We just got our team together and were about to make our first submission, but didn't realize the deadline was in UTC. Is there any way can exception can be made to get us in?
", "messages": [ { "id": "90276", "postDate": "08/25/2015 01:38:26", "content": "We just got our team together and were about to make our first submission, but didn't realize the deadline was in UTC. Is there any way can exception can be made to get us in?
", "rawMarkdown": "We just got our team together and were about to make our first submission, but didn't realize the deadline was in UTC. Is there any way can exception can be made to get us in?", "votes": null }, { "id": "90328", "postDate": "08/25/2015 14:03:49", "content": "I missed the 1st submission deadline too. It is such a shame, because I spent so much time on developing and validating my models locally. I would not like all these efforts to be wasted. I would really appreciate if you could let me in.
", "rawMarkdown": "I missed the 1st submission deadline too. It is such a shame, because I spent so much time on developing and validating my models locally. I would not like all these efforts to be wasted. I would really appreciate if you could let me in.", "votes": null }, { "id": "90330", "postDate": "08/25/2015 14:15:53", "content": "I didn't even know there's a 1st submission deadline earlier than the deadline. Should have read the rules more carefully.
", "rawMarkdown": "I didn't even know there's a 1st submission deadline earlier than the deadline. Should have read the rules more carefully.", "votes": null }, { "id": "90377", "postDate": "08/25/2015 23:01:18", "content": "I totally made this mistake before in other Kaggle competitions. (the Higgs Boson one.) Kaggle does a poor job making this point clear, and frankly, I'm not sure how much the competition experience is improved as a result of forcing this "deadline-before-the-deadline." I think it turns a lot of people off from Kaggle because it seems too finicky. Sorry guys. I feel for you. I wish Kaggle just did away with this mechanism.
", "rawMarkdown": "I totally made this mistake before in other Kaggle competitions. (the Higgs Boson one.) Kaggle does a poor job making this point clear, and frankly, I'm not sure how much the competition experience is improved as a result of forcing this \"deadline-before-the-deadline.\" I think it turns a lot of people off from Kaggle because it seems too finicky. Sorry guys. I feel for you. I wish Kaggle just did away with this mechanism.", "votes": null }, { "id": "90381", "postDate": "08/25/2015 23:13:33", "content": "[quote=JAR1986;90377]
\n\nI totally made this mistake before in other Kaggle competitions. (the Higgs Boson one.) Kaggle does a poor job making this point clear, and frankly, I'm not sure how much the competition experience is improved as a result of forcing this "deadline-before-the-deadline." I think it turns a lot of people off from Kaggle because it seems too finicky. Sorry guys. I feel for you. I wish Kaggle just did away with this mechanism.
\n\n[/quote]
\n\nIt's funny how people who say this sort of stuff have not competed in many competitions. This rule really helped reduce the amount of chaos and cheating that happened in the last week as panic led to temptation. I think most people who competed regularly before the rule would not want to go back.
\n\nAs data scientist, the first rule of business should be understand the problem definition. That includes understanding the rules and constraints you have to live within.
", "rawMarkdown": "[quote=JAR1986;90377]\r\n\r\nI totally made this mistake before in other Kaggle competitions. (the Higgs Boson one.) Kaggle does a poor job making this point clear, and frankly, I'm not sure how much the competition experience is improved as a result of forcing this \"deadline-before-the-deadline.\" I think it turns a lot of people off from Kaggle because it seems too finicky. Sorry guys. I feel for you. I wish Kaggle just did away with this mechanism.\r\n\r\n[/quote]\r\n\r\nIt's funny how people who say this sort of stuff have not competed in many competitions. This rule really helped reduce the amount of chaos and cheating that happened in the last week as panic led to temptation. I think most people who competed regularly before the rule would not want to go back.\r\n\r\nAs data scientist, the first rule of business should be understand the problem definition. That includes understanding the rules and constraints you have to live within.", "votes": null }, { "id": "90408", "postDate": "08/26/2015 08:59:58", "content": "Hello Neil,
\n\nI respectfully take your point that you have found the policy to be valuable, and I will retract my statement that it should be undone. Here, I will elaborate on the second point that I made.
