{
  "id": 207134,
  "title": "Do you use Kaggle Feed? Is it helping plagiarism? ",
  "url": "/competitions/riiid-test-answer-prediction/discussion/207134",
  "author_name": "",
  "post_date": "2020-12-28T11:44:05.886468100Z",
  "votes": 6,
  "comment_count": 3,
  "views": 0,
  "content": "<p>Hi,</p>\n<p>I just opened Kaggle home page and tried to scroll through the feed. One thing stuck to my mind. When we publish any notebook in the Kaggle feed, the notebook's we copied from doesn't appear. An example is below. I just stumbled across my own notebook in feed and didn't see the reference (I am not sure if other people have noted this. ). </p>\n<p><img src=\"https://i.imgur.com/vDQAWsG.png\" alt=\"Notebook on Kaggle Feed\"></p>\n<p>My point is that I was wondering how come <a href=\"https://www.kaggle.com/c/riiid-test-answer-prediction/discussion/206217#1124537\" target=\"_blank\">some people who are Experts and Masters upvote low effort kernels</a>. I thought that they should be able to notice the amount of value the kernel added to the original kernel if the original kernel is pretty popular. At that time I didn't consider the possibility that they might have upvoted it if they saw that in <strong>their feed</strong>. The Kaggle feed might just also become like LinkedIn feed where people are now a days copy pasting other people ideas without proper tributes. I am not sure how many of the views came from notebook section vs feed. </p>\n<p>I hope Kaggle add a small section with the link to the original kernel in the feed itself but it is too much ask. So I request you if you are upvote everything that comes into your feed, understand that you are making the copied work genuine and are promoting it. For me if someone upvotes my comments or notebooks, it makes me feel good and work harder but when I see the same people just upvote everything it makes me discount the upvote. <strong>Please don't lower the value of your upvote</strong>. </p>",
  "messages": [
    {
      "id": "1129498",
      "postDate": "12/28/2020 11:44:05",
      "content": "<p>Hi,</p>\n<p>I just opened Kaggle home page and tried to scroll through the feed. One thing stuck to my mind. When we publish any notebook in the Kaggle feed, the notebook's we copied from doesn't appear. An example is below. I just stumbled across my own notebook in feed and didn't see the reference (I am not sure if other people have noted this. ). </p>\n<p><img src=\"https://i.imgur.com/vDQAWsG.png\" alt=\"Notebook on Kaggle Feed\"></p>\n<p>My point is that I was wondering how come <a href=\"https://www.kaggle.com/c/riiid-test-answer-prediction/discussion/206217#1124537\" target=\"_blank\">some people who are Experts and Masters upvote low effort kernels</a>. I thought that they should be able to notice the amount of value the kernel added to the original kernel if the original kernel is pretty popular. At that time I didn't consider the possibility that they might have upvoted it if they saw that in <strong>their feed</strong>. The Kaggle feed might just also become like LinkedIn feed where people are now a days copy pasting other people ideas without proper tributes. I am not sure how many of the views came from notebook section vs feed. </p>\n<p>I hope Kaggle add a small section with the link to the original kernel in the feed itself but it is too much ask. So I request you if you are upvote everything that comes into your feed, understand that you are making the copied work genuine and are promoting it. For me if someone upvotes my comments or notebooks, it makes me feel good and work harder but when I see the same people just upvote everything it makes me discount the upvote. <strong>Please don't lower the value of your upvote</strong>. </p>",
      "rawMarkdown": "Hi,\n\nI just opened Kaggle home page and tried to scroll through the feed. One thing stuck to my mind. When we publish any notebook in the Kaggle feed, the notebook's we copied from doesn't appear. An example is below. I just stumbled across my own notebook in feed and didn't see the reference (I am not sure if other people have noted this. ). \n\n![Notebook on Kaggle Feed](https://i.imgur.com/vDQAWsG.png)\n\nMy point is that I was wondering how come [some people who are Experts and Masters upvote low effort kernels](https://www.kaggle.com/c/riiid-test-answer-prediction/discussion/206217#1124537). I thought that they should be able to notice the amount of value the kernel added to the original kernel if the original kernel is pretty popular. At that time I didn't consider the possibility that they might have upvoted it if they saw that in **their feed**. The Kaggle feed might just also become like LinkedIn feed where people are now a days copy pasting other people ideas without proper tributes. I am not sure how many of the views came from notebook section vs feed. \n\nI hope Kaggle add a small section with the link to the original kernel in the feed itself but it is too much ask. So I request you if you are upvote everything that comes into your feed, understand that you are making the copied work genuine and are promoting it. For me if someone upvotes my comments or notebooks, it makes me feel good and work harder but when I see the same people just upvote everything it makes me discount the upvote. **Please don't lower the value of your upvote**.",
