Closed alvi-khan closed 2 months ago
@alvi-khan It is not offered as a service, and we don't provide any guarantees w.r.t availability or uptime of the online tool. By introducing a privacy policy, I feel it is slipping into the "service" territory from the "tool" territory.
Or put it another way, it feels like we are taking some responsibility for how the software works. Not wanting to deal with that responsibility is the reason for choosing MIT License.
My ideal response, to "I want to get approval to use this tool", would be to suggest forking the source code and building it in their own machines and make it "their tool" instead of using the version hosted on Github Pages. They can use whatever auditing system to certify it for their internal use.
I don't see any issues with including the Privacy Policy itself. It is simply stating what the software does with the data. I just want to pause and think about how to do this pragmatically without slipping into the territory where we end up becoming a "supplier of goods and services" instead of "makers of goods".
So, I am not approving the PR yet. It would be great if you let me know what you think as well, as you handle a lot of the day to day maintanence, which I only look at this occassionally.
@tecoholic I see what you mean.
My intention was to address situations where people wanting to use the tool are facing push back from their institutions asking for 'official' documents. Although I think the MIT license addresses the fact that we don't bear responsibility for the tool, I can see how including a privacy policy could make people think that we do. After all, most open source tools don't have one.
We can try updating the policy to remind users that, although we don't actively collect data, we also don't bear responsibility for the tool. Otherwise, we can close this PR for now and revisit the issue in the future if we get further requests for an official privacy policy.
@alvi-khan
Although I think the MIT license addresses the fact that we don't bear responsibility for the tool, I can see how including a privacy policy could make people think that we do. After all, most open source tools don't have one.
I think the issue is NER Annotator is a desktop tool as well as a web app. So, the privacy policy requirement is probably about using the web app.
We can try updating the policy to remind users that, although we don't actively collect data, we also don't bear responsibility for the tool.
I am onboard with adding a privacy policy, if we explicitly state in the Privacy Policy, that it is specific to the GH Pages hosted one. Let us also put in, anyone who has concerns should download the desktop version or build from scratch based on their comfort level.
What do you think? If you agree, I can add the changes or you are welcome to add the changes yourself. :)
@tecoholic Sounds good to me. I've made some changes. Please take a look and modify the content or wording as you see fit. Thanks!
Following the the discussion on our data usage policy (#114), I'd like to add this simple privacy policy document stating that we do not collect any data from users.
Although an open-source tool does not necessarily need this, I can see how it would be useful to people using the tool in academic or professional environments.
Requesting your opinions on this @tecoholic.