•Drop’s</em><strong><em> </em></strong><em>list :) About the process of getting XRT will be written later in the </em><a href="https://blog.aira.life/"><em>Robonomics blog</em></a><em>. For now, read the cyber•Drop analysis report”, — Sergei Lonshakov, Robonomics network architect.</em></p><p>In the analysis, we considered the period from <strong>4,404,374 to 6,558,806 blocks</strong> (from 2017–10–22 to 2018–10–21). All values and conclusions are given for this period, unless otherwise specified.</p><p>The<strong> mining pool</strong> is an address that receives a reward from the system.</p><p>The <strong>miner</strong> is the person performing mining and receiving reward from the mining pool.</p><p>Data on the addresses of miners and the amount of reward received by them is not directly contained in the chain. We set ourselves the task to identify the most active addresses of miners.</p><p>The distribution of rewards among mining pools is uneven, TOP 20 of mining pools get 95% of rewards:</p><figure><img alt="" src="https://cdn-images-1.medium.com/proxy/1*Nqba17bocYsj4EOXlWO7yA.png" /></figure><p>When analyzing the transactions of the mining pools, we found that they are divided into two classes according to the average size of the outgoing transaction:</p><ol><li>less than <strong>10 ETH</strong> — <strong>91%</strong> from the volume of rewards and <strong>12</strong> from TOP 20</li><li>more than <strong>100 ETH</strong> — <strong>8%</strong> from the volume of rewards and <strong>8</strong> from TOP 20</li></ol><p><strong>We decided to analyze the first group, from now on all the data are given for it.</strong></p><p>The typical structure of the mining pool from the first group is as follows:</p><figure><img alt="" src="https://cdn-images-1.medium.com/proxy/0*dZ4FfIrmuWsWOa-L" /></figure><p>The problem is that other transactions are also present:</p><ul><li>Operations with cold wallets of pools;</li><li>Operations with exchanges.</li></ul><p>In our work, we tried to isolate only transactions specific to miners.</p><p>We analyzed <strong>756</strong> mining pools, <strong>1M</strong> addresses and more than <strong>30M</strong> transactions.</p><p>The results we got:</p><ol><li>Throughout all of the blockchain:</li></ol><ul><li>Rewards — <strong>7.8M ETH <br></strong>(<strong>7.5M ETH </strong>static and uncle rewards, <strong>0.3M ETH </strong>fees)</li></ul><p>2. Throughout the first group:</p><ul><li>Rewards — <strong>7.1M ETH</strong> or <strong>91%</strong> from general volume of the reward;</li><li>Incoming transactions — <strong>183K ETH</strong>;</li><li>Outgoing transactions — <strong>7.2M ETH.</strong></li></ul><p>When analyzing the incoming transactions of the first group, it was found that from <strong>183K ETH</strong>:</p><ul><li><strong>113K ETH</strong> transferred by addresses that received funds from mining pools;</li><li><strong>40K ETH</strong> received from other mining pools;</li><li><strong>30K ETH</strong> other.</li></ul><p>We took into account these revenues and did not consider them in determining the amounts transferred to miners.</p><p>Among the addresses that received funds from mining pools, we left addresses with a share of incoming funds from mining pools more than <strong>50%</strong>.</p><p>After this cleaning, the volume of transactions amounted to <strong>6.6M ETH </strong>or <strong>85%</strong> of the volume of rewards for mining.</p><p>The number of miner addresses was <strong>708K</strong>.</p><p>For the purpose of airdrop, it is desirable to identify addresses that have a long lifetime, since otherwise it is likely to send funds to an address that no one else will use and track.</p><p>By the lifetime of an address, we mean the difference between the time of the first and the last transaction.</p><figure><img alt="" src="https://cdn-images-1.medium.com/proxy/1*a0kEvc4CviQk8e510093yg.png" /></figure><p>The amount of reward received by addresses with a lifetime of up to one month is 2.9%. However, the number of these addresses is 26%.</p><figure><img alt="" src="https://cdn-images-1.medium.com/proxy/0*3py5xbCUm-6INxmU" /></figure><p>The addresses with the last transaction made less than 1 month ago account for <strong>64%</strong> of the total reward received, and they constitute <strong>32%</strong> of the number of miner addresses.</p><p>The number of addresses with a lifetime of more than 1 month and with the date of the last transaction less than 1 month ago is <strong>209K</strong> addresses and corresponds to <strong>4M ETH</strong> of reward.