\n\nThe second point that I made was simply that the double-deadline system, as it is currently communicated, can be counter-intuitive to new competitors. As an example, 2 of the top 4 competitions that are listed as "active" on the site can not be entered by a new user, as of today, because they are each past the first submission deadline. When you click on those competitions, you can see a time bar at the top that indicates that there are still 'X' days remaining before the deadline, even though there is no option for new entrants to make a submission.
\n\nWhen newer participants misunderstand the double-deadline system, it doesn't necessarily mean that they are idiots. It likely means that they got excited about digging into the data and thought they had some time before they had to worry about the rest of the competition details because the top of the competition page said they still had 'X' days left. (When in reality, they had (X-7) days left.)
\n\nThe fact remains that people keep getting confused about this simple matter, and I am not in the camp that believes it is because they are dumb. I agree with you that "As data scientist, the first rule of business should be understand the problem definition..." However, why should we make this hard when we can make this easy?
\n\nSome possible remedies include (1) simply denoting on the time bar the date of the 1st submission deadline in addition to the (already-displayed) date of the final deadline and/or (2) a statement somewhere at the top of the competition page (without having to click on a link) stating that the competition "closes to new entrants in X days" or is "currently closed to new entrants".
\n\nI know that if one or both of these things had been implemented, I would not have made the mistake that I made a year ago. I suspect it would clear up the confusion that others are having also. I understand you may think this is excessive hand-holding, but I would prefer to compete against more people than fewer.
\n\nThank you for replying to me.
\n\nJames
", "rawMarkdown": "Hello Neil,\r\n\r\nI respectfully take your point that you have found the policy to be valuable, and I will retract my statement that it should be undone. Here, I will elaborate on the second point that I made.\r\n\r\nThe second point that I made was simply that the double-deadline system, as it is currently communicated, can be counter-intuitive to new competitors. As an example, 2 of the top 4 competitions that are listed as \"active\" on the site can not be entered by a new user, as of today, because they are each past the first submission deadline. When you click on those competitions, you can see a time bar at the top that indicates that there are still 'X' days remaining before the deadline, even though there is no option for new entrants to make a submission.\r\n\r\nWhen newer participants misunderstand the double-deadline system, it doesn't necessarily mean that they are idiots. It likely means that they got excited about digging into the data and thought they had some time before they had to worry about the rest of the competition details because the top of the competition page said they still had 'X' days left. (When in reality, they had (X-7) days left.)\r\n\r\nThe fact remains that people keep getting confused about this simple matter, and I am not in the camp that believes it is because they are dumb. I agree with you that \"As data scientist, the first rule of business should be understand the problem definition...\" However, why should we make this hard when we can make this easy?\r\n\r\nSome possible remedies include (1) simply denoting on the time bar the date of the 1st submission deadline in addition to the (already-displayed) date of the final deadline and/or (2) a statement somewhere at the top of the competition page (without having to click on a link) stating that the competition \"closes to new entrants in X days\" or is \"currently closed to new entrants\". \r\n\r\nI know that if one or both of these things had been implemented, I would not have made the mistake that I made a year ago. I suspect it would clear up the confusion that others are having also. I understand you may think this is excessive hand-holding, but I would prefer to compete against more people than fewer.\r\n\r\nThank you for replying to me.\r\n\r\nJames", "votes": null }, { "id": "90410", "postDate": "08/26/2015 09:12:28", "content": "I was just asking for a small exception because I didn't notice the "UTC" aspect of it. I was more or less about 2 hours too late because of it. No big deal, though. I'll play around with the data and just say that I participated :)
", "rawMarkdown": "I was just asking for a small exception because I didn't notice the \"UTC\" aspect of it. I was more or less about 2 hours too late because of it. No big deal, though. I'll play around with the data and just say that I participated :)", "votes": null }, { "id": "90411", "postDate": "08/26/2015 09:18:36", "content": "In my case it's my fault because I didn't read the rules with enough attention. If I had submitted something it would have been for the experience and not because I was likely to win, I'm just trying to learn how to analyse data and it's not something I do full-time (far from it).\nBut I admire people who stand up for newbies, and it would be good to highlight the different deadlines (although it's unlikely others will make the same mistake I made).
