<?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[Stories by Sukrit Singh on Medium]]></title>
        <description><![CDATA[Stories by Sukrit Singh on Medium]]></description>
        <link>https://medium.com/@sukritsingh?source=rss-a96bc79c415c------2</link>
        <image>
            <url>https://cdn-images-1.medium.com/fit/c/150/150/0*Ipv-ag2gz9kyWxCm.</url>
            <title>Stories by Sukrit Singh on Medium</title>
            <link>https://medium.com/@sukritsingh?source=rss-a96bc79c415c------2</link>
        </image>
        <generator>Medium</generator>
        <lastBuildDate>Fri, 20 Nov 2026 05:53:30 GMT</lastBuildDate>
        <atom:link href="https://medium.com/@sukritsingh/feed" rel="self" type="application/rss+xml"/>
        <webMaster><![CDATA[yourfriends@medium.com]]></webMaster>
        <atom:link href="http://medium.superfeedr.com/" rel="hub"/>
        <item>
            <title><![CDATA[Rail Trail]]></title>
            <link>https://medium.com/@sukritsingh/rail-trail-a9aebb83a986?source=rss-a96bc79c415c------2</link>
            <guid isPermaLink="false">https://medium.com/p/a9aebb83a986</guid>
            <category><![CDATA[travel]]></category>
            <category><![CDATA[optimization]]></category>
            <category><![CDATA[india]]></category>
            <category><![CDATA[indian-railways]]></category>
            <category><![CDATA[railways]]></category>
            <dc:creator><![CDATA[Sukrit Singh]]></dc:creator>
            <pubDate>Wed, 21 Nov 2018 21:25:59 GMT</pubDate>
            <atom:updated>2019-01-12T17:12:16.866Z</atom:updated>
            <content:encoded><![CDATA[<figure><img alt="" src="https://cdn-images-1.medium.com/max/467/0*ki46d2yn3iaYxkhn" /></figure><h3>Problem</h3><p>Indian Railways witnessed one of its worst performance in terms of punctuality in 2017–18. Around 30% of the trains were late during this period, which is an extremely disturbing figure considering that trains which are late for up to 15 minutes are graded as “on-time”. But a more disturbing figuring is that of average Train Delay Index, which is as high as 35 minutes. This means that every passenger has to wait for an average of 35 minutes for a train beyond its scheduled arrival time.</p><p>So using a little approximation,</p><p>Average waiting time per passenger: 35 minutes</p><p>Number of passengers per day: 23 million</p><p>GDP per capita : 1939 USD/annum</p><p>Working hours per day: 8</p><p>Working days per year: 261</p><p>The loss in GDP due to train delays = 1939*23 million*35*365/(60 * 8 * 261) = <strong>4.5 billion USD</strong></p><p>There can be several reasons for train delays. Some of them are –</p><ul><li>Inadequate platform/train infrastructure (saturation of network)</li><li>Engineering works</li><li>Signal failures</li><li>Mismanagement/ passenger agitations</li><li>External factors eg: weather/accidents/chain pulling</li></ul><p>Let’s take a closer look at the points mentioned above. It can be seen that except the first, others are temporary in nature. Signal failures, accidents, mismanagement, engineering works happen at a micro level and although they can affect a small portion of the vast railway network, they are definitely not strong enough reasons to result in around 30% of the trains running behind schedule throughout the year. But if the first point is the main problem, shouldn’t the railway timetable already take that into account?</p><h3>Back to basics</h3><p>Let’s try to build the railway network from scratch assuming that all trains run with the same speed and the stations are equidistant, and see whether our railway system would run perfectly if none of the temporary issues crop up.</p><p>Case 1 — Two nodes and a double track</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/827/0*P6JDgcKfG52r_TQZ" /></figure><p>This is the ideal scenario where two adjacent stations are connected via double railway track (assuming well-maintained tracks using latest technology). Two trains starting simultaneously in opposite directions should reach their destinations on time.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/827/0*L2hVMMBss6KVUF_s" /></figure><p>Case 2 — Two independent pairs</p><p>Now let’s imagine that we have another similar pair of stations connected via double track. This too should work perfectly well.