</p><p>For the purpose of airdrop, it is advisable to avoid sending tokens to addresses that are smart contracts. Otherwise, there is a possibility that the recipient of the tokens will not be able to use them.</p><p>The number of <strong>wallets</strong> was <strong>836K</strong>, the volume of funds sent from mining pools was<strong> 6.7M ETH</strong> or<strong> 93%</strong>.</p><p>The number of <strong>smart contracts</strong> was <strong>59K</strong>, the volume of funds sent from the mining pools was <strong>0.5M ETH</strong> or <strong>7%</strong>.</p><figure><img alt="" src="https://cdn-images-1.medium.com/proxy/1*-Y1Aq9ugQXbDkI6C76fKUw.png" /></figure><p>All mining pools and selected miner addresses were included in the airdrop list.</p><p>We selected the addresses of miners based on the following conditions:</p><ul><li>Miner reward from mining pools with an average outgoing transaction value of less than 10 ETH was considered.</li><li>The amount of the miner’s reward is reduced by the amount of transactions sent by the miner to the mining pool.</li><li>The share of incoming transactions from mining pools must be at least 50% of all incoming transactions.</li><li>The difference between the time of the first and the last transaction must be at least 30 days.</li><li>The last transaction must be made later than 30 days before the end of the period in question (later than 09/21/2018).</li><li>The address has to be a wallet.</li></ul><p>Conditions are met by <strong>226K</strong> addresses of miners who received <strong>3.9M ETH</strong> rewards.</p><p>The list of addresses and the number of tokens you can see <a href="https://ipfs.io/ipfs/QmXrxGXfNnsZvdB3WvNWkWAkLoPx5ANQkFMzSqTS13hCHb"><strong>here</strong></a><strong>.</strong></p><p><em>Originally published at </em><a href="https://blog.aira.life/how-we-searched-for-miners-for-robonomics-token-airdrop-4271b25e2d6a"><em>blog.aira.life</em></a><em> on November 1, 2018.</em></p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=9b64e5c70f68" width="1" height="1" alt=""><hr><p><a href="https://medium.com/cyber-drop/how-we-searched-for-miners-for-robonomics-token-airdrop-9b64e5c70f68">How we searched for miners for Robonomics token airdrop</a> was originally published in <a href="https://medium.com/cyber-drop">cyber•Drop</a> on Medium, where people are continuing the conversation by highlighting and responding to this story.</p>]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[DApp Transactions Analysis: Etheroll suspicious gambler]]></title>
            <link>https://medium.com/cyber-drop/dapp-transactions-analysis-etheroll-suspicious-gambler-f37a7263da5?source=rss----fb4b117b4346---4</link>
            <guid isPermaLink="false">https://medium.com/p/f37a7263da5</guid>
            <category><![CDATA[scam]]></category>
            <category><![CDATA[casino]]></category>
            <category><![CDATA[blockchain]]></category>
            <category><![CDATA[data-analysis]]></category>
            <category><![CDATA[ethereum]]></category>
            <dc:creator><![CDATA[cyber•Drop]]></dc:creator>
            <pubDate>Fri, 30 Nov 2018 05:50:16 GMT</pubDate>
            <atom:updated>2018-10-29T10:53:48.440Z</atom:updated>
            <content:encoded><![CDATA[<p><em>by </em><a href="https://github.com/ninjascant"><em>Maxim Odegov</em></a><em>, </em><a href="https://github.com/belya"><em>Anatoli Belchikov</em></a><em>, </em><a href="https://github.com/Snedashkovsky"><em>Serge Nedashkovsky</em></a></p><p>Recently we have <a href="https://medium.com/@cyber_drop/dapp-transactions-analysis-dice2win-case-study-cbbf0f29f2f0">analyzed players’ transactions on dice2win casino</a> and found a system of interconnected addresses that have placed more than half of all bets. Having found this mechanism, we decided to check if the same cases exist in other DApps related to bets. Therefore, we analyzed the interconnections between <a href="https://etheroll.com/">Etheroll</a> players in the same way.</p><h3>Game mechanics</h3><p>Here are the steps how Etheroll bets are placed:</p><ol><li>A user calls the playerRollDice method of Etheroll contract with the parameter rollUnder. This parameter is used in the contract to define the winning probability and the size of the prize.</li><li>From the method rollUnder the contract calls Oraclize API to get a random number from random.org.</li><li>The result, received from Oraclize, is processed with the method __callback, which checks if the bet has won. If it wins, the contract sends a reward to the player address, otherwise the player gets 1 Wei.