{
  "id": 10899,
  "title": "deadline was misleading",
  "url": "/competitions/seizure-prediction/discussion/10899",
  "author_name": "",
  "post_date": "2014-11-11T10:53:27.073Z",
  "votes": -2,
  "comment_count": 21,
  "views": 3545,
  "content": "<p>Hi,</p>\n<p>I wanted to upload my predictions today (already subscribe for the competition before) but apparently you had to put your first prediction until yesterday (&quot;team merge deadline&quot;). This is not fair. I worked a lot on this competition! The headline says 6.5 days to go but I cannot upload my predictions!</p>\n\n<p>Hanan.</p>",
  "messages": [
    {
      "id": "57765",
      "postDate": "11/11/2014 10:53:27",
      "content": "<p>Hi,</p>\n<p>I wanted to upload my predictions today (already subscribe for the competition before) but apparently you had to put your first prediction until yesterday (&quot;team merge deadline&quot;). This is not fair. I worked a lot on this competition! The headline says 6.5 days to go but I cannot upload my predictions!</p>\n\n<p>Hanan.</p>",
      "rawMarkdown": "",
      "votes": null
    },
    {
      "id": "57766",
      "postDate": "11/11/2014 11:01:40",
      "content": "<p>I would like to complain that I subscribed to the competition awhile back but I cannot upload my prediction files although it says there are 6.5 days left. This is really not fair as it was not mentioned clearly that I must make my first submission before the team merge date, only subscribe. You defined the November 10th as &quot;deadline to new entry &amp; team merges&quot;. I understood entry as in subscribe to the competition and not submission of a prediction. I worked very hard on this competition and it is not fair I cannot participate.</p>",
      "rawMarkdown": "",
      "votes": null
    },
    {
      "id": "57768",
      "postDate": "11/11/2014 11:09:40",
      "content": "<p>Hi, First submission deadline is in the rules </p>\n<p><br>Competition Timeline<br>Start Date: 8/25/2014 4:15:33 PM UTC<br>Merger Deadline: 11/10/2014 11:59:00 PM UTC<br>First Submission Deadline: 11/10/2014 11:59:00 PM UTC<br>End Date: 11/17/2014 11:59:00 PM UTC</p>\n\n<p>I</p>",
      "rawMarkdown": "",
      "votes": null
    },
    {
      "id": "57769",
      "postDate": "11/11/2014 11:15:46",
      "content": "<p>Yeah. &nbsp;It will happen if you don't read the rules properly!&nbsp;</p>",
      "rawMarkdown": "",
      "votes": null
    },
    {
      "id": "57773",
      "postDate": "11/11/2014 11:37:16",
      "content": "<p>Hi, I've also made the same mistake... I worked so hard but I didn t pay attention regarding the first submission date. So at least I would like to write a scientific paper using this database and data. So I would ask if it is possible to have at the end of competition the &quot;test data&quot; (the correct labelling) in order to obtain the results and write a scientific paper.</p>\n\n<p>Thanks,</p>\n<p>Luca</p>",
      "rawMarkdown": "",
      "votes": null
    },
    {
      "id": "57775",
      "postDate": "11/11/2014 11:47:04",
      "content": "<p>I think the submission will be open again when the competition ends, so you can test your ideas &quot;after&quot; the competition.</p>",
      "rawMarkdown": "",
      "votes": null
    },
    {
      "id": "57783",
      "postDate": "11/11/2014 12:07:33",
      "content": "<p>Well the rules are misleading and make no sense as I could just have submitted an empty set just to get first entry on time. In this sense the gate of first submission is actually meaningless.</p>",
      "rawMarkdown": "",
      "votes": null
    },
    {
      "id": "57784",
      "postDate": "11/11/2014 12:14:02",