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/0*4iXOnlmNNHxZogLP" /></figure><p>This is where the trains will end up:</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/0*9mxH6Zlrxz_ipi9o" /></figure><p>Case 3: Introducing a single track</p><p>Now let’s imagine that there is an increasing demand for the passengers to travel between Station 1 and Station 4. Hence, the ministry decides to extend routes of Train 1 and Train 4 till Station 4 and Station 1 respectively. Due to the limitation of the budget, only a single track is added between Station 2 and Station 3. This is what happens now:</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/0*B_LyTnakj8CAHEUD" /></figure><p>Train 1 starts its journey towards Station 3 while Train 4 waits for Train 1 to free-up the track. Hence, Train 4 gets delayed !!</p><p>A simple solution to fix the delay above is to simply adjust the timings of the trains:</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/0*li7gKvv-RwTEQt7N" /></figure><p>Train 4 starts after Train 1 reaches Station 2:</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/0*GZIjpUBgvIlghluo" /></figure><p>And the journey continues:</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/0*iVmTIr0DJwmN-hrt" /></figure><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/0*6W6EFFGNhnkJJ86D" /></figure><p>Sounds like a simple solution right? There are two very interesting points to note here:</p><ol><li>Although the problem arises due to the unavailability of additional track between Track 2 and Track 3, the fix can be provided without adding an additional track between the stations</li><li>The delay happens because the train is made to wait at Station 3, but the solution is to adjust the timing at Station 4</li></ol><p>So what’s the point that I am trying to make? In a network setup, the source of the solution may not be the same as the source of the problem.</p><p>Moving on, adding more complexity to the problem:</p><p>Case 4: Unavailability of a platform at Station 3</p><p>Let’s assume that Station 3 has one just platform. This means that Train 4 cannot wait at this platform.</p><p>So from here :</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/0*t2s3Jd6mfvvHtf1n" /></figure><p>We go here:</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/0*sbo1tg_qkf9BKYUq" /></figure><p>This shows that even though Train 4 has reached Station 3, it has to wait outside as there are not enough platforms on the station. Therefore, now Train 4 is even delayed at Station 3, where it was reaching on time earlier. Also note how the double line between Station 3 and Station 4 is largely a waste now.</p><p>So what could have been a better solution? To invest in an additional platform on Station 3 instead of a double track between Station 3 and Station 4.</p><p>Another thing to note in the above example is that Train 4 would occupy a platform at Station 3 and the track between Station 2 and Station 3 later than its scheduled time. In an actual scenario, this might delay another lower priority train using the same track and station. Result? <strong>Delay cascading</strong>!!</p><p>Now, if we expand this little example to the entire Indian Railway network, we can extrapolate our findings to see that several delays can be avoided if we adjust train timings and use our resources more effectively.</p><h3>Hypothesis</h3><p><em>The Indian Railway network is not optimized.</em></p><p>I initially came up with this hypothesis when I observed the train timings. It is worth noting that <strong>almost all trains </strong>start at pretty looking times that are multiples of 5. eg: 20:00, 21:35, 10:15, 18:20 etc. Why are there hardly any trains with times like 18:32, 21:01, 10:16 ?</p><p>This makes me wonder how train timings are actually decided? Is it purely based on intuition? Is it demand-based? Is it demand-based + resource availability-based? Does it consider the resource availability on the train route or the entire network?</p><p>To find answers to these questions, and to validate the hypothesis, I decided to simulate the Indian Railway network.