{
  "id": 364921,
  "title": "Can we extend the deadline time for submission as the problem is little complex?",
  "url": "/competitions/open-problems-multimodal/discussion/364921",
  "author_name": "",
  "post_date": "2022-11-09T03:18:47.628669Z",
  "votes": -11,
  "comment_count": 3,
  "views": 0,
  "content": "<p><a href=\"https://www.kaggle.com/danielburkhardt\" target=\"_blank\">@danielburkhardt</a> : can you please consider extending deadline as the problem is hard to resolve?</p>",
  "messages": [
    {
      "id": "2022430",
      "postDate": "11/09/2022 03:18:47",
      "content": "<p><a href=\"https://www.kaggle.com/danielburkhardt\" target=\"_blank\">@danielburkhardt</a> : can you please consider extending deadline as the problem is hard to resolve?</p>",
      "rawMarkdown": "danielburkhardt : can you please consider extending deadline as the problem is hard to resolve?",
      "votes": null
    },
    {
      "id": "2023059",
      "postDate": "11/09/2022 14:00:58",
      "content": "<p>I won't downvote this but I will explain the downvotes and why they won't extend a competition… especially for the provided reason of it being 'hard to resolve'.</p>\n<ol>\n<li><p>This competition problem statement looks relatively solved. i.e. The top of the leaderboard has converged to a pretty firm asymptote (~0.815) that could be approaching the bayes limit of performance (minimum possible error). As such more time is not really required from the hosts side of things.</p></li>\n<li><p>The hosts double dipped on this competition by essentially wrapping two competitions in one. This is fine, but it obviously makes this competition more complex than others may be. As such, the risk of users not being able to resolve/understand the problem in the given time is a risk that falls on the hosts.</p></li>\n<li><p>If you want to understand/resolve this problem after the competition ends, there's nothing stopping you from continuing to work on the problem. The private data remains private and you can keep developing better and better solutions if you so choose.</p></li>\n</ol>\n<p>Finally, Kaggle is a platform that charges to host competitions. The terms are set early on and are based on some agreement between the hosts and Kaggle. If the hosts thought this was long enough to solve this problem then they are bound to that (unless they want to host a second competition – like the feedback prize competition which has done multiple competitions).</p>\n<p>I hope this helps you understand. I hope you take it on yourself to continue trying to resolve the problem, even after the competition closes.</p>\n<p>Cheers!</p>",
      "rawMarkdown": "I won't downvote this but I will explain the downvotes and why they won't extend a competition... especially for the provided reason of it being 'hard to resolve'.\n\n1. This competition problem statement looks relatively solved. i.e. The top of the leaderboard has converged to a pretty firm asymptote (~0.815) that could be approaching the bayes limit of performance (minimum possible error). As such more time is not really required from the hosts side of things.\n\n2. The hosts double dipped on this competition by essentially wrapping two competitions in one. This is fine, but it obviously makes this competition more complex than others may be. As such, the risk of users not being able to resolve/understand the problem in the given time is a risk that falls on the hosts.\n\n3. If you want to understand/resolve this problem after the competition ends, there's nothing stopping you from continuing to work on the problem. The private data remains private and you can keep developing better and better solutions if you so choose.\n\nFinally, Kaggle is a platform that charges to host competitions. The terms are set early on and are based on some agreement between the hosts and Kaggle. If the hosts thought this was long enough to solve this problem then they are bound to that (unless they want to host a second competition – like the feedback prize competition which has done multiple competitions).\n\nI hope this helps you understand. I hope you take it on yourself to continue trying to resolve the problem, even after the competition closes.\n\nCheers!",
      "votes": null
    },
    {
      "id": "2026713",
      "postDate": "11/12/2022 09:58:30",
      "content": "<p>Everyday trying to improve the ranking, only to wake up and find myself back in the previous day ranking again… </p>\n<p>Can’t wait for this to end now soon, running out of ways to improve and trying new ideas, and working for two problem simultaneously, brainstorming on them is a tiring process.</p>",
      "rawMarkdown": "Everyday trying to improve the ranking, only to wake up and find myself back in the previous day ranking again… \n\nCan’t wait for this to end now soon, running out of ways to improve and trying new ideas, and working for two problem simultaneously, brainstorming on them is a tiring process.",
      "votes": null