", "rawMarkdown": "In my case it's my fault because I didn't read the rules with enough attention. If I had submitted something it would have been for the experience and not because I was likely to win, I'm just trying to learn how to analyse data and it's not something I do full-time (far from it).\r\nBut I admire people who stand up for newbies, and it would be good to highlight the different deadlines (although it's unlikely others will make the same mistake I made).", "votes": null }, { "id": "90428", "postDate": "08/26/2015 13:46:42", "content": "@JAR1986 Yes I agree that the policy is still not communicated well from a user experience perspective on the website.
\n\n@ZhangBanger Yes it harsh, and from a single user perspective it is nice to think that they could make an exception. But you have to consider the competition as a whole, if they started to make an exception for individual cases then it would open the flood gates and they would be getting requests all the time. Where do you draw the line - easiest on the rule itself.
\n\nFor all those that missed the deadline, you will still be able to see how well your submission would have fared. After the final deadline the submission process opens up again and you will get feedback on your submission on what the leaderboard would have scored, but you will just not be ranked on the leaderboard itself.
", "rawMarkdown": "JAR1986 Yes I agree that the policy is still not communicated well from a user experience perspective on the website.\r\n\r\n@ZhangBanger Yes it harsh, and from a single user perspective it is nice to think that they could make an exception. But you have to consider the competition as a whole, if they started to make an exception for individual cases then it would open the flood gates and they would be getting requests all the time. Where do you draw the line - easiest on the rule itself.\r\n\r\nFor all those that missed the deadline, you will still be able to see how well your submission would have fared. After the final deadline the submission process opens up again and you will get feedback on your submission on what the leaderboard would have scored, but you will just not be ranked on the leaderboard itself.", "votes": null }, { "id": "90440", "postDate": "08/26/2015 15:42:19", "content": "We hate to see people miss the first submission deadline. What would make it clearer? We currently mention the deadline in the rules, the timeline page, and on the graphical timeline at the top of the competition pages. We plan to implement a reminder email to people who have accepted rules but not submitted. Think that is sufficient?
\n\nNo amount of flashing lights and blaring sirens will eliminate this thread (competitions have thousands of participants, so someone's dog is always going to eat their homework), but we do want to meet you half way.
", "rawMarkdown": "We hate to see people miss the first submission deadline. What would make it clearer? We currently mention the deadline in the rules, the timeline page, and on the graphical timeline at the top of the competition pages. We plan to implement a reminder email to people who have accepted rules but not submitted. Think that is sufficient?\r\n\r\nNo amount of flashing lights and blaring sirens will eliminate this thread (competitions have thousands of participants, so someone's dog is always going to eat their homework), but we do want to meet you half way.", "votes": null }, { "id": "90443", "postDate": "08/26/2015 15:56:52", "content": "@Will\nYes, you guys do a lot, and you are right that no matter how much you do, someone will always miss the deadline for whatever reason. How about a competition to predict which entrants will miss the deadline. You must have a good training set on that by now!
\n\nAll kidding aside, quick couple of things to make the timeline bar at the top clearer.
\n\n1) A countdown timer on the first submission deadline too.
\n\n2) Making the text of the first submission deadline right justified to the point 1 week from the end. That way it is clearer that it corresponds to the earlier point. First glance is not that clear that it refers to the earlier point.
\n\n3) Adding midnight UTC on the end times, as it is not really the same day for everyone, so that can be confusing.
\n\nI think the reminder email will probably help a lot.
", "rawMarkdown": "Will\r\nYes, you guys do a lot, and you are right that no matter how much you do, someone will always miss the deadline for whatever reason. How about a competition to predict which entrants will miss the deadline. You must have a good training set on that by now!\r\n\r\nAll kidding aside, quick couple of things to make the timeline bar at the top clearer.\r\n\r\n1) A countdown timer on the first submission deadline too.\r\n\r\n2) Making the text of the first submission deadline right justified to the point 1 week from the end. That way it is clearer that it corresponds to the earlier point. First glance is not that clear that it refers to the earlier point.\r\n\r\n3) Adding midnight UTC on the end times, as it is not really the same day for everyone, so that can be confusing.\r\n\r\nI think the reminder email will probably help a lot.", "votes": null } ], "comments": [ { "id": 90328, "author_name": "aydarkhan", "author_url": "", "post_date": "08/25/2015 14:03:49", "content": "I missed the 1st submission deadline too. It is such a shame, because I spent so much time on developing and validating my models locally. I would not like all these efforts to be wasted. I would really appreciate if you could let me in.