</li></ol><p>As you can see from the description, the main feature that distinguishes Etheroll from other services is the usage of random.org through Oraclize, as it guarantees the randomness of numbers used to define the outcome of the bet.</p><h3>General statistics</h3><p>In Etheroll 872 players have placed bets, the number of bets equals 75,473. An average bet is 1.15 ETH, the total amount of bets is 86,936 ETH. Etheroll turnover displayed in figure 1.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/923/1*G0rRTxdz5WjdVt-0qgPjjg.png" /><figcaption><em>Figure 1. Etheroll overall amount of bets over time</em></figcaption></figure><p>The total amount of bets by dates 92 out of 872 players have placed bets over 100 ETH each. The amount of these bets equals 92.1% (80,074 ETH). Figure 2 represents the proportions between different metrics for top addresses and for rest of them.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/980/1*ReeYm_MJYj3wMP_KAySUkQ.png" /><figcaption>Figure 2. Dice2Win overall bet volume and amount for players from top players and the rest of them</figcaption></figure><h3>Suspicious addresses</h3><p>Analyzing the graph on incoming players’ transactions we have found the same case as in dice2win. We have detected a large connected component consisting of 354 addresses, 126 of which are Etheroll players. You can see the visualization of this graph in figure 3.</p><p>Interestingly, the graph shows that Ethereum senders in Etheroll have more receivers than dice2win ones.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*wjPcaICN2iT05qYF6PTkgA.png" /></figure><figure><img alt="" src="https://cdn-images-1.medium.com/max/517/1*dobu9M-nus4lJ3iv-Qz4nA.png" /><figcaption><em>Figure 3. Suspicious players and their ties in relation with the rest of casino players’ interconnections</em></figcaption></figure><p>The amount of bets made by the addresses of this group is smaller than in dice2win where the amount of bets made by a suspicious group reached more than half of the total bet (52.1%), but it is still significant and equals <strong>36.9% (32,065 ETH)</strong>. The number of bets was also low and equals <strong>40.6% (30,629)</strong>. These proportions represented in figure 4. Figures 5 and 6 shows how the amount and number of bets are distributed over time.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/950/0*VD2z4x8zvso_wnGV" /><figcaption><em>Figure 4. Etheroll overall volume and number of bets for common and suspicious players</em></figcaption></figure><figure><img alt="" src="https://cdn-images-1.medium.com/max/923/1*koiGDITa9x2ppWuOeq5cjA.png" /><figcaption><em>Figure 5. Amount of bets for common and suspicious players over time</em></figcaption></figure><figure><img alt="" src="https://cdn-images-1.medium.com/max/923/1*Vgz_u1uS8UWPRYbYFQin1g.png" /><figcaption><em>Figure 6. Number of bets for common and suspicious players over time</em></figcaption></figure><p>It also worth noting that the players included into the group of suspicious addresses take the first two places in the list of top 10 players (see table 1). The amount of bets placed by these players is more than half amount of total bets placed by suspicious players.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1011/1*3h5Ox71SzSKb1xxi81owCQ.png" /><figcaption><em>Table 1. TOP-10 players and their participation in rigged turnover</em></figcaption></figure><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=f37a7263da5" width="1" height="1" alt=""><hr><p><a href="https://medium.com/cyber-drop/dapp-transactions-analysis-etheroll-suspicious-gambler-f37a7263da5">DApp Transactions Analysis: Etheroll suspicious gambler</a> was originally published in <a href="https://medium.com/cyber-drop">cyber•Drop</a> on Medium, where people are continuing the conversation by highlighting and responding to this story.</p>]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[DApp Transactions Analysis: Dice2Win Case Study]]></title>
            <link>https://medium.com/cyber-drop/dapp-transactions-analysis-dice2win-case-study-cbbf0f29f2f0?source=rss----fb4b117b4346---4</link>
            <guid isPermaLink="false">https://medium.com/p/cbbf0f29f2f0</guid>
            <category><![CDATA[dapps]]></category>
            <category><![CDATA[scam]]></category>
            <category><![CDATA[ethereum]]></category>
            <category><![CDATA[blockchain]]></category>
            <category><![CDATA[data-science]]></category>
            <dc:creator><![CDATA[cyber•Drop]]></dc:creator>
            <pubDate>Fri, 30 Nov 2018 05:48:19 GMT</pubDate>