</p><h3>The Indian Railway Simulator — Rail Trail</h3><p>The concept of railway simulation is not new. The Open Track project (<a href="http://www.opentrack.ch/mobile/opentrack_e/opentrack_e.html">http://www.opentrack.ch/mobile/opentrack_e/opentrack_e.html</a>) started this implementation in the 1990s. It was/is used to do the following:</p><ul><li>Analyze capacity of existing infrastructure</li><li>Time-table planning</li><li>Evaluate energy consumption</li><li>Track signaling system</li></ul><p>I tried to find a similar system for the Indian railways but I couldn’t, so I decided to create a simple version of my own. This version can be accessed <a href="http://railtrail.ga/">here</a>. To do this, the following data points were collected:</p><ul><li>Railway timetable</li><li>List of all railway stations in India</li><li>Number of platforms in each station</li><li>The route followed by every train</li><li>Type of railway track between any two stations</li></ul><p>After this data was gathered, a simulator was implemented to replicate the movement of every train at its designated time, keeping the constraints into consideration. As this implementation was supposed to be a Proof of Concept (PoC), I decided to ignore factors like signaling, weather, disruptions, accidents. Only the following constraints were considered:</p><ul><li>Number of platforms on each station</li><li>Type of track (Single/Double) between any two stations</li></ul><h3>The Techie side</h3><p>The entire railway network was constructed like a graph, where each station was represented by a node and the path between two nodes was represented by an edge. Hence, each directed pair of nodes was maintained as a separate entry in an SQL database. The JAVA back-end was exposed as REST APIs, which accepted a range of time period as request parameters and the current graph position of all trains in the network. These APIs then returned the new location of all trains, along with the locations of newly started trains during this time period.</p><p>The back-end logic was implemented with the following assumptions:</p><ul><li>The speed of a train between any two nodes remains constant</li><li>A train occupies a platform of the last station for 15 minutes before disappearing</li><li>Two trains cannot exist on the same track between two nodes at the same time</li><li>Trains start running at their designated starting times only after the simulator starts</li></ul><p>With these assumptions, the movement of all trains was simulated. If any train came close to violating a constraint, it was made to wait; <strong>resulting in delays</strong>!!</p><h3>Results and Suggestions</h3><p>The simulator started working as expected, but as more and more trains entered the network, my laptop’s computing power gave up. But even during this period, where I could simulate the movement of trains for 6 hours starting midnight of a Sunday, I could see trains getting delayed. It is worth noting here that not even 500 trains were on the network at that point. This result was extremely surprising especially when it is known that the Indian Railway network can easily have more than 4000 trains running together. Conclusion ? <strong>Delays are inevitable with our current timetable, and the existing railway network is not optimized</strong></p><p>It might be true that our infrastructure is not good enough to have no train delays at all. But is it that bad that around 30% of our trains are late ? Surely not. So what can be done ?</p><ol><li>Revisit our railway timetable and adjust the train timings using full network simulation. I understand that there are certain trains for which train timings cannot be changed significantly, especially the morning and evening trains on which people commute daily. But even a minor tweak can have a large impact on the entire network (as explained in the section above). In fact, it would be ideal if the simulator can also take passenger traffic into account, to effectively balance out loss in passenger traffic vs overall network delay avoided due to any timing change.</li><li>Use simulator to find the next best infrastructure development project. As mentioned earlier, an additional track or platform at a traffic-heavy destination might provide some relief, but it might not be the most effective use of the resource. It might be more effective to use it at some other destination which might have a larger overall impact. This can again be calculated using a simulator to predict the reduction in net delay of trains, and it can be combined with passenger traffic to calculate the net reduction in train delay index.</li></ol><p>It took me a little over two months to build a basic simulator. Enhancing this project using traffic and cost parameters can considerably improve our network.</p><h3>Links and Resources</h3><p>Indian Railway Simulator PoC: <a href="http://railtrail.ga/">http://railtrail.ga/</a></p><p>React front-end : <a href="https://github.com/Sukrit/Railways---UI">https://github.com/Sukrit/Railways—UI</a></p><p>Java Back-end: <a href="https://github.com/Sukrit/Railways-Back-end">https://github.com/Sukrit/Railways-Back-end</a></p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=a9aebb83a986" width="1" height="1" alt="">]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[Cricket : Curious case of batting numbers]]></title>