", "votes": null, "replies": [] }, { "id": 90330, "author_name": "ana4min", "author_url": "", "post_date": "08/25/2015 14:15:53", "content": "I didn't even know there's a 1st submission deadline earlier than the deadline. Should have read the rules more carefully.
", "votes": null, "replies": [] }, { "id": 90377, "author_name": "jarolfsen", "author_url": "", "post_date": "08/25/2015 23:01:18", "content": "I totally made this mistake before in other Kaggle competitions. (the Higgs Boson one.) Kaggle does a poor job making this point clear, and frankly, I'm not sure how much the competition experience is improved as a result of forcing this "deadline-before-the-deadline." I think it turns a lot of people off from Kaggle because it seems too finicky. Sorry guys. I feel for you. I wish Kaggle just did away with this mechanism.
", "votes": null, "replies": [] }, { "id": 90381, "author_name": "neilsummers", "author_url": "", "post_date": "08/25/2015 23:13:33", "content": "[quote=JAR1986;90377]
\n\nI totally made this mistake before in other Kaggle competitions. (the Higgs Boson one.) Kaggle does a poor job making this point clear, and frankly, I'm not sure how much the competition experience is improved as a result of forcing this "deadline-before-the-deadline." I think it turns a lot of people off from Kaggle because it seems too finicky. Sorry guys. I feel for you. I wish Kaggle just did away with this mechanism.
\n\n[/quote]
\n\nIt's funny how people who say this sort of stuff have not competed in many competitions. This rule really helped reduce the amount of chaos and cheating that happened in the last week as panic led to temptation. I think most people who competed regularly before the rule would not want to go back.
\n\nAs data scientist, the first rule of business should be understand the problem definition. That includes understanding the rules and constraints you have to live within.
", "votes": null, "replies": [] }, { "id": 90408, "author_name": "jarolfsen", "author_url": "", "post_date": "08/26/2015 08:59:58", "content": "Hello Neil,
\n\nI respectfully take your point that you have found the policy to be valuable, and I will retract my statement that it should be undone. Here, I will elaborate on the second point that I made.
\n\nThe second point that I made was simply that the double-deadline system, as it is currently communicated, can be counter-intuitive to new competitors. As an example, 2 of the top 4 competitions that are listed as "active" on the site can not be entered by a new user, as of today, because they are each past the first submission deadline. When you click on those competitions, you can see a time bar at the top that indicates that there are still 'X' days remaining before the deadline, even though there is no option for new entrants to make a submission.
\n\nWhen newer participants misunderstand the double-deadline system, it doesn't necessarily mean that they are idiots. It likely means that they got excited about digging into the data and thought they had some time before they had to worry about the rest of the competition details because the top of the competition page said they still had 'X' days left. (When in reality, they had (X-7) days left.)
\n\nThe fact remains that people keep getting confused about this simple matter, and I am not in the camp that believes it is because they are dumb. I agree with you that "As data scientist, the first rule of business should be understand the problem definition..." However, why should we make this hard when we can make this easy?
\n\nSome possible remedies include (1) simply denoting on the time bar the date of the 1st submission deadline in addition to the (already-displayed) date of the final deadline and/or (2) a statement somewhere at the top of the competition page (without having to click on a link) stating that the competition "closes to new entrants in X days" or is "currently closed to new entrants".
\n\nI know that if one or both of these things had been implemented, I would not have made the mistake that I made a year ago. I suspect it would clear up the confusion that others are having also. I understand you may think this is excessive hand-holding, but I would prefer to compete against more people than fewer.
\n\nThank you for replying to me.
\n\nJames
", "votes": null, "replies": [] }, { "id": 90410, "author_name": "", "author_url": "", "post_date": "08/26/2015 09:12:28", "content": "I was just asking for a small exception because I didn't notice the "UTC" aspect of it. I was more or less about 2 hours too late because of it. No big deal, though. I'll play around with the data and just say that I participated :)
", "votes": null, "replies": [] }, { "id": 90411, "author_name": "ana4min", "author_url": "", "post_date": "08/26/2015 09:18:36", "content": "In my case it's my fault because I didn't read the rules with enough attention. If I had submitted something it would have been for the experience and not because I was likely to win, I'm just trying to learn how to analyse data and it's not something I do full-time (far from it).\nBut I admire people who stand up for newbies, and it would be good to highlight the different deadlines (although it's unlikely others will make the same mistake I made).