      "content": "<p>OK, thanks. In any case I re-read the rules and in fact it is not so clear...only in this short paragraph without explanation there is written:</p>\n<p>Start Date: 8/25/2014 4:15:33 PM UTC<br>Merger Deadline: 11/10/2014 11:59:00 PM UTC<br>First Submission Deadline: 11/10/2014 11:59:00 PM UTC<br>End Date: 11/17/2014 11:59:00 PM UTC</p>\n<p><strong>ANd it is not clear if first submission is related only for who MERGED</strong>.&nbsp;(As you can note this deadline is specified under the merger deadline, which is the same date!!&nbsp;Also the date it is wrong: 11/10/2014 means&nbsp; eleven/october/2014!</p>\n<p>As I never saw a condition like this in any competition, (also because one for the first submission can put random label to say...), I think, maybe at this point for the future, please explain or describe this FOUNDAMENTAL deadline.</p>\n<p>Thanks,</p>\n<p>Luca</p>",
      "rawMarkdown": "",
      "votes": null
    },
    {
      "id": "57791",
      "postDate": "11/11/2014 13:12:08",
      "content": "<p>I agree with Sindi.</p>\n<p>The instructions were not clear.</p>\n<p>It is annoying to work on something for weeks and then miss the deadline.</p>\n<p>See image below: &quot;Deadline for new entry&quot; it seems like entry means subscribing as a contestant as it is in the same line as &quot;team mergers&quot;.</p>\n\n<p><img src=\"http://oi62.tinypic.com/nq49sg.jpg\" alt width=\"770\" height=\"164\"></p>",
      "rawMarkdown": "",
      "votes": null
    },
    {
      "id": "57794",
      "postDate": "11/11/2014 14:01:31",
      "content": "<p>Apparently it's an anti-cheating rule because they had problems with people registering new accounts at the last minute and trying a bunch of submissions with separate accounts. Since it seems like almost all new participants run into this problem, I might suggest that 'entry' should be changed to 'accepting the terms and conditions of the competition.' Both would take about the same amount of effort for an experienced cheater, but the second would be less likely to catch a bunch of newbies scrambling to get their first entry in on time.</p>",
      "rawMarkdown": "",
      "votes": null
    },
    {
      "id": "57796",
      "postDate": "11/11/2014 14:04:36",
      "content": "<p>All you have to do is submit the samplesubmission.csv after accepting the rules.</p>",
      "rawMarkdown": "",
      "votes": null
    },
    {
      "id": "57798",
      "postDate": "11/11/2014 14:37:08",
      "content": "<p>As a data miner with a signal processing background, I found this Epilepsy competition quite interesting, but&nbsp;I've been&nbsp;burned by that persnickety rule too.</p>\n<p>Considering the time and effort&nbsp;I put into coming up with my predictive system, I'm pretty miffed that I'm not allowed to submit it !</p>\n<p>I've been tuning my predictions&nbsp;using my own cross-validation procedures, so as not to create useless load on&nbsp;Kaggle's servers, and only discovered today that I'm not allowed to make my first submission due to some bizarro rule.</p>\n<p>I fail to see the point of that rule, if it's indeed intended to deter cheaters, as nothing prevents cheaters from having registered&nbsp;lots of new accounts before Nov. 11 to make a bunch of submissions.</p>\n<p>As it stands, that arbitrary, obscure deadline traps&nbsp;many users who were considerate for Kaggle's resources, and used their own systems to cross-validate / score their predictions until they were ready to submit their preferred&nbsp;results for Kaggle's evaluation.</p>\n<p>I'd urge Kaggle's competition managers to waive that capricious limitation and actually allow submissions, including a user's first ones, right until the actual closing date of the competition.</p>\n<p>Please ?</p>",
      "rawMarkdown": "",
      "votes": null
    },
    {
      "id": "57800",
      "postDate": "11/11/2014 15:01:14",
      "content": "<p>[quote=Chronognosis;57798]</p>\n<p>I fail to see the point of that rule, if it's indeed intended to deter cheaters, as nothing prevents cheaters from having registered&nbsp;lots of new accounts before Nov. 11 to make a bunch of submissions.</p>\n<p>[/quote]</p>\n<p>The admins have explained this previously. Cheaters tend to be lazy by nature, and don't plan ahead. So, while this doesn't prevent all cheating, it does reduce it.</p>",
      "rawMarkdown": "",
      "votes": null
    },
    {
      "id": "57801",
      "postDate": "11/11/2014 15:09:04",
