{
  "id": 23728,
  "title": "Questions about the Rules",
  "url": "/competitions/bosch-production-line-performance/discussion/23728",
  "author_name": "",
  "post_date": "2016-09-17T17:59:11.667Z",
  "votes": null,
  "comment_count": 1,
  "views": 563,
  "content": "<p>Hello,</p>\n\n<p>This is a question for the Admins.</p>\n\n<p>I am pretty new to Kaggle, so forgive me if these questions have already been answered.\nMy teammates and I have read through the rules, but a few things are still not completely clear to us.</p>\n\n<p>Before asking my questions, I want to emphasize that we have <em>not</em> shared data and we have <em>not</em> used any proprietary software. </p>\n\n<p>We just wanted to be proactive about this, and understand the rules before we begin.</p>\n\n<ol>\n<li><p>Are we allowed to share competition data with teammates? \nIn particular, can I share the following with my teammates:\n\n\n</p><blockquote>\n  <p></p>\n</blockquote>\n\n<blockquote>\n  <p>(a) the raw data, </p>\n  \n  <p>(b) the raw data stored in a different open-source compression format, </p>\n  \n  <p>(c) the raw data with additional engineered features?</p>\n</blockquote>\n\n<ol start=\"2\"><li>Are we allowed to use proprietary software in the <em>infrastructure</em> of our setup? </li></ol></li>\n</ol>\n\n<blockquote>\n  <p>For example, can we use Amazon EC2 Instances to train models, Amazon S3 for data and feature sharing, Amazon RDS for sharing models and results, etc.?</p>\n  \n  <p>We wanted to use this setup to facilitate collaboration and expedite development. \n  This infrastructure would not be necessary to create/train the final model(s)--one could use any Hadoop cluster running Apache Spark. </p>\n  \n  <p>However, the use of this infrastructure (EC2, S3, RDS) could shape the decisions we make for model exploration and feature selection, and their use might be necessary to <em>recreate</em> our full development process. </p>\n  \n  <p>So those are our concerns. What do you think?</p>\n</blockquote>\n\n<p>I know I am overthinking things, but I thought it would be worthwhile to get this settled. \nThanks very much.</p>",
  "messages": [
    {
      "id": "135857",
      "postDate": "09/17/2016 17:59:11",
      "content": "<p>Hello,</p>\n\n<p>This is a question for the Admins.</p>\n\n<p>I am pretty new to Kaggle, so forgive me if these questions have already been answered.\nMy teammates and I have read through the rules, but a few things are still not completely clear to us.</p>\n\n<p>Before asking my questions, I want to emphasize that we have <em>not</em> shared data and we have <em>not</em> used any proprietary software. </p>\n\n<p>We just wanted to be proactive about this, and understand the rules before we begin.</p>\n\n<ol>\n<li><p>Are we allowed to share competition data with teammates? \nIn particular, can I share the following with my teammates:\n\n\n</p><blockquote>\n  <p></p>\n</blockquote>\n\n<blockquote>\n  <p>(a) the raw data, </p>\n  \n  <p>(b) the raw data stored in a different open-source compression format, </p>\n  \n  <p>(c) the raw data with additional engineered features?</p>\n</blockquote>\n\n<ol start=\"2\"><li>Are we allowed to use proprietary software in the <em>infrastructure</em> of our setup? </li></ol></li>\n</ol>\n\n<blockquote>\n  <p>For example, can we use Amazon EC2 Instances to train models, Amazon S3 for data and feature sharing, Amazon RDS for sharing models and results, etc.?</p>\n  \n  <p>We wanted to use this setup to facilitate collaboration and expedite development. \n  This infrastructure would not be necessary to create/train the final model(s)--one could use any Hadoop cluster running Apache Spark. </p>\n  \n  <p>However, the use of this infrastructure (EC2, S3, RDS) could shape the decisions we make for model exploration and feature selection, and their use might be necessary to <em>recreate</em> our full development process. </p>\n  \n  <p>So those are our concerns. What do you think?</p>\n</blockquote>\n\n<p>I know I am overthinking things, but I thought it would be worthwhile to get this settled. \nThanks very much.</p>",
      "rawMarkdown": "Hello,\r\n\r\nThis is a question for the Admins.\r\n\r\nI am pretty new to Kaggle, so forgive me if these questions have already been answered.\r\nMy teammates and I have read through the rules, but a few things are still not completely clear to us.\r\n\r\nBefore asking my questions, I want to emphasize that we have *not* shared data and we have *not* used any proprietary software. \r\n\r\nWe just wanted to be proactive about this, and understand the rules before we begin.\r\n\r\n 1. Are we allowed to share competition data with teammates? \r\nIn particular, can I share the following with my teammates:\r\n>\r\n> (a) the raw data, \r\n> \r\n> (b) the raw data stored in a different open-source compression format, \r\n> \r\n> (c) the raw data with additional engineered features?\r\n\r\n2. Are we allowed to use proprietary software in the *infrastructure* of our setup? \r\n\r\n> For example, can we use Amazon EC2 Instances to train models, Amazon S3 for data and feature sharing, Amazon RDS for sharing models and results, etc.?\r\n>\r\n> We wanted to use this setup to facilitate collaboration and expedite development. \r\n> This infrastructure would not be necessary to create/train the final model(s)--one could use any Hadoop cluster running Apache Spark. \r\n>\r\n> However, the use of this infrastructure (EC2, S3, RDS) could shape the decisions we make for model exploration and feature selection, and their use might be necessary to *recreate* our full development process. \r\n>\r\n> So those are our concerns. What do you think?\r\n\r\nI know I am overthinking things, but I thought it would be worthwhile to get this settled. \r\nThanks very much.",