      "votes": null
    },
    {
      "id": "1129557",
      "postDate": "12/28/2020 12:34:43",
      "content": "<p>That would help but not always. People can also just download the code and upload it into a new notebook. That is not a fork….</p>\n<p>I would by the way also strongly encourage Kaggle to check if votes on private notebooks still count. I am aware of 2 people who seems to have deleted Gold medals. Now, why would you do that? Very likely because you know that you have stolen them from somewhere and am feeling the heat from plagiarism hunters. However, the rank of those 2 people never went down badly. Therefore, I suspect that they did not delete those notebooks but made them private…..and votes still count.</p>",
      "rawMarkdown": "That would help but not always. People can also just download the code and upload it into a new notebook. That is not a fork....\n\nI would by the way also strongly encourage Kaggle to check if votes on private notebooks still count. I am aware of 2 people who seems to have deleted Gold medals. Now, why would you do that? Very likely because you know that you have stolen them from somewhere and am feeling the heat from plagiarism hunters. However, the rank of those 2 people never went down badly. Therefore, I suspect that they did not delete those notebooks but made them private.....and votes still count.",
      "votes": null
    },
    {
      "id": "1129564",
      "postDate": "12/28/2020 12:42:50",
      "content": "<p>Thanks Erik. I missed that people could just re-upload notebooks. This would make sense why Kaggle hasn't tried this feature out. I also didn't know that private notebooks upvotes are counted.</p>\n<p>It's great that you have been trying hard to provide feedback to Kaggle teams. I saw some of your threads on how people could abuse Hotness algorithm. Thanks for the time you spend on providing feedback. I have learnt a lot from your discussions. </p>",
      "rawMarkdown": "Thanks Erik. I missed that people could just re-upload notebooks. This would make sense why Kaggle hasn't tried this feature out. I also didn't know that private notebooks upvotes are counted.\n\nIt's great that you have been trying hard to provide feedback to Kaggle teams. I saw some of your threads on how people could abuse Hotness algorithm. Thanks for the time you spend on providing feedback. I have learnt a lot from your discussions.",
      "votes": null
    },
    {
      "id": "1129571",
      "postDate": "12/28/2020 12:52:35",
      "content": "<p>Abuse of hotness is unfortunately going on for years. Up untill about a month ago somebody was still spamming the list with unsignificant updates. However, something seems to have changed recently. I found out when I was finishing off my EDA. Don't know what happened exactly but older stuff with hunderds of votes does not show up anymore. I hope that Kaggle thought it out well ;-).</p>",
      "rawMarkdown": "Abuse of hotness is unfortunately going on for years. Up untill about a month ago somebody was still spamming the list with unsignificant updates. However, something seems to have changed recently. I found out when I was finishing off my EDA. Don't know what happened exactly but older stuff with hunderds of votes does not show up anymore. I hope that Kaggle thought it out well ;-).",
      "votes": null
    }
  ],
  "comments": [
    {
      "id": 1129557,
      "author_name": "erikbruin",
      "author_url": "",
      "post_date": "12/28/2020 12:34:43",
      "content": "<p>That would help but not always. People can also just download the code and upload it into a new notebook. That is not a fork….</p>\n<p>I would by the way also strongly encourage Kaggle to check if votes on private notebooks still count. I am aware of 2 people who seems to have deleted Gold medals. Now, why would you do that? Very likely because you know that you have stolen them from somewhere and am feeling the heat from plagiarism hunters. However, the rank of those 2 people never went down badly. Therefore, I suspect that they did not delete those notebooks but made them private…..and votes still count.</p>",
