{
  "id": 18508,
  "title": "Regulatory/ethical issues with using open source code for new medical procedure?",
  "url": "/competitions/second-annual-data-science-bowl/discussion/18508",
  "author_name": "",
  "post_date": "2016-01-22T04:07:30.417Z",
  "votes": -8,
  "comment_count": 3,
  "views": 726,
  "content": "<p>Where does the FDA stand when it comes to regulating new medical procedures developed with open source code such as R or Python vs. commercial software such as SAS or Matlab? </p>\n\n<p>How would you convince a physician to implement the winning methodology in this competition if he or she feels uncomfortable with relying on open source code? For that matter, given that people's lives may be at stake, <em>should</em> we be using open source code?</p>",
  "messages": [
    {
      "id": "105325",
      "postDate": "01/22/2016 04:07:30",
      "content": "<p>Where does the FDA stand when it comes to regulating new medical procedures developed with open source code such as R or Python vs. commercial software such as SAS or Matlab? </p>\n\n<p>How would you convince a physician to implement the winning methodology in this competition if he or she feels uncomfortable with relying on open source code? For that matter, given that people's lives may be at stake, <em>should</em> we be using open source code?</p>",
      "rawMarkdown": "Where does the FDA stand when it comes to regulating new medical procedures developed with open source code such as R or Python vs. commercial software such as SAS or Matlab? \r\n\r\nHow would you convince a physician to implement the winning methodology in this competition if he or she feels uncomfortable with relying on open source code? For that matter, given that people's lives may be at stake, *should* we be using open source code?",
      "votes": null
    },
    {
      "id": "105567",
      "postDate": "01/24/2016 21:46:46",
      "content": "<blockquote>\n  <p>For that matter, given that people's lives may be at stake, should we be using open source code?</p>\n</blockquote>\n\n<p>I think you misspelled &quot;closed&quot; (oops!). I agree with your sentiment: in a situation where people's lives are at stake, it would be ridiculous to use black-box, closed-source software that cannot be verified to work as advertised. Especially if there are financial incentives on the line to falsely advertise the capability of the software and swipe bugs under the rug, as is often the case with closed-source software. Think of the VW cheating software scandal: this is what you get when you use closed-source for important matters.</p>\n\n<p>&quot;Just trust us&quot; doesn't cut it. For important matters, you want code that can be openly scrutinized. Especially when private incentives are not aligned with the public interest, and especially in the medical sector.</p>",
      "rawMarkdown": "> For that matter, given that people's lives may be at stake, should we be using open source code?\r\n\r\nI think you misspelled \"closed\" (oops!). I agree with your sentiment: in a situation where people's lives are at stake, it would be ridiculous to use black-box, closed-source software that cannot be verified to work as advertised. Especially if there are financial incentives on the line to falsely advertise the capability of the software and swipe bugs under the rug, as is often the case with closed-source software. Think of the VW cheating software scandal: this is what you get when you use closed-source for important matters.\r\n\r\n\"Just trust us\" doesn't cut it. For important matters, you want code that can be openly scrutinized. Especially when private incentives are not aligned with the public interest, and especially in the medical sector.",
      "votes": null
    },
    {
      "id": "105568",
      "postDate": "01/24/2016 21:56:35",
      "content": "<p>fchollet: +1, +100 if I could. Reliability of closed vs open was a worthwhile discussion in the mid nineties, it's not worth rehashing it.</p>",
      "rawMarkdown": "fchollet: +1, +100 if I could. Reliability of closed vs open was a worthwhile discussion in the mid nineties, it's not worth rehashing it.",
      "votes": null
    },
    {
      "id": "105580",
      "postDate": "01/25/2016 00:29:18",
      "content": "<p>fchollet - yes, good points in favor of open source.</p>\n\n<p>Found this Revolutions blog article from four years ago:</p>\n\n<p><a href=\"http://blog.revolutionanalytics.com/2012/06/fda-r-ok.html\">http://blog.revolutionanalytics.com/2012/06/fda-r-ok.html</a></p>",
      "rawMarkdown": "fchollet - yes, good points in favor of open source.\r\n\r\nFound this Revolutions blog article from four years ago:\r\n\r\nhttp://blog.revolutionanalytics.com/2012/06/fda-r-ok.html",
      "votes": null
    }
  ],
  "comments": [
    {
      "id": 105567,
      "author_name": "fchollet",
      "author_url": "",
      "post_date": "01/24/2016 21:46:46",
      "content": "<blockquote>\n  <p>For that matter, given that people's lives may be at stake, should we be using open source code?</p>\n</blockquote>\n\n<p>I think you misspelled &quot;closed&quot; (oops!). I agree with your sentiment: in a situation where people's lives are at stake, it would be ridiculous to use black-box, closed-source software that cannot be verified to work as advertised. Especially if there are financial incentives on the line to falsely advertise the capability of the software and swipe bugs under the rug, as is often the case with closed-source software. Think of the VW cheating software scandal: this is what you get when you use closed-source for important matters.</p>\n\n<p>&quot;Just trust us&quot; doesn't cut it. For important matters, you want code that can be openly scrutinized. Especially when private incentives are not aligned with the public interest, and especially in the medical sector.</p>",
      "votes": null,
      "replies": []
    },
    {
      "id": 105568,
      "author_name": "jfkingiii",
      "author_url": "",
      "post_date": "01/24/2016 21:56:35",
      "content": "<p>fchollet: +1, +100 if I could. Reliability of closed vs open was a worthwhile discussion in the mid nineties, it's not worth rehashing it.</p>",
      "votes": null,
      "replies": []
    },
    {
      "id": 105580,
      "author_name": "robertfeyerharm",
      "author_url": "",
      "post_date": "01/25/2016 00:29:18",
      "content": "<p>fchollet - yes, good points in favor of open source.</p>\n\n<p>Found this Revolutions blog article from four years ago:</p>\n\n<p><a href=\"http://blog.revolutionanalytics.com/2012/06/fda-r-ok.html\">http://blog.revolutionanalytics.com/2012/06/fda-r-ok.html</a></p>",
      "votes": null,
      "replies": []
    }
  ],
  "raw_markdown_by_id": {
    "105325": "Where does the FDA stand when it comes to regulating new medical procedures developed with open source code such as R or Python vs. commercial software such as SAS or Matlab? \r\n\r\nHow would you convince a physician to implement the winning methodology in this competition if he or she feels uncomfortable with relying on open source code? For that matter, given that people's lives may be at stake, *should* we be using open source code?",
    "105567": "> For that matter, given that people's lives may be at stake, should we be using open source code?\r\n\r\nI think you misspelled \"closed\" (oops!). I agree with your sentiment: in a situation where people's lives are at stake, it would be ridiculous to use black-box, closed-source software that cannot be verified to work as advertised. Especially if there are financial incentives on the line to falsely advertise the capability of the software and swipe bugs under the rug, as is often the case with closed-source software. Think of the VW cheating software scandal: this is what you get when you use closed-source for important matters.\r\n\r\n\"Just trust us\" doesn't cut it. For important matters, you want code that can be openly scrutinized. Especially when private incentives are not aligned with the public interest, and especially in the medical sector.",
    "105568": "fchollet: +1, +100 if I could. Reliability of closed vs open was a worthwhile discussion in the mid nineties, it's not worth rehashing it.",
    "105580": "fchollet - yes, good points in favor of open source.\r\n\r\nFound this Revolutions blog article from four years ago:\r\n\r\nhttp://blog.revolutionanalytics.com/2012/06/fda-r-ok.html"
  },
  "source": "meta"
}