      "content": "<p>[quote=ACS69;57796]</p>\n<p>All you have to do is submit the samplesubmission.csv after accepting the rules.</p>\n<p>[/quote]</p>\n<p>This is the obvious workaround, but the point is that first time users wont know to do this until they've already been burned by it.</p>",
      "rawMarkdown": "",
      "votes": null
    },
    {
      "id": "57803",
      "postDate": "11/11/2014 15:26:52",
      "content": "<p>I agree with ACS69. Nobody can discuss the rules, kaggle has surely good reasons to put some rules! <strong>But please let the rules be clear</strong>!! Not just a row without meaning or explanation and ambiguous... Every rule is OK, but it is a rule only if it is clear and explained from the beginning.</p>\n<p>This is my personal opinion. In any case strarting from the next competition I know what I have to do, but not because I&nbsp; read it, but because I learned it by practice! (machine Learning !!! ;) )</p>\n\n<p>Bye!!</p>",
      "rawMarkdown": "",
      "votes": null
    },
    {
      "id": "57821",
      "postDate": "11/11/2014 17:28:50",
      "content": "<h2>Corresponding with Kaggle:</h2>\n<p>On Tue, Nov 11, 2014 at 4:48 PM, Jerad Rose &lt;support@kaggle.com&gt; wrote:</p>\n<p>Hi Hanan,</p>\n<p>I understand your frustration, and do apologize for this. There are some known issues with how we communicate the two submission deadlines, and the confusion that these can cause. However, we do not make exceptions to our submission deadlines as these are part of the agreement when joining a competition.</p>\n<p>Having said that, we understand the issues this is causing for our users, and are working on some changes that will make sure these deadlines are clear from the beginning.</p>\n<p>Kind regards,<br>Jerad<br>Kaggle Team</p>\n\n<h3>Answer:</h3>\n<p>Dear Jerad,<br>I believe you are damaging Kaggle name when you hold participants to submit just due to the lack of a &quot;meaningless upload&quot;: one could make a first submission easily by uploading a dummy csv file.<br>I think it is in Kaggle's interest that you could have consider an exception of this special case.<br>I really like Kaggle until this incident.</p>\n<p>HS</p>",
      "rawMarkdown": "",
      "votes": null
    },
    {
      "id": "57822",
      "postDate": "11/11/2014 17:29:12",
      "content": "<p>I&nbsp; faced same problem in my first competition.</p>\n\n<p>Some comments here:</p>\n\n<p>1) It is easy to put a reminder&nbsp; there like:</p>\n<p>Please submit your first submission by XXX,just like please select your 2 best models.</p>\n<p>2) Or change the &quot;deadline for new entry&quot; to deadline for first submission&quot;</p>\n<p>3) Or have a button somewhere to let user join competition, anyway using first submission as entry point is not the best solution.</p>\n<p>3.1) first submission for most of people is the sample_submission.csv, time/cpu/network/storage space wasting for both Kaggle and each individual.</p>\n<p>4)better to have a &quot;watching&quot; function like github</p>",
      "rawMarkdown": "",
      "votes": null
    },
    {
      "id": "57831",
      "postDate": "11/11/2014 18:07:47",
      "content": "<p>[quote=Hanan Shteingart;57821]</p>\n<h2>Corresponding with Kaggle:</h2>\n<p>On Tue, Nov 11, 2014 at 4:48 PM, Jerad Rose &lt;support@kaggle.com&gt; wrote:</p>\n<p>Hi Hanan,</p>\n<p>Having said that, we understand the issues this is causing for our users, and are working on some changes that will make sure these deadlines are clear from the beginning.</p>\n<p>[/quote]</p>\n\n<p>[quote=Steven Du;57822]</p>\n<p>3) Or have a button somewhere to let user join competition, anyway using first submission as entry point is not the best solution.</p>\n<p>[/quote]</p>\n\n<p>It's good they are looking into changes. As to Steven's point 3, as I mentioned earlier, they already have a button. To download the data, you have to agree to the terms of the competition (data privacy etc.). Why not just make this the entry point?</p>",
      "rawMarkdown": "",
      "votes": null
    },
    {
      "id": "57834",
      "postDate": "11/11/2014 18:27:19",
