Archived What are your thoughts on open sourcing datasets / codebases in science? (AskScience)
submitted ago by eddotman
Posted by: eddotman
Posting time: 5.4 years ago on
Last edit time: never edited.
Archived on: 2/12/2017 1:51:00 AM
Views: 250
SCP: 8
8 upvotes, 0 downvotes (100% upvoted it)
Archived What are your thoughts on open sourcing datasets / codebases in science? (AskScience)
submitted ago by eddotman
view the rest of the comments →
[–] GuruFault 0 points 1 point 1 point (+1|-0) ago
I am a scientist and I have to agree that it is /really/ hard to give up a competitive code-base for free. Others feel the same way about their datasets. In addition, it can be challenging to take the time to polish things up enough that there is no sense of professional pride being put on the line by sharing something that doesn't look perfect. However, you can rest easy for three reasons. First, at least within my discipline and I suspect many others, unless your work is entirely and groundbreakingly novel, few are likely to make use of what you make available. This is especially true if your released work doesn't have a lot of polish on it. Second, it is common practice to release work under some sort of attributional license. That means that others who use your work must cite it. The advantage in this is that if what you have done is leveraged by many in the scientific community, you will get due credit for it. This will be reflected in citation metrics like your h-index. Finally, you retain a huge advantage. You already understand the work that you are publishing and you have all of the time from the end of your data collection to publication in order to get started on the next project in the series. Frequently this will provide you with at least a second publication from your source data / source code prior to anybody else even beginning to try to wrap their heads around your first publication.