            <link>https://medium.com/@sukritsingh/cricket-curious-case-of-batting-numbers-c8fe8e44c153?source=rss-a96bc79c415c------2</link>
            <guid isPermaLink="false">https://medium.com/p/c8fe8e44c153</guid>
            <category><![CDATA[cricket]]></category>
            <category><![CDATA[statistics]]></category>
            <category><![CDATA[ms-dhoni]]></category>
            <dc:creator><![CDATA[Sukrit Singh]]></dc:creator>
            <pubDate>Wed, 21 Nov 2018 21:23:27 GMT</pubDate>
            <atom:updated>2018-11-21T21:23:27.079Z</atom:updated>
            <content:encoded><![CDATA[<h3>Cricket : Curious case of batting numbers</h3><h3>Problem</h3><p>Let me jump straight to the point. A player’s effectiveness as a batsman is often represented through his/her batting average and strike rate. By definition, they represent the following:</p><p>Batting Average: Runs scored between every dismissal</p><p>Strike Rate: Runs scored per hundred balls faced</p><p>Batting Average supposedly represents quantity of runs scored, while Strike Rate quantifies the attacking ability of a player. Although both these metrics fairly accurately represent a batsman’s greatness, they fail to show their <strong>effectiveness</strong> as a batsman in the context of the game. Let me illustrate this using an example:</p><p><em>Assuming a team is trying to chase a score of 350 runs in 50 overs (Required Run Rate of 7 runs/over). While chasing, a batsman scores 150 runs off 150 balls and gets out, and eventually the chasing team ends with a score of 330/5 in 50 overs. The team falls short of target by 19 runs.</em></p><p><em>Assuming no extras, the centurion scored at an average of 150 and a strike rate of 100, while others scored at a strike rate of 120. Fairly good numbers. Yet, the team lost the game. So what exactly went wrong ?</em></p><p><em>Two possible explanations:</em></p><ul><li><em>The centurion’s role was to stay at the crease. While he performed his role perfectly, others didn’t</em></li><li><em>The centurion consumed too many balls and other batsmen couldn’t keep up with the required rate</em></li></ul><p>In either case, the record books wouldn’t care much. They would only show a spectacular average of 150 and a strike rate of 100 for that player in that game, and if we extrapolate this scenario to many games, this player would go down as one of the greats of the game, irrespective of the team’s performance. This conclusion seems odd, especially for a team game like cricket, where a team’s think tank would surely revisit the roles of the players if the above scenario happens frequently due to the first explanation. But what if the second one is the more likely explanation ? His average and strike rate are phenomenal, remember ?</p><h3>So what’s a better method ?</h3><p>Batting statistics can become more meaningful if the match context is also taken into account. Let’s consider various scenarios and solve this problem step by step. But before we jump into scenarios, we must break down our problem:</p><p>The effectiveness of a batsman in test matches, which although are more complex by nature, is relatively easier to analyse because the weightage of strike rate falls significantly in most cases. Hence, we should rather restrict ourselves to limited over versions of cricket. For a good balance between average and strike rate analysis, I am going ahead with 50 over ODIs (One Day Internationals) alone.</p><p>As we are trying to take context of a match into account, it is easier to establish the same in the second innings for an ODI. Reason ? The target is known, and it is easier to track progress at every step of the innings. In comparison, it is harder to define a good score while batting first because numerous factors are involved (This discussion can be taken up later). So let’s narrow down our problem statement to only consider Second Innings in ODIs. Also, to remove complexity, let’s ignore rain affected games.</p><p>So here’s how we will proceed, we will consider two options in every scenario and choose a winner. Once we have some basic rules established, we can quantify winning scores later.