", "votes": null, "replies": [] }, { "id": 90428, "author_name": "neilsummers", "author_url": "", "post_date": "08/26/2015 13:46:42", "content": "@JAR1986 Yes I agree that the policy is still not communicated well from a user experience perspective on the website.
\n\n@ZhangBanger Yes it harsh, and from a single user perspective it is nice to think that they could make an exception. But you have to consider the competition as a whole, if they started to make an exception for individual cases then it would open the flood gates and they would be getting requests all the time. Where do you draw the line - easiest on the rule itself.
\n\nFor all those that missed the deadline, you will still be able to see how well your submission would have fared. After the final deadline the submission process opens up again and you will get feedback on your submission on what the leaderboard would have scored, but you will just not be ranked on the leaderboard itself.
", "votes": null, "replies": [] }, { "id": 90440, "author_name": "wcukierski", "author_url": "", "post_date": "08/26/2015 15:42:19", "content": "We hate to see people miss the first submission deadline. What would make it clearer? We currently mention the deadline in the rules, the timeline page, and on the graphical timeline at the top of the competition pages. We plan to implement a reminder email to people who have accepted rules but not submitted. Think that is sufficient?
\n\nNo amount of flashing lights and blaring sirens will eliminate this thread (competitions have thousands of participants, so someone's dog is always going to eat their homework), but we do want to meet you half way.
", "votes": null, "replies": [] }, { "id": 90443, "author_name": "neilsummers", "author_url": "", "post_date": "08/26/2015 15:56:52", "content": "@Will\nYes, you guys do a lot, and you are right that no matter how much you do, someone will always miss the deadline for whatever reason. How about a competition to predict which entrants will miss the deadline. You must have a good training set on that by now!
\n\nAll kidding aside, quick couple of things to make the timeline bar at the top clearer.
\n\n1) A countdown timer on the first submission deadline too.
\n\n2) Making the text of the first submission deadline right justified to the point 1 week from the end. That way it is clearer that it corresponds to the earlier point. First glance is not that clear that it refers to the earlier point.
\n\n3) Adding midnight UTC on the end times, as it is not really the same day for everyone, so that can be confusing.
\n\nI think the reminder email will probably help a lot.
", "votes": null, "replies": [] } ], "raw_markdown_by_id": { "90276": "We just got our team together and were about to make our first submission, but didn't realize the deadline was in UTC. Is there any way can exception can be made to get us in?", "90328": "I missed the 1st submission deadline too. It is such a shame, because I spent so much time on developing and validating my models locally. I would not like all these efforts to be wasted. I would really appreciate if you could let me in.", "90330": "I didn't even know there's a 1st submission deadline earlier than the deadline. Should have read the rules more carefully.", "90377": "I totally made this mistake before in other Kaggle competitions. (the Higgs Boson one.) Kaggle does a poor job making this point clear, and frankly, I'm not sure how much the competition experience is improved as a result of forcing this \"deadline-before-the-deadline.\" I think it turns a lot of people off from Kaggle because it seems too finicky. Sorry guys. I feel for you. I wish Kaggle just did away with this mechanism.", "90381": "[quote=JAR1986;90377]\r\n\r\nI totally made this mistake before in other Kaggle competitions. (the Higgs Boson one.) Kaggle does a poor job making this point clear, and frankly, I'm not sure how much the competition experience is improved as a result of forcing this \"deadline-before-the-deadline.\" I think it turns a lot of people off from Kaggle because it seems too finicky. Sorry guys. I feel for you. I wish Kaggle just did away with this mechanism.\r\n\r\n[/quote]\r\n\r\nIt's funny how people who say this sort of stuff have not competed in many competitions. This rule really helped reduce the amount of chaos and cheating that happened in the last week as panic led to temptation. I think most people who competed regularly before the rule would not want to go back.\r\n\r\nAs data scientist, the first rule of business should be understand the problem definition. That includes understanding the rules and constraints you have to live within.", "90408": "Hello Neil,\r\n\r\nI respectfully take your point that you have found the policy to be valuable, and I will retract my statement that it should be undone. Here, I will elaborate on the second point that I made.