{
  "id": 55092,
  "title": "Kaggle mismanages $50+ million of research ",
  "url": "/competitions/data-science-bowl-2018/discussion/55092",
  "author_name": "Leith Akkawi",
  "post_date": "2018-04-21T23:44:49.174000",
  "votes": 3,
  "comment_count": 0,
  "views": 0,
  "content": "<p>Open Letter:</p>\n\n<p>Here’s a fact: On April 10th, the status on Kaggle showed the following: 3,570 teams, 4,490 competitors and 58,100 entries. A back-of-the-envelope calculation indicates that with 58,100 entries taking at least each 10 hours of research each by an intermediate level data scientist, which is billable at $100/hour, the total sum and value of the research can be estimated at 58,100,000 dollars. </p>\n\n<p>Here’s another set of facts: Kaggle announced that they changed the rules after the competition ended to disqualify those who haven’t submitted a model. Not only did Kaggle change the rule, but they also validated their managerial decision with some inaccurate statements about sample size sets in Stage2.</p>\n\n<p>Aren’t we all trying to advance medical research to improve the detection of nuclei? Wouldn’t it be wise at this point of Kaggle’s management of the 2018 data science bowl to open it up for participants to submit their models and incentivize everyone with a different medal or reward? Instead, Kaggle continues to misrepresent the competition details and discourages participants with inaccurate statements about the competition and rules. </p>\n\n<p>I’m sure many others like me have submitted their models to contribute to the success of the research initiative. Many others like me who are disqualfied feel unappreciated now. Many others like me would like to see this situation rectified. So let’s not waste $50+ million of research. Let’s improve the process at Kaggle. Let’s help cure diseases. Let’s help solve the next big challenge!</p>\n\n<p>Thank you,\nTEAMPDITI </p>",
  "messages": [
    {
      "id": 317590,
      "postDate": "2018-04-21T23:44:49.173Z",
      "content": "<p>Open Letter:</p>\n\n<p>Here’s a fact: On April 10th, the status on Kaggle showed the following: 3,570 teams, 4,490 competitors and 58,100 entries. A back-of-the-envelope calculation indicates that with 58,100 entries taking at least each 10 hours of research each by an intermediate level data scientist, which is billable at $100/hour, the total sum and value of the research can be estimated at 58,100,000 dollars. </p>\n\n<p>Here’s another set of facts: Kaggle announced that they changed the rules after the competition ended to disqualify those who haven’t submitted a model. Not only did Kaggle change the rule, but they also validated their managerial decision with some inaccurate statements about sample size sets in Stage2.</p>\n\n<p>Aren’t we all trying to advance medical research to improve the detection of nuclei? Wouldn’t it be wise at this point of Kaggle’s management of the 2018 data science bowl to open it up for participants to submit their models and incentivize everyone with a different medal or reward? Instead, Kaggle continues to misrepresent the competition details and discourages participants with inaccurate statements about the competition and rules. </p>\n\n<p>I’m sure many others like me have submitted their models to contribute to the success of the research initiative. Many others like me who are disqualfied feel unappreciated now. Many others like me would like to see this situation rectified. So let’s not waste $50+ million of research. Let’s improve the process at Kaggle. Let’s help cure diseases. Let’s help solve the next big challenge!</p>\n\n<p>Thank you,\nTEAMPDITI </p>",
      "rawMarkdown": "Open Letter:\n \nHere’s a fact: On April 10th, the status on Kaggle showed the following: 3,570 teams, 4,490 competitors and 58,100 entries. A back-of-the-envelope calculation indicates that with 58,100 entries taking at least each 10 hours of research each by an intermediate level data scientist, which is billable at $100/hour, the total sum and value of the research can be estimated at 58,100,000 dollars. \n \nHere’s another set of facts: Kaggle announced that they changed the rules after the competition ended to disqualify those who haven’t submitted a model. Not only did Kaggle change the rule, but they also validated their managerial decision with some inaccurate statements about sample size sets in Stage2.\n \nAren’t we all trying to advance medical research to improve the detection of nuclei? Wouldn’t it be wise at this point of Kaggle’s management of the 2018 data science bowl to open it up for participants to submit their models and incentivize everyone with a different medal or reward? Instead, Kaggle continues to misrepresent the competition details and discourages participants with inaccurate statements about the competition and rules. \n \nI’m sure many others like me have submitted their models to contribute to the success of the research initiative. Many others like me who are disqualfied feel unappreciated now. Many others like me would like to see this situation rectified. So let’s not waste $50+ million of research. Let’s improve the process at Kaggle. Let’s help cure diseases. Let’s help solve the next big challenge!\n \nThank you,\nTEAMPDITI ",
      "votes": 3
    }
  ],
  "comments": [],
  "raw_markdown_by_id": {
    "317590": "Open Letter:\n \nHere’s a fact: On April 10th, the status on Kaggle showed the following: 3,570 teams, 4,490 competitors and 58,100 entries. A back-of-the-envelope calculation indicates that with 58,100 entries taking at least each 10 hours of research each by an intermediate level data scientist, which is billable at $100/hour, the total sum and value of the research can be estimated at 58,100,000 dollars. \n \nHere’s another set of facts: Kaggle announced that they changed the rules after the competition ended to disqualify those who haven’t submitted a model. Not only did Kaggle change the rule, but they also validated their managerial decision with some inaccurate statements about sample size sets in Stage2.\n \nAren’t we all trying to advance medical research to improve the detection of nuclei? Wouldn’t it be wise at this point of Kaggle’s management of the 2018 data science bowl to open it up for participants to submit their models and incentivize everyone with a different medal or reward? Instead, Kaggle continues to misrepresent the competition details and discourages participants with inaccurate statements about the competition and rules. \n \nI’m sure many others like me have submitted their models to contribute to the success of the research initiative. Many others like me who are disqualfied feel unappreciated now. Many others like me would like to see this situation rectified. So let’s not waste $50+ million of research. Let’s improve the process at Kaggle. Let’s help cure diseases. Let’s help solve the next big challenge!\n \nThank you,\nTEAMPDITI "
  }
}