We got great feedback on our proposed Tidepool License (see the blog post). Most of the commentary came from the constructive and helpful folks on the Open Source Initiative list. We had an exciting and productive discussion, but came away with the feeling that we have a lot more work still to do on it.
The process has made it clear is that getting a new open source license right is hard — especially one with novel concepts such as the Tidepool License. We need to spend more time carefully considering all of the use cases. Different communities (academics, medical researchers, device manufacturers, patients, etc) have different requirements, and it’s a challenge balancing all of those needs.
We had been hoping that the process would happen relatively quickly, and we have been waiting until the license was ready before opening up our code publicly. But not developing our code in the open is slowing us down. Tidepool’s vision for an open data platform depends critically on having many eyes on it — user interface design, API design, testing, and development will all benefit from community commentary and contributions. We have had several people who have offered to contribute, but it’s been challenging to effectively integrate them into the process when that process is happening behind closed doors.
We’ve decided we can’t wait any longer. Here’s what we’re going to do for now:
- We’re going to keep working with the community on the Tidepool License. If you’d like to participate in the conversation, please join the tidepool-license-discuss mailing list. If you’d like to propose changes, we have a GitHub repository with the license text available and we will gladly consider pull requests for changes. We can also use GitHub issues for discussion.
- For most of the existing Tidepool code base, we will be releasing the code under the BSD 2-Clause license (a.k.a. BSD-2).
- For the code where keeping changes secret or private would be a safety concern, we will release the code under GPLv2. This will compel people who edit this code to release their changes back to the public.
At some point in the future, when we feel we have achieved a reasonable consensus on the Tidepool License, we’ll relicense future versions of the code under that license. But this plan allows us all to move forward. If you’re motivated to help, you’ll now have a variety of ways to do so.
Our VP of Engineering, Kent Quirk, will be following up this blog post with one that explains what we’re going to do and how.
Thanks again for all of your support. We’ve made great progress in establishing an open data platform and great applications to help reduce the burden of type 1 diabetes. We’re excited to get even more community involvement going forward.