\r\n\r\nThe second point that I made was simply that the double-deadline system, as it is currently communicated, can be counter-intuitive to new competitors. As an example, 2 of the top 4 competitions that are listed as \"active\" on the site can not be entered by a new user, as of today, because they are each past the first submission deadline. When you click on those competitions, you can see a time bar at the top that indicates that there are still 'X' days remaining before the deadline, even though there is no option for new entrants to make a submission.\r\n\r\nWhen newer participants misunderstand the double-deadline system, it doesn't necessarily mean that they are idiots. It likely means that they got excited about digging into the data and thought they had some time before they had to worry about the rest of the competition details because the top of the competition page said they still had 'X' days left. (When in reality, they had (X-7) days left.)\r\n\r\nThe fact remains that people keep getting confused about this simple matter, and I am not in the camp that believes it is because they are dumb. I agree with you that \"As data scientist, the first rule of business should be understand the problem definition...\" However, why should we make this hard when we can make this easy?\r\n\r\nSome possible remedies include (1) simply denoting on the time bar the date of the 1st submission deadline in addition to the (already-displayed) date of the final deadline and/or (2) a statement somewhere at the top of the competition page (without having to click on a link) stating that the competition \"closes to new entrants in X days\" or is \"currently closed to new entrants\". \r\n\r\nI know that if one or both of these things had been implemented, I would not have made the mistake that I made a year ago. I suspect it would clear up the confusion that others are having also. I understand you may think this is excessive hand-holding, but I would prefer to compete against more people than fewer.\r\n\r\nThank you for replying to me.\r\n\r\nJames", "90410": "I was just asking for a small exception because I didn't notice the \"UTC\" aspect of it. I was more or less about 2 hours too late because of it. No big deal, though. I'll play around with the data and just say that I participated :)", "90411": "In my case it's my fault because I didn't read the rules with enough attention. If I had submitted something it would have been for the experience and not because I was likely to win, I'm just trying to learn how to analyse data and it's not something I do full-time (far from it).\r\nBut I admire people who stand up for newbies, and it would be good to highlight the different deadlines (although it's unlikely others will make the same mistake I made).", "90428": "JAR1986 Yes I agree that the policy is still not communicated well from a user experience perspective on the website.\r\n\r\n@ZhangBanger Yes it harsh, and from a single user perspective it is nice to think that they could make an exception. But you have to consider the competition as a whole, if they started to make an exception for individual cases then it would open the flood gates and they would be getting requests all the time. Where do you draw the line - easiest on the rule itself.\r\n\r\nFor all those that missed the deadline, you will still be able to see how well your submission would have fared. After the final deadline the submission process opens up again and you will get feedback on your submission on what the leaderboard would have scored, but you will just not be ranked on the leaderboard itself.", "90440": "We hate to see people miss the first submission deadline. What would make it clearer? We currently mention the deadline in the rules, the timeline page, and on the graphical timeline at the top of the competition pages. We plan to implement a reminder email to people who have accepted rules but not submitted. Think that is sufficient?\r\n\r\nNo amount of flashing lights and blaring sirens will eliminate this thread (competitions have thousands of participants, so someone's dog is always going to eat their homework), but we do want to meet you half way.", "90443": "Will\r\nYes, you guys do a lot, and you are right that no matter how much you do, someone will always miss the deadline for whatever reason. How about a competition to predict which entrants will miss the deadline. You must have a good training set on that by now!\r\n\r\nAll kidding aside, quick couple of things to make the timeline bar at the top clearer.\r\n\r\n1) A countdown timer on the first submission deadline too.\r\n\r\n2) Making the text of the first submission deadline right justified to the point 1 week from the end. That way it is clearer that it corresponds to the earlier point. First glance is not that clear that it refers to the earlier point.\r\n\r\n3) Adding midnight UTC on the end times, as it is not really the same day for everyone, so that can be confusing.\r\n\r\nI think the reminder email will probably help a lot." }, "source": "meta" }