      "votes": null,
      "replies": [
        {
          "id": 1129564,
          "author_name": "manikanthr5",
          "author_url": "",
          "post_date": "12/28/2020 12:42:50",
          "content": "<p>Thanks Erik. I missed that people could just re-upload notebooks. This would make sense why Kaggle hasn't tried this feature out. I also didn't know that private notebooks upvotes are counted.</p>\n<p>It's great that you have been trying hard to provide feedback to Kaggle teams. I saw some of your threads on how people could abuse Hotness algorithm. Thanks for the time you spend on providing feedback. I have learnt a lot from your discussions. </p>",
          "votes": null,
          "replies": []
        },
        {
          "id": 1129571,
          "author_name": "erikbruin",
          "author_url": "",
          "post_date": "12/28/2020 12:52:35",
          "content": "<p>Abuse of hotness is unfortunately going on for years. Up untill about a month ago somebody was still spamming the list with unsignificant updates. However, something seems to have changed recently. I found out when I was finishing off my EDA. Don't know what happened exactly but older stuff with hunderds of votes does not show up anymore. I hope that Kaggle thought it out well ;-).</p>",
          "votes": null,
          "replies": []
        }
      ]
    }
  ],
  "raw_markdown_by_id": {
    "1129498": "Hi,\n\nI just opened Kaggle home page and tried to scroll through the feed. One thing stuck to my mind. When we publish any notebook in the Kaggle feed, the notebook's we copied from doesn't appear. An example is below. I just stumbled across my own notebook in feed and didn't see the reference (I am not sure if other people have noted this. ). \n\n![Notebook on Kaggle Feed](https://i.imgur.com/vDQAWsG.png)\n\nMy point is that I was wondering how come [some people who are Experts and Masters upvote low effort kernels](https://www.kaggle.com/c/riiid-test-answer-prediction/discussion/206217#1124537). I thought that they should be able to notice the amount of value the kernel added to the original kernel if the original kernel is pretty popular. At that time I didn't consider the possibility that they might have upvoted it if they saw that in **their feed**. The Kaggle feed might just also become like LinkedIn feed where people are now a days copy pasting other people ideas without proper tributes. I am not sure how many of the views came from notebook section vs feed. \n\nI hope Kaggle add a small section with the link to the original kernel in the feed itself but it is too much ask. So I request you if you are upvote everything that comes into your feed, understand that you are making the copied work genuine and are promoting it. For me if someone upvotes my comments or notebooks, it makes me feel good and work harder but when I see the same people just upvote everything it makes me discount the upvote. **Please don't lower the value of your upvote**.",
    "1129557": "That would help but not always. People can also just download the code and upload it into a new notebook. That is not a fork....\n\nI would by the way also strongly encourage Kaggle to check if votes on private notebooks still count. I am aware of 2 people who seems to have deleted Gold medals. Now, why would you do that? Very likely because you know that you have stolen them from somewhere and am feeling the heat from plagiarism hunters. However, the rank of those 2 people never went down badly. Therefore, I suspect that they did not delete those notebooks but made them private.....and votes still count.",
    "1129564": "Thanks Erik. I missed that people could just re-upload notebooks. This would make sense why Kaggle hasn't tried this feature out. I also didn't know that private notebooks upvotes are counted.\n\nIt's great that you have been trying hard to provide feedback to Kaggle teams. I saw some of your threads on how people could abuse Hotness algorithm. Thanks for the time you spend on providing feedback. I have learnt a lot from your discussions.",
    "1129571": "Abuse of hotness is unfortunately going on for years. Up untill about a month ago somebody was still spamming the list with unsignificant updates. However, something seems to have changed recently. I found out when I was finishing off my EDA. Don't know what happened exactly but older stuff with hunderds of votes does not show up anymore. I hope that Kaggle thought it out well ;-)."
  },
  "source": "meta"
}