            <atom:updated>2018-11-29T18:07:49.672Z</atom:updated>
            <content:encoded><![CDATA[<p><em>by </em><a href="https://github.com/ninjascant"><em>Maxim Odegov</em></a><em>, </em><a href="https://github.com/belya"><em>Anatoli Belchikov</em></a><em>, </em><a href="https://github.com/Snedashkovsky"><em>Serge Nedashkovsky</em></a></p><p>Blockchains have introduced unprecedented transparency of economic agents activity that helps analyse this activity from different perspectives. <a href="https://blog.cyber.fund/huge-ethereum-mixer-6cf98680ee6c">One</a> of our previous articles shows how one system influences the metrics of the whole Ethereum network.</p><p>The structures containing a relatively low number of tightly connected addresses, that have a great influence in the network, also exist in some DApps. The objective of this article is to check the presence of such structures in <a href="https://dice2.win/">Dice2Win</a> that might mean the casino uses the rigged turnover.</p><h3>Game mechanics</h3><p>The general bet mechanics in Dice2Win is the following:</p><ol><li>A user sends a transaction to casino contract by calling the placeBet method. In the parameters of the call the user selects a number of sides on a simulated dice — modulo, selected sides for the bet — betMask, and commit — the number, provided by Dice2Win and used at a later stage to define the outcome of the bet.</li><li>A croupier bot sends a transaction by calling the settleBet method. This method combines the number-commit with the block hash related to the bet. Based on this combination the keccak256-hash is calculated and with its help the outcome of the bet is defined. If the bet wins, the player gets a reward, otherwise he receives 1 Wei as a consolation prize.</li></ol><p>At present Dice2Win offers the following bets:</p><ul><li>coin flip: in this case modulo equals 2;</li><li>roll a dice: modulo equals 6;</li><li>two dice: modulo equals 36</li><li>etheroll: modulo equals 100</li></ul><p>The amount of Ethereum used in bets should be over 0.05 ETH.</p><h3>General statistics</h3><p>Dice2Win contract, which has been changed five times, is the main contract of the service. In parallel there was at least one contract with the same interface that co-existed with the main one. Figure 1 illustrates the period of activity for each contract version. In our research we used only the actual version of this contract with the address <a href="https://etherscan.io/address/0xd1ceeeeee83f8bcf3bedad437202b6154e9f5405">0xd1ceeeeee83f8bcf3bedad437202b6154e9f5405</a>.</p><figure><img alt="" src="https://cdn-images-1.medium.com/proxy/0*EGrn2x4lbjKD_ijR.png" /><figcaption><em>Figure 1. Number of bets over time for Dice2Win contract versions</em></figcaption></figure><p>In total 817 players have placed their bets, the total number of bets is 93,380. An average bet equals 0.26 ETH, the total amount of all bets equals 24,607.4 ETH. Dice2Win turnover displayed in figure 2.</p><figure><img alt="" src="https://cdn-images-1.medium.com/proxy/0*84sew9KRtDQ1o_D8.png" /><figcaption><em>Figure 2. Dice2Win overall amount of bets over time.</em></figcaption></figure><p>The most popular bet is coin flip: this type of game is leading both by the number of transactions and by the number of bets. Figure 3 represents the popularity of each game type.</p><figure><img alt="" src="https://cdn-images-1.medium.com/proxy/0*wBaqpP-iuD6FsA_0.png" /><figcaption><em>Figure 3. Dice2Win overall bet volume and amount for each game type.</em></figcaption></figure><p>48 out of 817 players have places the bets in the amount of more than 100 ETH. The volume of bets placed by these players equals 83.7% (28,317 ETH), whereas their amount is only 5.9% (48 addresses) of the total number. Figure 4 represents the proportions between different metrics for top addresses and for rest of them.</p><figure><img alt="" src="https://cdn-images-1.medium.com/proxy/0*Ryv1hb8KUWIu8cBf.png" /><figcaption><em>Figure 4. Dice2Win overall bet volume and amount for players from top players and the rest of them</em></figcaption></figure><h3>Suspicious addresses</h3><p>Taking a closer look at top-addresses statistics we could define a group of wallets which have a relatively low number of Ethereum recipients and senders:</p><ul><li>0x292eb4da428d13a2b13ea9257a58781f9037de65 — 4 unique recipients and senders;</li><li>0x34241da47768670cf372da49849c0bb86e6b3aae — 2 unique recipients and 3 senders;</li><li>0x768052831f753f7aa532b0d8751d5a5c22377681 — 3 unique recipients and 5 senders;</li><li>0x355e2479ca4b494296b4d3c08e5c3c57cffd0dd1 – 6 unique recipients and 4 senders;</li></ul><p>Interestingly, the last address, unlike the first three ones, interacted more often with usual wallets than with stock markets. Having analyzed its interactions, we came to the conclusion that among its recipients and senders there was a large number of addresses, which also placed bets in Dice2Win. To define the scale of these interactions and their influence on the economy of the service we have analyzed the graph of transactions between the addresses, which had sent the bets in Dice2Win.