</p><p>Scenario 1:</p><p>Chasing a score of 300 in 50 overs; which option should fetch more points off the first ball of the innings ?</p><p>Option A : Scores a single</p><p>Option B : Plays a dot ball</p><p>Winner: Option A wins obviously, it reduces target and keeps the scoreboard on track with the run rate.</p><p>Scenario 2:</p><p>Chasing a score of 300 in 50 overs, which option should fetch more points ?</p><p>Option A : First ball dot, second ball single</p><p>Option B : First ball single, second ball single</p><p>Winner: Option B wins. Same logic as Scenario 1, it reduces target and keeps the scoreboard on track with the run rate</p><p>Scenario 3:</p><p>Chasing a score of 300 in 50 overs, which option should fetch more points ?</p><p>Option A : ‘Blaster’ scoring a boundary off the first ball of the innings</p><p>Option B : ‘Finisher’ scoring a boundary off the last ball of the innings (assuming 4 runs were needed)</p><p>Winner: Option B wins. Two reasons:</p><p>– Option B decides the game. It can be argued that Option A contributed equally towards the score, but more was at stake towards the end. Scoring a boundary was optional for ‘Blaster’, but it was mandatory for ‘Finisher’</p><p>– ‘Finisher’, who bats later in the innings should be awarded more points than a ‘Blaster’ for a similar score. ‘Blaster’ has the luxury of facing more balls, getting his eye in, and making up for dot balls, and has the cushion of more batsmen to follow. The same doesn’t apply for a finisher</p><p>Scenario 4:</p><p>Option A : First ball dot in a 300 run chase</p><p>Option B : Last ball dot in a 300 run chase</p><p>Winner: Option A wins, or Option B loses more. The more at stake, the more you lose</p><p>Scenario 5:</p><p>Option A : First ball boundary in a 300 run chase</p><p>Option B : First ball boundary in a 150 run chase</p><p>Winner: Option B wins. The weightage of every run scored depends on the Required Run Rate</p><p>So after going through the possible scenarios, we can establish two facts:</p><ul><li>A run scored later hold more weightage than a run scored earlier</li><li>The weightage of a run depends on the required run rate</li></ul><h3>How to quantify ?</h3><p>The score of a player in every game can be determined by the aggregate of points earned on every ball faced. The points should depend on the two factors mentioned above. Let’s start with the second factor:</p><p>Assuming that the batsman fetches exactly one point for correctly following the RRR, he must score:</p><p>score = r — RR + 1</p><p>where:</p><p>r — run scored</p><p>RR — Instantaneous required run rate</p><p>This score must be multiplied to a factor to grant higher weightage to every run scored later. To achieve this, the maximum possible weightage must be defined (some subjectivity here). It should be fair to say that a last ball run should not hold more than twice the weightage of a run scored earlier. However, this multiplication factor should also normalize the rapid change of RR in the later part of the innings. Hence, this factor should represent an increasing graph at a decreasing rate with a maximum value of 2.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/525/0*B_ruLG29v7JoGQAs" /></figure><p>The above graph of y = sqrt(over/14 + 0.5) addresses the requirements perfectly. Therefore our final score per ball comes out to be:</p><p>Score = sqrt(over/14 + 0.5) * (r — RR +1)</p><p>Subsequently, the aggregated score for a player in a match comes out to be the sum of scores for all balls faced. It is worth noting here that the above formula can fetch negative scores for negative contributions towards a team’s cause.</p><h3>Does it really work ?</h3><p>To check if this formula works, I started working with some ODI data for years 2006 to 2017. Here’s what the aggregate second innings looked like:</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/0*5Hb3o9ux0VUO0_UT" /></figure><p>Note how team aggregate second inning scores vary from around 900 to -3.4k. The median score is 538, which also gives a good indication of what can be considered as a good effectiveness score for an individual. Considering that the top seven batsmen contribute the most towards the effectivity score, a score of 70 for an individual can be considered average.</p><p>But wait, how can a team possibly score -3.4 k ?