Maybe I stayed in academia for too long to grow the habit of crediting everyone who I borrow code from, or at least try to pin down the original source.
\nI complained before
\nhttps://www.kaggle.com/c/riiid-test-answer-prediction/discussion/206217
Yet I feel Kaggle took little effort to protect and nurture the fantastic sharing culture of notebooks.
\nFor example, I happened to click one of the notebooks (because the user has a cat avatar)
\nhttps://www.kaggle.com/firefliesqn/g2net-gpu-newbie
Apparently this person felt the urge to copy but without copy the starter I prepared:
\nhttps://www.kaggle.com/scaomath/g2net-1d-cnn-gem-pool-pytorch-train-inference
This person literally copied (without using the fork function which points to the original notebook), before version 20, this notebook is just an ensembling notebook that takes 10 secs to run.
", "messages": [ { "id": "1521818", "postDate": "09/23/2021 14:51:45", "content": "Maybe I stayed in academia for too long to grow the habit of crediting everyone who I borrow code from, or at least try to pin down the original source.
\nI complained before
\nhttps://www.kaggle.com/c/riiid-test-answer-prediction/discussion/206217
Yet I feel Kaggle took little effort to protect and nurture the fantastic sharing culture of notebooks.
\nFor example, I happened to click one of the notebooks (because the user has a cat avatar)
\nhttps://www.kaggle.com/firefliesqn/g2net-gpu-newbie
Apparently this person felt the urge to copy but without copy the starter I prepared:
\nhttps://www.kaggle.com/scaomath/g2net-1d-cnn-gem-pool-pytorch-train-inference
This person literally copied (without using the fork function which points to the original notebook), before version 20, this notebook is just an ensembling notebook that takes 10 secs to run.
", "rawMarkdown": "Maybe I stayed in academia for too long to grow the habit of crediting everyone who I borrow code from, or at least try to pin down the original source. \n\nI complained before \nhttps://www.kaggle.com/c/riiid-test-answer-prediction/discussion/206217\n\nYet I feel Kaggle took little effort to protect and nurture the fantastic sharing culture of notebooks.\n\nFor example, I happened to click one of the notebooks (because the user has a cat avatar)\nhttps://www.kaggle.com/firefliesqn/g2net-gpu-newbie\n\nApparently this person felt the urge to copy but without copy the starter I prepared:\nhttps://www.kaggle.com/scaomath/g2net-1d-cnn-gem-pool-pytorch-train-inference \n\nThis person literally copied (without using the fork function which points to the original notebook), before version 20, this notebook is just an ensembling notebook that takes 10 secs to run.", "votes": null }, { "id": "1521904", "postDate": "09/23/2021 16:30:26", "content": "I proposed to enable downvotes on notebooks for this kind of useless shallow copy without added value. Feel free to uppote and support if you think it could be useful. I do think that if we were down voting shallow copies then they would quickly disappear.
\nhttps://www.kaggle.com/product-feedback/76394
", "rawMarkdown": "I proposed to enable downvotes on notebooks for this kind of useless shallow copy without added value. Feel free to uppote and support if you think it could be useful. I do think that if we were down voting shallow copies then they would quickly disappear. \n\nhttps://www.kaggle.com/product-feedback/76394", "votes": null }, { "id": "1521919", "postDate": "09/23/2021 16:42:46", "content": "I'm sorry to hear that. Your works are awesome.
\nSo I checked the notebook function 'Report Notebook'. [Plagiarism/copied content that is not meaningfully different]
\nI will report that notebook.
If you look at firefliesqn's other notebooks, it is usually the same thing. Copy a notebook, change the random state, get upvotes. In this one, they copied one of my notebooks: https://www.kaggle.com/firefliesqn/chaiii-qa-starter but then version 2 copies Abhishek's notebook, and version 3 copies someone else's (julian3833) notebook. It also seems like the same group of people are upvoting these notebooks
\nEdit: I've also noticed that firefliesqn frequently changes their profile picture, which I would pessimistically interpret as an attempt to not be recognized. Maybe I'm reading into it too much.
\nI think Kaggle can compare the code in a notebook to other public notebooks and indicate at the top that the work heavily features content from another notebook(s), linking that notebook(s). This could just be an endless rabbit hole and it could be hard to know who is the \"original\" author. Perhaps the comparisons should be limited to other public notebooks within the same competition and not all of Kaggle. Food for thought.
", "rawMarkdown": "If you look at firefliesqn's other notebooks, it is usually the same thing. Copy a notebook, change the random state, get upvotes. In this one, they copied one of my notebooks: https://www.kaggle.com/firefliesqn/chaiii-qa-starter but then version 2 copies Abhishek's notebook, and version 3 copies someone else's (julian3833) notebook. It also seems like the same group of people are upvoting these notebooks\n\nEdit: I've also noticed that firefliesqn frequently changes their profile picture, which I would pessimistically interpret as an attempt to not be recognized. Maybe I'm reading into it too much.\n\nI think Kaggle can compare the code in a notebook to other public notebooks and indicate at the top that the work heavily features content from another notebook(s), linking that notebook(s). This could just be an endless rabbit hole and it could be hard to know who is the \"original\" author. Perhaps the comparisons should be limited to other public notebooks within the same competition and not all of Kaggle. Food for thought.", "votes": null }, { "id": "1522049", "postDate": "09/23/2021 19:19:51", "content": "I agree downvotes should be enabled just because. It's strange that hey aren't. However, I doubt it'd solve he issue. If certain individuals are maliciously removing fork history and C&P or uploading a notebook then releasing it as their own, the masses who vote for a kernel that produces results won't be bothered to deal with a bibliography.