</p><p>In order to show how the service creates rigged transactions, let’s take a look at the top player’s wallet 0xda2101021a7937e4c12036205f16694ef1e7a3c0, as its turnover seemed suspicious to us. Figure 5 shows the change of its balance. As you can see from the chart, its profit from gambling is negative, but the balance stays positive for a long time. Further research shows there is a group of addresses that charge up the wallet’s balance each time it falls under 8 ETH. It is worth noting that the address returns Ethereum when its balance is high enough to place bets (supposedly, it happens when it overcomes the barrier of 45 ETH).</p><figure><img alt="" src="https://cdn-images-1.medium.com/proxy/0*fUf-4eM6Dc-BYaQg.png" /><figcaption><em>Figure 5. Balance of the suspicious player over time</em></figcaption></figure><p>Taking a closer look at the environment of the address, it is possible to note that 54 ETH was received from 0xdf95de30cdff4381b69f9e4fa8dddce31a0128df. It is indicative that apart from this transaction this address sent Ethereum to other top players, for instance, to 0x17d8e86af16cd6a5fdd5eb04ace17a4461151781 or 0x40584ddb22779dc7280cd09deecca4771f785542. With its help, 28 dice2win players got 388.5 ETH in total, however no players sent transactions in response.</p><h3>Graph analysis: transactions between players</h3><p>To find tightly connected groups of addresses, we did the following:</p><ol><li>Collected a list of addresses that sent Ethereum to Dice2Win players. A player is anyone who calls the sendBet method of Dice2Win contract.</li><li>Removed smart contracts and named etherscan.io addresses from the dataset.</li><li>Built a graph with players and senders of Ethereum represented as nodes. The nodes are connected with an edge if there is at least one transaction between them.</li></ol><p>As a result, we can detect 234 players, that have received Ethereum from the addresses that are not contracts and don’t have names on etherscan.io. If we show their interconnections, we will get a graph presented in <a href="https://image.ibb.co/d1WQ0q/graph-susp.png">figure 6</a>.</p><figure><img alt="" src="https://cdn-images-1.medium.com/proxy/0*GXg1830gJ9aoPjF6.png" /></figure><figure><img alt="" src="https://cdn-images-1.medium.com/proxy/0*IE97ZhCe6lz9jIx0.png" /><figcaption><em>Figure 6. Suspicious players and their ties in relation with the rest of casino players’ interconnections</em></figcaption></figure><p>As we can see, most of addresses are situated within the biggest connected component. It contains 406 addresses, 24 addresses are outside of this group. The number of players in the group equals 215. Further analysis showed that the volume of all bets in the main component equals 52.1% of the total, and the number of bets is 52.4% of the total, these proportions represented in figure 7. The changes of the numbers of bets and their volume over time within the main component and in total are displayed in figures 8 and 9.</p><figure><img alt="" src="https://cdn-images-1.medium.com/proxy/0*aO5XnT5oGgBfZ-Bq.png" /><figcaption><em>Figure 7. Dice2Win overall volume and number of bets for common and suspicious players</em></figcaption></figure><figure><img alt="" src="https://cdn-images-1.medium.com/proxy/0*ahvxmjVnRsQPAgjK.png" /><figcaption><em>Figure 8. Amount of bets for common and suspicious players over time</em></figcaption></figure><figure><img alt="" src="https://cdn-images-1.medium.com/proxy/0*U21Q_Sb_fhVVUiIA.png" /><figcaption><em>Figure 9. Number of bets for common and suspicious players over time</em></figcaption></figure><h3>Conclusion</h3><p>The presence of an influential suspicious agent in Dice2Win makes it possible to suggest about the rigged number of contract transactions. According to the research, more than a half of top casino players participate in the suspicious turnover (see table 1), and its volume is more than 50% of total Dice2Win turnover.