{
  "id": 3321,
  "title": "flighthistory.csv dates",
  "url": "/competitions/flight/discussion/3321",
  "author_name": "Gábor Melis",
  "post_date": "2012-12-13T15:00:18.177000",
  "votes": 0,
  "comment_count": 0,
  "views": 1626,
  "content": "<p>Where do the dates in flighthistory.csv come from? I'm trying to see if the actual/scheduled gate/runway departure/arrival dates can be calculated from flighthistoryevents.csv.</p>\r\n<p>So far it seems that most of the time the value (after &quot;New=&quot;) of last ARA event matches the date in actual_runway_arrival. There are some mismatches though:</p>\r\n<pre>WARNING:<br>   Flight 280246329 (diverted) has value 3561813240 where event AGA has 3561809340<br>WARNING:<br>   Flight 280246329 (diverted) has value 3561813120 where event ARA has 3561809100<br>WARNING: Flight 280462983 has value 3562034399 where event ARA has 3562034340<br>WARNING: Flight 280474122 has value 3562041599 where event ARA has 3562041540<br>WARNING: Flight 280754886 has value 3562300799 where event ARA has 3562300740<br>WARNING: Flight 280848254 has value 3562379999 where event ARA has 3562379940<br>WARNING: Flight 280904396 has value 3562462799 where event ARA has 3562462740<br>WARNING: Flight 281131835 has value 3562473599 where event ARA has 3562473540<br>WARNING: Flight 281259821 has value 3562811999 where event ARA has 3562811940<br>WARNING: Flight 281328141 has value 3562898399 where event ARA has 3562898340</pre>\r\n<p>The values are second since the the unix epoch.</p>\r\n<p>The two diverted flights have bigger discrepancies. Shouldn't they be dropped from the dataset? We don't know where they actually landed.</p>\r\n<p>The rest have 59 seconds difference.</p>",
  "messages": [
    {
      "id": 17779,
      "postDate": "2012-12-13T15:00:18.177Z",
      "content": "<p>Where do the dates in flighthistory.csv come from? I'm trying to see if the actual/scheduled gate/runway departure/arrival dates can be calculated from flighthistoryevents.csv.</p>\r\n<p>So far it seems that most of the time the value (after &quot;New=&quot;) of last ARA event matches the date in actual_runway_arrival. There are some mismatches though:</p>\r\n<pre>WARNING:<br>   Flight 280246329 (diverted) has value 3561813240 where event AGA has 3561809340<br>WARNING:<br>   Flight 280246329 (diverted) has value 3561813120 where event ARA has 3561809100<br>WARNING: Flight 280462983 has value 3562034399 where event ARA has 3562034340<br>WARNING: Flight 280474122 has value 3562041599 where event ARA has 3562041540<br>WARNING: Flight 280754886 has value 3562300799 where event ARA has 3562300740<br>WARNING: Flight 280848254 has value 3562379999 where event ARA has 3562379940<br>WARNING: Flight 280904396 has value 3562462799 where event ARA has 3562462740<br>WARNING: Flight 281131835 has value 3562473599 where event ARA has 3562473540<br>WARNING: Flight 281259821 has value 3562811999 where event ARA has 3562811940<br>WARNING: Flight 281328141 has value 3562898399 where event ARA has 3562898340</pre>\r\n<p>The values are second since the the unix epoch.</p>\r\n<p>The two diverted flights have bigger discrepancies. Shouldn't they be dropped from the dataset? We don't know where they actually landed.</p>\r\n<p>The rest have 59 seconds difference.</p>"
    }
  ],
  "comments": [],
  "raw_markdown_by_id": {
    "17779": ""
  }
}