    },
    {
      "id": "2030881",
      "postDate": "11/15/2022 17:16:13",
      "content": "<p>I do not think the deadline can be extended. Cheers!</p>",
      "rawMarkdown": "I do not think the deadline can be extended. Cheers!",
      "votes": null
    }
  ],
  "comments": [
    {
      "id": 2023059,
      "author_name": "dschettler8845",
      "author_url": "",
      "post_date": "11/09/2022 14:00:58",
      "content": "<p>I won't downvote this but I will explain the downvotes and why they won't extend a competition… especially for the provided reason of it being 'hard to resolve'.</p>\n<ol>\n<li><p>This competition problem statement looks relatively solved. i.e. The top of the leaderboard has converged to a pretty firm asymptote (~0.815) that could be approaching the bayes limit of performance (minimum possible error). As such more time is not really required from the hosts side of things.</p></li>\n<li><p>The hosts double dipped on this competition by essentially wrapping two competitions in one. This is fine, but it obviously makes this competition more complex than others may be. As such, the risk of users not being able to resolve/understand the problem in the given time is a risk that falls on the hosts.</p></li>\n<li><p>If you want to understand/resolve this problem after the competition ends, there's nothing stopping you from continuing to work on the problem. The private data remains private and you can keep developing better and better solutions if you so choose.</p></li>\n</ol>\n<p>Finally, Kaggle is a platform that charges to host competitions. The terms are set early on and are based on some agreement between the hosts and Kaggle. If the hosts thought this was long enough to solve this problem then they are bound to that (unless they want to host a second competition – like the feedback prize competition which has done multiple competitions).</p>\n<p>I hope this helps you understand. I hope you take it on yourself to continue trying to resolve the problem, even after the competition closes.</p>\n<p>Cheers!</p>",
      "votes": null,
      "replies": []
    },
    {
      "id": 2026713,
      "author_name": "tarique7",
      "author_url": "",
      "post_date": "11/12/2022 09:58:30",
      "content": "<p>Everyday trying to improve the ranking, only to wake up and find myself back in the previous day ranking again… </p>\n<p>Can’t wait for this to end now soon, running out of ways to improve and trying new ideas, and working for two problem simultaneously, brainstorming on them is a tiring process.</p>",
      "votes": null,
      "replies": []
    },
    {
      "id": 2030881,
      "author_name": "juliankim",
      "author_url": "",
      "post_date": "11/15/2022 17:16:13",
      "content": "<p>I do not think the deadline can be extended. Cheers!</p>",
      "votes": null,
      "replies": []
    }
  ],
  "raw_markdown_by_id": {
    "2022430": "danielburkhardt : can you please consider extending deadline as the problem is hard to resolve?",
    "2023059": "I won't downvote this but I will explain the downvotes and why they won't extend a competition... especially for the provided reason of it being 'hard to resolve'.\n\n1. This competition problem statement looks relatively solved. i.e. The top of the leaderboard has converged to a pretty firm asymptote (~0.815) that could be approaching the bayes limit of performance (minimum possible error). As such more time is not really required from the hosts side of things.\n\n2. The hosts double dipped on this competition by essentially wrapping two competitions in one. This is fine, but it obviously makes this competition more complex than others may be. As such, the risk of users not being able to resolve/understand the problem in the given time is a risk that falls on the hosts.\n\n3. If you want to understand/resolve this problem after the competition ends, there's nothing stopping you from continuing to work on the problem. The private data remains private and you can keep developing better and better solutions if you so choose.\n\nFinally, Kaggle is a platform that charges to host competitions. The terms are set early on and are based on some agreement between the hosts and Kaggle. If the hosts thought this was long enough to solve this problem then they are bound to that (unless they want to host a second competition – like the feedback prize competition which has done multiple competitions).\n\nI hope this helps you understand. I hope you take it on yourself to continue trying to resolve the problem, even after the competition closes.\n\nCheers!",
    "2026713": "Everyday trying to improve the ranking, only to wake up and find myself back in the previous day ranking again… \n\nCan’t wait for this to end now soon, running out of ways to improve and trying new ideas, and working for two problem simultaneously, brainstorming on them is a tiring process.",
    "2030881": "I do not think the deadline can be extended. Cheers!"
  },
  "source": "meta"
}