      "votes": null
    },
    {
      "id": "136055",
      "postDate": "09/19/2016 19:17:55",
      "content": "<p>Hi Robert,</p>\n\n<ol>\n<li>Yes, provided your teammates have Kaggle accounts and have agreed to the competition rules, it's fine to share with them (&quot;[participants] agree not to transmit, duplicate, publish, redistribute or otherwise provide or make available the Data to any party not participating in the Competition&quot;).</li>\n<li>Yes, it's fine to use common cloud infrastructure like you mentioned. The most important consideration is that you are able to reproduce your methodology and that you have the rights to grant a license to the parts that you develop.</li>\n</ol>",
      "rawMarkdown": "Hi Robert,\r\n\r\n 1. Yes, provided your teammates have Kaggle accounts and have agreed to the competition rules, it's fine to share with them (\"[participants] agree not to transmit, duplicate, publish, redistribute or otherwise provide or make available the Data to any party not participating in the Competition\").\r\n 2. Yes, it's fine to use common cloud infrastructure like you mentioned. The most important consideration is that you are able to reproduce your methodology and that you have the rights to grant a license to the parts that you develop.",
      "votes": null
    }
  ],
  "comments": [
    {
      "id": 136055,
      "author_name": "wcukierski",
      "author_url": "",
      "post_date": "09/19/2016 19:17:55",
      "content": "<p>Hi Robert,</p>\n\n<ol>\n<li>Yes, provided your teammates have Kaggle accounts and have agreed to the competition rules, it's fine to share with them (&quot;[participants] agree not to transmit, duplicate, publish, redistribute or otherwise provide or make available the Data to any party not participating in the Competition&quot;).</li>\n<li>Yes, it's fine to use common cloud infrastructure like you mentioned. The most important consideration is that you are able to reproduce your methodology and that you have the rights to grant a license to the parts that you develop.</li>\n</ol>",
      "votes": null,
      "replies": []
    }
  ],
  "raw_markdown_by_id": {
    "135857": "Hello,\r\n\r\nThis is a question for the Admins.\r\n\r\nI am pretty new to Kaggle, so forgive me if these questions have already been answered.\r\nMy teammates and I have read through the rules, but a few things are still not completely clear to us.\r\n\r\nBefore asking my questions, I want to emphasize that we have *not* shared data and we have *not* used any proprietary software. \r\n\r\nWe just wanted to be proactive about this, and understand the rules before we begin.\r\n\r\n 1. Are we allowed to share competition data with teammates? \r\nIn particular, can I share the following with my teammates:\r\n>\r\n> (a) the raw data, \r\n> \r\n> (b) the raw data stored in a different open-source compression format, \r\n> \r\n> (c) the raw data with additional engineered features?\r\n\r\n2. Are we allowed to use proprietary software in the *infrastructure* of our setup? \r\n\r\n> For example, can we use Amazon EC2 Instances to train models, Amazon S3 for data and feature sharing, Amazon RDS for sharing models and results, etc.?\r\n>\r\n> We wanted to use this setup to facilitate collaboration and expedite development. \r\n> This infrastructure would not be necessary to create/train the final model(s)--one could use any Hadoop cluster running Apache Spark. \r\n>\r\n> However, the use of this infrastructure (EC2, S3, RDS) could shape the decisions we make for model exploration and feature selection, and their use might be necessary to *recreate* our full development process. \r\n>\r\n> So those are our concerns. What do you think?\r\n\r\nI know I am overthinking things, but I thought it would be worthwhile to get this settled. \r\nThanks very much.",
    "136055": "Hi Robert,\r\n\r\n 1. Yes, provided your teammates have Kaggle accounts and have agreed to the competition rules, it's fine to share with them (\"[participants] agree not to transmit, duplicate, publish, redistribute or otherwise provide or make available the Data to any party not participating in the Competition\").\r\n 2. Yes, it's fine to use common cloud infrastructure like you mentioned. The most important consideration is that you are able to reproduce your methodology and that you have the rights to grant a license to the parts that you develop."
  },
  "source": "meta"
}