</p><p>We haven’t found a fixed address which charges up the players’ balance when they are lack of Ethereum. However it’s possible that a group of wallets created with the same purpose might exist. One of these wallets has been found and described in the research.</p><figure><img alt="" src="https://cdn-images-1.medium.com/proxy/0*jWg4otdrfxY6lCp4.png" /><figcaption><em>Table 1. TOP-10 players and their participation in rigged turnover</em></figcaption></figure><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=cbbf0f29f2f0" width="1" height="1" alt=""><hr><p><a href="https://medium.com/cyber-drop/dapp-transactions-analysis-dice2win-case-study-cbbf0f29f2f0">DApp Transactions Analysis: Dice2Win Case Study</a> was originally published in <a href="https://medium.com/cyber-drop">cyber•Drop</a> on Medium, where people are continuing the conversation by highlighting and responding to this story.</p>]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[cyber•Drop — Smart Solution for Your airdrop]]></title>
            <link>https://medium.com/cyber-drop/cyber-drop-smart-solution-for-your-airdrop-580402711c6a?source=rss----fb4b117b4346---4</link>
            <guid isPermaLink="false">https://medium.com/p/580402711c6a</guid>
            <category><![CDATA[blockchain]]></category>
            <dc:creator><![CDATA[cyber•Drop]]></dc:creator>
            <pubDate>Fri, 30 Nov 2018 05:47:06 GMT</pubDate>
            <atom:updated>2018-08-03T17:15:48.038Z</atom:updated>
            <content:encoded><![CDATA[<figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*4OXONQnnTNBYFsiwPl_u-Q.jpeg" /></figure><p>Service for Your Airdrop campaign is available now!</p><h3>Who we are?</h3><p>We are the interplanetary team of developers and researchers in the field of blockchain and data science.<br>Our team unites humans, blockchains and other thinking creations.<br>The Ethereum blockchain is one of our team member. We communicate with it through full node.<br>Another member of our team is a Smart Contract. He works with us for the Gas.<br>We are consider ourselves a part of the <strong>Worldwide Blockchain Community</strong>, and share the ideas of decentralization, transparency and privacy.</p><h3>What did we do?</h3><p>We spent one year to find an effective solutions for working with data from the Ethereum blockchain.</p><h3>What did we get?</h3><p>As a result of our research, we created a set of services that can “read” the Ethereum blockchain.</p><h3>What we can give?</h3><p>cyber•Drop Services can get any data from the Ethereum blockchain:</p><ul><li>Addresses-list of holders ETH and ERC20-tokens</li><li>Mapping addresses and balances</li><li>Activity of addresses</li><li>Profitability of addresses</li><li>Investments of addresses</li><li>Methods of calling smart contracts</li></ul><h3>How you can use it?</h3><p>A capabilities of <strong>cyber•Drop Services</strong> are truly great!</p><ul><li>Launch an airdrop campaign by the ready-made parameters: we’ve found for you the best combination of indicators and amounts of addresses</li><li>Customize your own package for an airdrop by the individual parameters: choose your target group, model of distribution and quantity of addresses</li><li>Distribute your tokens after an ICO or Bounty campaign: upload ready address list to filter it, or to count the amounts of tokens for distribution</li><li>Calculate effectiveness of the airdrop campaign: our Services will give you expert review in any convenient form</li><li>Get raw data in a understandable form: our Services are perfect for researching and analyzing processes inside of the Ethereum blockchain</li></ul><h3>What to do now?</h3><ul><li>Visit our website: <a href="https://cyberdrop.ai/">https://cyberdrop.ai</a></li><li>Download our OnePage</li><li>Join our community in Telegram: <a href="https://t.me/cyberDrop_community">https://t.me/cyberDrop_community</a></li><li>Visit our github: <a href="https://github.com/cyber-drop">https://github.com/cyber-drop</a></li></ul><h3>Need help?</h3><p>Our technical support ready to help you 24/7:<br>Telegram — <a href="https://t.me/cyberdrop_support">https://t.me/cyberdrop_support</a><br>E-mail — support@cyberdrop.ai</p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=580402711c6a" width="1" height="1" alt=""><hr><p><a href="https://medium.com/cyber-drop/cyber-drop-smart-solution-for-your-airdrop-580402711c6a">cyber•Drop — Smart Solution for Your airdrop</a> was originally published in <a href="https://medium.com/cyber-drop">cyber•Drop</a> on Medium, where people are continuing the conversation by highlighting and responding to this story.</p>]]></content:encoded>
        </item>
    </channel>
</rss>