      "content": "<p>Shoot! &nbsp;I made the same mistake too. &nbsp;I was just about to submit my first prediction. &nbsp;</p>",
      "rawMarkdown": "",
      "votes": null
    },
    {
      "id": "57869",
      "postDate": "11/12/2014 03:01:04",
      "content": "<p>If you post your solution csv and source code here, I will make a submission on behalf of your team. If I win using an ensemble which is more than 5% of this solution, I will give you 5% of the prize money. Furthermore as&nbsp;Kaggle competitor tinrtgu has demonstrated, sharing of good code is highly regarded by the Kaggle community regardless of prize money won. Good luck!</p>",
      "rawMarkdown": "",
      "votes": null
    },
    {
      "id": "57883",
      "postDate": "11/12/2014 08:36:22",
      "content": "<p>[quote=Mike Kim;57869]</p>\n<p>If you post your solution csv and source code here, I will make a submission on behalf of your team. If I win using an ensemble which is more than 5% of this solution, I will give you 5% of the prize money. Furthermore as&nbsp;Kaggle competitor tinrtgu has demonstrated, sharing of good code is highly regarded by the Kaggle community regardless of prize money won. Good luck!</p>\n<p>[/quote]</p>\n<p>If the submission deadline has already ended isn't it illegal to use submissions from people who aren't in the competition? Well of course people are free to share code and ideas.</p>",
      "rawMarkdown": "",
      "votes": null
    },
    {
      "id": "57916",
      "postDate": "11/12/2014 15:17:55",
      "content": "<p>Nah, if he posts the csv here, it's basically the same as posting free to use source code. Anyone can use it alone or as part of an ensemble so it's not unfair advantage.</p>",
      "rawMarkdown": "",
      "votes": null
    }
  ],
  "comments": [
    {
      "id": 57766,
      "author_name": "hananshteingart",
      "author_url": "",
      "post_date": "11/11/2014 11:01:40",
      "content": "<p>I would like to complain that I subscribed to the competition awhile back but I cannot upload my prediction files although it says there are 6.5 days left. This is really not fair as it was not mentioned clearly that I must make my first submission before the team merge date, only subscribe. You defined the November 10th as &quot;deadline to new entry &amp; team merges&quot;. I understood entry as in subscribe to the competition and not submission of a prediction. I worked very hard on this competition and it is not fair I cannot participate.</p>",
      "votes": null,
      "replies": []
    },
    {
      "id": 57768,
      "author_name": "domcastro",
      "author_url": "",
      "post_date": "11/11/2014 11:09:40",
      "content": "<p>Hi, First submission deadline is in the rules </p>\n<p><br>Competition Timeline<br>Start Date: 8/25/2014 4:15:33 PM UTC<br>Merger Deadline: 11/10/2014 11:59:00 PM UTC<br>First Submission Deadline: 11/10/2014 11:59:00 PM UTC<br>End Date: 11/17/2014 11:59:00 PM UTC</p>\n\n<p>I</p>",
      "votes": null,
      "replies": []
    },
    {
      "id": 57769,
      "author_name": "abhishek",
      "author_url": "",
      "post_date": "11/11/2014 11:15:46",
      "content": "<p>Yeah. &nbsp;It will happen if you don't read the rules properly!&nbsp;</p>",
      "votes": null,
      "replies": []
    },
    {
      "id": 57773,
      "author_name": "sindi112709",
      "author_url": "",
      "post_date": "11/11/2014 11:37:16",
      "content": "<p>Hi, I've also made the same mistake... I worked so hard but I didn t pay attention regarding the first submission date. So at least I would like to write a scientific paper using this database and data. So I would ask if it is possible to have at the end of competition the &quot;test data&quot; (the correct labelling) in order to obtain the results and write a scientific paper.</p>\n\n<p>Thanks,</p>\n<p>Luca</p>",
      "votes": null,
      "replies": []
    },
    {
      "id": 57775,
      "author_name": "pakozm",
      "author_url": "",
      "post_date": "11/11/2014 11:47:04",
