Glenn Chapman

22 Comments

  1. The other thing they left out is a recent experience I had where the authors couldn't share the data, even though it was from a government database (not US) but a country that officially supports open source.
    In the animation, another under-current is that published papers do not allow for enough room to fully detail the research. This needs to end in this day of electronic publishing. Maybe print publishers should just print the abstracts with links to full papers?๏ปฟ

  2. Aaaargh! I have been on both sides of this. I will thorougly organize my data for the article I'm putting together now!

  3. Yes! Hilarious! "Where is your USB drive. It is in a box. It is in a box at home… ๐Ÿ™‚ ;)"

  4. "so many boxes"…..cracks me up every time! I love this youtube clip, thanks

  5. I've seen a number of papers where the Supplemental Materials were more like the "real paper" than the actual paper (which was more like an extended abstract).
    Yes… Everyone should include their data (and metadata, models, ect) in the Supplemental Materials. However, that won't happen in a lot of cases because the same data is used to generate multiple papers (analyzing the data in different ways and/or adding/expanding to the data set).

  6. this is why Nature's new scientific data publication, and others, are so topical npg blogs nature.com scientificdata

  7. Hilarious and at the same time so true. I wish I thought of making this video (and having a sense of humor about it). I sent to a friend who recently walked off my project because he couldn't work with the dataset. I explained this was typical in clinical data and that I'd waited 15 years to get access to datasets like these and I was going to work with them. Luckily I can extract what I need for the particular analysis. Thanks for the video, I sent it to him. Going forward: Provenance! & RDF

  8. As a data coordinator/manager of a large scientific data base, I've seen this happened many times. There are several inexcusable causes: lack of enforcement for data archival/documentation, job security, delegation of data documentation/archival to unqualified personnel, etc. My take on the matter is: IF one gets paid to do the research, he/she is OBLIGATED to deliver, document and make the data available.ย 

  9. I dont understand, people are really like this? This isn't how I do it. I would have converted the output to a CSV file, labeled all fields appropiately, made a README to describe the file, put copies everywhere, like lab shared computers, main work computer, home computer, Time Machine backups, external drive backups, dropbox, my version control system running on the lab server, and of course, actually on the lab website. All of this before we send the paper in for peer review. No one would have to ask for the data. If the data doesn't have legal restrictions on it, it would be on our website since the begining. If there were legal restrictions on the data, then the website would have a form to fill out to start the process of getting approval.

Leave a Reply

Your email address will not be published. Required fields are marked *