", "rawMarkdown": "I agree downvotes should be enabled just because. It's strange that hey aren't. However, I doubt it'd solve he issue. If certain individuals are maliciously removing fork history and C&P or uploading a notebook then releasing it as their own, the masses who vote for a kernel that produces results won't be bothered to deal with a bibliography.", "votes": null }, { "id": "1522066", "postDate": "09/23/2021 19:40:35", "content": "Done my duty a long time ago upvoting.
", "rawMarkdown": "Done my duty a long time ago upvoting.", "votes": null }, { "id": "1523954", "postDate": "09/25/2021 22:34:16", "content": "Also, I think that just like the compare version feature in notebooks if we could have some way to compare any 2 notebooks on Kaggle then it would help. Many times it is difficult to find where exactly the code lines have been modified.
", "rawMarkdown": "Also, I think that just like the compare version feature in notebooks if we could have some way to compare any 2 notebooks on Kaggle then it would help. Many times it is difficult to find where exactly the code lines have been modified.", "votes": null }, { "id": "1524054", "postDate": "09/26/2021 04:47:14", "content": "I think that just like the compare version feature in notebooks
", "rawMarkdown": "I think that just like the compare version feature in notebooks", "votes": null }, { "id": "1524762", "postDate": "09/26/2021 20:10:23", "content": "Thanks for sharing
", "rawMarkdown": "Thanks for sharing", "votes": null }, { "id": "1559735", "postDate": "10/27/2021 07:10:20", "content": "Hey All,
\nThank you all for taking part in our competition. The participation has been overwhelmingly positive. We are currently conducting a survey to gauge the demographic and outreach achieved. Kindly spare 2min and fill in this survey https://forms.gle/QP9L16niPexozyhu5.
\nThank you all,
\nRegards,
\nChris
I proposed to enable downvotes on notebooks for this kind of useless shallow copy without added value. Feel free to uppote and support if you think it could be useful. I do think that if we were down voting shallow copies then they would quickly disappear.
\nhttps://www.kaggle.com/product-feedback/76394
", "votes": null, "replies": [ { "id": 1522049, "author_name": "authman", "author_url": "", "post_date": "09/23/2021 19:19:51", "content": "I agree downvotes should be enabled just because. It's strange that hey aren't. However, I doubt it'd solve he issue. If certain individuals are maliciously removing fork history and C&P or uploading a notebook then releasing it as their own, the masses who vote for a kernel that produces results won't be bothered to deal with a bibliography.
", "votes": null, "replies": [] }, { "id": 1522066, "author_name": "scaomath", "author_url": "", "post_date": "09/23/2021 19:40:35", "content": "Done my duty a long time ago upvoting.
", "votes": null, "replies": [] } ] }, { "id": 1521919, "author_name": "jslim7", "author_url": "", "post_date": "09/23/2021 16:42:46", "content": "I'm sorry to hear that. Your works are awesome.
\nSo I checked the notebook function 'Report Notebook'. [Plagiarism/copied content that is not meaningfully different]
\nI will report that notebook.
If you look at firefliesqn's other notebooks, it is usually the same thing. Copy a notebook, change the random state, get upvotes. In this one, they copied one of my notebooks: https://www.kaggle.com/firefliesqn/chaiii-qa-starter but then version 2 copies Abhishek's notebook, and version 3 copies someone else's (julian3833) notebook. It also seems like the same group of people are upvoting these notebooks
\nEdit: I've also noticed that firefliesqn frequently changes their profile picture, which I would pessimistically interpret as an attempt to not be recognized. Maybe I'm reading into it too much.
\nI think Kaggle can compare the code in a notebook to other public notebooks and indicate at the top that the work heavily features content from another notebook(s), linking that notebook(s). This could just be an endless rabbit hole and it could be hard to know who is the \"original\" author. Perhaps the comparisons should be limited to other public notebooks within the same competition and not all of Kaggle. Food for thought.
", "votes": null, "replies": [] }, { "id": 1523954, "author_name": "debarshichanda", "author_url": "", "post_date": "09/25/2021 22:34:16", "content": "Also, I think that just like the compare version feature in notebooks if we could have some way to compare any 2 notebooks on Kaggle then it would help. Many times it is difficult to find where exactly the code lines have been modified.
", "votes": null, "replies": [] }, { "id": 1524054, "author_name": "mahbubur1043", "author_url": "", "post_date": "09/26/2021 04:47:14", "content": "I think that just like the compare version feature in notebooks
", "votes": null, "replies": [] }, { "id": 1524762, "author_name": "devilmaycry00", "author_url": "", "post_date": "09/26/2021 20:10:23", "content": "Thanks for sharing
", "votes": null, "replies": [] }, { "id": 1559735, "author_name": "zerafachris", "author_url": "", "post_date": "10/27/2021 07:10:20", "content": "Hey All,
\nThank you all for taking part in our competition. The participation has been overwhelmingly positive. We are currently conducting a survey to gauge the demographic and outreach achieved. Kindly spare 2min and fill in this survey https://forms.gle/QP9L16niPexozyhu5.
\nThank you all,
\nRegards,
\nChris