      "content": "<p>I think the submission will be open again when the competition ends, so you can test your ideas &quot;after&quot; the competition.</p>",
      "votes": null,
      "replies": []
    },
    {
      "id": 57783,
      "author_name": "hananshteingart",
      "author_url": "",
      "post_date": "11/11/2014 12:07:33",
      "content": "<p>Well the rules are misleading and make no sense as I could just have submitted an empty set just to get first entry on time. In this sense the gate of first submission is actually meaningless.</p>",
      "votes": null,
      "replies": []
    },
    {
      "id": 57784,
      "author_name": "sindi112709",
      "author_url": "",
      "post_date": "11/11/2014 12:14:02",
      "content": "<p>OK, thanks. In any case I re-read the rules and in fact it is not so clear...only in this short paragraph without explanation there is written:</p>\n<p>Start Date: 8/25/2014 4:15:33 PM UTC<br>Merger Deadline: 11/10/2014 11:59:00 PM UTC<br>First Submission Deadline: 11/10/2014 11:59:00 PM UTC<br>End Date: 11/17/2014 11:59:00 PM UTC</p>\n<p><strong>ANd it is not clear if first submission is related only for who MERGED</strong>.&nbsp;(As you can note this deadline is specified under the merger deadline, which is the same date!!&nbsp;Also the date it is wrong: 11/10/2014 means&nbsp; eleven/october/2014!</p>\n<p>As I never saw a condition like this in any competition, (also because one for the first submission can put random label to say...), I think, maybe at this point for the future, please explain or describe this FOUNDAMENTAL deadline.</p>\n<p>Thanks,</p>\n<p>Luca</p>",
      "votes": null,
      "replies": []
    },
    {
      "id": 57791,
      "author_name": "hananshteingart",
      "author_url": "",
      "post_date": "11/11/2014 13:12:08",
      "content": "<p>I agree with Sindi.</p>\n<p>The instructions were not clear.</p>\n<p>It is annoying to work on something for weeks and then miss the deadline.</p>\n<p>See image below: &quot;Deadline for new entry&quot; it seems like entry means subscribing as a contestant as it is in the same line as &quot;team mergers&quot;.</p>\n\n<p><img src=\"http://oi62.tinypic.com/nq49sg.jpg\" alt width=\"770\" height=\"164\"></p>",
      "votes": null,
      "replies": []
    },
    {
      "id": 57794,
      "author_name": "piinformatics",
      "author_url": "",
      "post_date": "11/11/2014 14:01:31",
      "content": "<p>Apparently it's an anti-cheating rule because they had problems with people registering new accounts at the last minute and trying a bunch of submissions with separate accounts. Since it seems like almost all new participants run into this problem, I might suggest that 'entry' should be changed to 'accepting the terms and conditions of the competition.' Both would take about the same amount of effort for an experienced cheater, but the second would be less likely to catch a bunch of newbies scrambling to get their first entry in on time.</p>",
      "votes": null,
      "replies": []
    },
    {
      "id": 57796,
      "author_name": "domcastro",
      "author_url": "",
      "post_date": "11/11/2014 14:04:36",
      "content": "<p>All you have to do is submit the samplesubmission.csv after accepting the rules.</p>",
      "votes": null,
      "replies": []
    },
    {
      "id": 57798,
      "author_name": "chronognosis",
      "author_url": "",
      "post_date": "11/11/2014 14:37:08",
      "content": "<p>As a data miner with a signal processing background, I found this Epilepsy competition quite interesting, but&nbsp;I've been&nbsp;burned by that persnickety rule too.</p>\n<p>Considering the time and effort&nbsp;I put into coming up with my predictive system, I'm pretty miffed that I'm not allowed to submit it !</p>\n<p>I've been tuning my predictions&nbsp;using my own cross-validation procedures, so as not to create useless load on&nbsp;Kaggle's servers, and only discovered today that I'm not allowed to make my first submission due to some bizarro rule.</p>\n<p>I fail to see the point of that rule, if it's indeed intended to deter cheaters, as nothing prevents cheaters from having registered&nbsp;lots of new accounts before Nov. 11 to make a bunch of submissions.</p>\n<p>As it stands, that arbitrary, obscure deadline traps&nbsp;many users who were considerate for Kaggle's resources, and used their own systems to cross-validate / score their predictions until they were ready to submit their preferred&nbsp;results for Kaggle's evaluation.</p>\n<p>I'd urge Kaggle's competition managers to waive that capricious limitation and actually allow submissions, including a user's first ones, right until the actual closing date of the competition.</p>\n<p>Please ?</p>",