</p><p>Let’s revisit the primary objective of this article. We want to measure how <strong>effective</strong> a batsman is in the <strong>context of the game</strong>. So if players score heavily with no intentions of winning the game, they should be penalized. Now let’s take a look at the scorecard of Match Id: 1089524</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/0*QkYvncaPc2KxVKBe" /></figure><p>Notice how the Ireland team played out the entire 50 overs with no intention of getting close to the target. Sorry to say, but they deserved to be penalized.</p><p>Now it’s time to jump to individual scores. Why not start with arguably one of the best chasers of the game: M S Dhoni. This is what a snapshot of his career scores looks like:</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/0*cUR4caZpFksd_czv" /></figure><p>The blues represents the scores where Dhoni remained unbeaten. Not surprisingly, there are plenty of 100+ scores during the initial stages of his career. But what happened later ?</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/0*3kRW1LSHIpGvPzRd" /></figure><p>The graph above shows the average effectivity score of M S Dhoni every year. Although, the drop in his effectiveness should come as no surprise to us, but what’s surprising is that his effectiveness has been <strong>below average </strong>as a batsman for the last 6 years. But do his batting numbers really say so? Let’s take a look at this example from WC2015 Semi-Final:</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/0*z0-RJOBv0ebIl-5L" /></figure><p>Chasing 329, India was bowled out for mere 233 runs. No one, except M S Dhoni, put up a fight against the Aussies (as seen from the scorecard above). But in reality,</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/0*GUF61udytq67slbV" /></figure><p>Dhoni’s effectivity score was <strong>negative</strong>. It was because when he walked in, the Required Rate was over 8. But Dhoni started playing his shots only when the rate went past 15, reducing the effectiveness score of even hitting a <strong>six,</strong> significantly. India were in the game only when Rohit and Dhawan were at the crease.</p><p>Here’s another, remember the famous 112m six by Dhoni in Adelaide ? When he walked in to bat, the required rate was less than 6 with around 15 overs to go. But he hit this six when the team needed 12 off 4 balls.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/0*9fwy0CYdtMga-rKV" /></figure><p>The effectiveness score chart recognizes his finishing abilities, but also gives its due credit to Gambhir (who openly criticized Dhoni for leaving the chase for a bit too late).</p><p>And what about Virat Kohli ?</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/0*oSX9y6gyrrOWSeZN" /></figure><p>As seen from the red bars, King Kohli is on the rise and still going higher. But surprise surprise, his effectivity score is not the highest among India players.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/0*J4pSI4oYpkEd3juG" /></figure><p>Ambati Rayudu’s effectiveness during chases outnumbers everyone else comfortably. In the limited number of matches that he has played for India, he has been able to close games successfully, which is aptly reflected in the graph above (Does anyone remember Dhoni denying a single against England in a T20, preventing Rayudu from taking the strike ?) . What is also worth noting here is that Ganguly, Gambhir and Yuvraj have higher chase effectiveness scores than Dhoni. But what about Tendulkar ? Well!!</p><p>Going back to Ambati Rayudu, here’s what his scores looked like till last year:</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/0*Ivip0vATiSLm6DsB" /></figure><p>The blues show the matches where he remained unbeaten. Seeing that five of them fetched him scores over 200, it won’t be unfair to call him a match winner for India in the limited opportunities that he has got. However, the poor quality of attack faced by him has resulted in the lack of recognition for his performances.</p><h3>Conclusion</h3><p>Historically, we have used averages and strike rates to judge the greatness of players and more often than not, these numbers don’t mislead. But these numbers don’t often show their usefulness towards a bigger objective, which is team’s victory.</p><p>This analysis doesn’t necessarily prove that players play to improve their individual batting scores. But it points out some glaring flaws in our statistics, through which players get away even after negatively contributing towards a team’s cause, while receiving all the glory with a positive outcome.</p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=c8fe8e44c153" width="1" height="1" alt="">]]></content:encoded>
        </item>
    </channel>
</rss>