      "votes": null,
      "replies": []
    },
    {
      "id": 57800,
      "author_name": "inversion",
      "author_url": "",
      "post_date": "11/11/2014 15:01:14",
      "content": "<p>[quote=Chronognosis;57798]</p>\n<p>I fail to see the point of that rule, if it's indeed intended to deter cheaters, as nothing prevents cheaters from having registered&nbsp;lots of new accounts before Nov. 11 to make a bunch of submissions.</p>\n<p>[/quote]</p>\n<p>The admins have explained this previously. Cheaters tend to be lazy by nature, and don't plan ahead. So, while this doesn't prevent all cheating, it does reduce it.</p>",
      "votes": null,
      "replies": []
    },
    {
      "id": 57801,
      "author_name": "piinformatics",
      "author_url": "",
      "post_date": "11/11/2014 15:09:04",
      "content": "<p>[quote=ACS69;57796]</p>\n<p>All you have to do is submit the samplesubmission.csv after accepting the rules.</p>\n<p>[/quote]</p>\n<p>This is the obvious workaround, but the point is that first time users wont know to do this until they've already been burned by it.</p>",
      "votes": null,
      "replies": []
    },
    {
      "id": 57803,
      "author_name": "sindi112709",
      "author_url": "",
      "post_date": "11/11/2014 15:26:52",
      "content": "<p>I agree with ACS69. Nobody can discuss the rules, kaggle has surely good reasons to put some rules! <strong>But please let the rules be clear</strong>!! Not just a row without meaning or explanation and ambiguous... Every rule is OK, but it is a rule only if it is clear and explained from the beginning.</p>\n<p>This is my personal opinion. In any case strarting from the next competition I know what I have to do, but not because I&nbsp; read it, but because I learned it by practice! (machine Learning !!! ;) )</p>\n\n<p>Bye!!</p>",
      "votes": null,
      "replies": []
    },
    {
      "id": 57821,
      "author_name": "hananshteingart",
      "author_url": "",
      "post_date": "11/11/2014 17:28:50",
      "content": "<h2>Corresponding with Kaggle:</h2>\n<p>On Tue, Nov 11, 2014 at 4:48 PM, Jerad Rose &lt;support@kaggle.com&gt; wrote:</p>\n<p>Hi Hanan,</p>\n<p>I understand your frustration, and do apologize for this. There are some known issues with how we communicate the two submission deadlines, and the confusion that these can cause. However, we do not make exceptions to our submission deadlines as these are part of the agreement when joining a competition.</p>\n<p>Having said that, we understand the issues this is causing for our users, and are working on some changes that will make sure these deadlines are clear from the beginning.</p>\n<p>Kind regards,<br>Jerad<br>Kaggle Team</p>\n\n<h3>Answer:</h3>\n<p>Dear Jerad,<br>I believe you are damaging Kaggle name when you hold participants to submit just due to the lack of a &quot;meaningless upload&quot;: one could make a first submission easily by uploading a dummy csv file.<br>I think it is in Kaggle's interest that you could have consider an exception of this special case.<br>I really like Kaggle until this incident.</p>\n<p>HS</p>",
      "votes": null,
      "replies": []
    },
    {
      "id": 57822,
      "author_name": "stevendu",
      "author_url": "",
      "post_date": "11/11/2014 17:29:12",
      "content": "<p>I&nbsp; faced same problem in my first competition.</p>\n\n<p>Some comments here:</p>\n\n<p>1) It is easy to put a reminder&nbsp; there like:</p>\n<p>Please submit your first submission by XXX,just like please select your 2 best models.</p>\n<p>2) Or change the &quot;deadline for new entry&quot; to deadline for first submission&quot;</p>\n<p>3) Or have a button somewhere to let user join competition, anyway using first submission as entry point is not the best solution.</p>\n<p>3.1) first submission for most of people is the sample_submission.csv, time/cpu/network/storage space wasting for both Kaggle and each individual.</p>\n<p>4)better to have a &quot;watching&quot; function like github</p>",
      "votes": null,
      "replies": []
    },
    {
      "id": 57831,
      "author_name": "piinformatics",
      "author_url": "",
      "post_date": "11/11/2014 18:07:47",
      "content": "<p>[quote=Hanan Shteingart;57821]</p>\n<h2>Corresponding with Kaggle:</h2>\n<p>On Tue, Nov 11, 2014 at 4:48 PM, Jerad Rose &lt;support@kaggle.com&gt; wrote:</p>\n<p>Hi Hanan,</p>\n<p>Having said that, we understand the issues this is causing for our users, and are working on some changes that will make sure these deadlines are clear from the beginning.</p>\n<p>[/quote]</p>\n\n<p>[quote=Steven Du;57822]</p>\n<p>3) Or have a button somewhere to let user join competition, anyway using first submission as entry point is not the best solution.</p>\n<p>[/quote]</p>\n\n<p>It's good they are looking into changes. As to Steven's point 3, as I mentioned earlier, they already have a button. To download the data, you have to agree to the terms of the competition (data privacy etc.). Why not just make this the entry point?</p>",
      "votes": null,
      "replies": []
    },
    {
      "id": 57834,
      "author_name": "raystorm0",
      "author_url": "",
      "post_date": "11/11/2014 18:27:19",
      "content": "<p>Shoot! &nbsp;I made the same mistake too. &nbsp;I was just about to submit my first prediction. &nbsp;</p>",
      "votes": null,
      "replies": []
    },
    {
      "id": 57869,
      "author_name": "mikeskim",
      "author_url": "",
      "post_date": "11/12/2014 03:01:04",
      "content": "<p>If you post your solution csv and source code here, I will make a submission on behalf of your team. If I win using an ensemble which is more than 5% of this solution, I will give you 5% of the prize money. Furthermore as&nbsp;Kaggle competitor tinrtgu has demonstrated, sharing of good code is highly regarded by the Kaggle community regardless of prize money won. Good luck!</p>",
      "votes": null,
      "replies": []
    },
    {
      "id": 57883,
      "author_name": "scatterbrain333",
      "author_url": "",
      "post_date": "11/12/2014 08:36:22",
      "content": "<p>[quote=Mike Kim;57869]</p>\n<p>If you post your solution csv and source code here, I will make a submission on behalf of your team. If I win using an ensemble which is more than 5% of this solution, I will give you 5% of the prize money. Furthermore as&nbsp;Kaggle competitor tinrtgu has demonstrated, sharing of good code is highly regarded by the Kaggle community regardless of prize money won. Good luck!</p>\n<p>[/quote]</p>\n<p>If the submission deadline has already ended isn't it illegal to use submissions from people who aren't in the competition? Well of course people are free to share code and ideas.</p>",
      "votes": null,
      "replies": []
    },
    {
      "id": 57916,
      "author_name": "piinformatics",
      "author_url": "",
      "post_date": "11/12/2014 15:17:55",
      "content": "<p>Nah, if he posts the csv here, it's basically the same as posting free to use source code. Anyone can use it alone or as part of an ensemble so it's not unfair advantage.</p>",
      "votes": null,
      "replies": []
    }
  ],
  "raw_markdown_by_id": {
    "57765": "",
    "57766": "",
    "57768": "",
    "57769": "",
    "57773": "",
    "57775": "",
    "57783": "",
    "57784": "",
    "57791": "",
    "57794": "",
    "57796": "",
    "57798": "",
    "57800": "",
    "57801": "",
    "57803": "",
    "57821": "",
    "57822": "",
    "57831": "",
    "57834": "",
    "57869": "",
    "57883": "",
    "57916": ""
  },
  "source": "meta"
}