Closed cbrenner04 closed 9 years ago
It's the responsibility of the researcher to set the password for their study. Given that PR is being used for purposes beyond CBITs patient studies, requiring a password creates issues for the other audiences that PR is serving. If the researcher cannot be bothered to set a password to keep their study settings immutable, that's on them, not PR.
We should also think about having a discussion later on about the visibility of PR with respect to its users (including patients). Given that we're deploying what's a very powerful piece of software that intentionally captures more information about the user than any other piece of Android software (or software they've used in the past), I think it's our ethical responsibility to find a way to make PR transparent to the end users in such a way that clearly lets them see (and understand) what information the app is collecting and where that data is going and how it's being used. While it may be expedient to hide as much of that as possible, that creates avery real ethical dilemma where we may be collecting information that the user has not and would not consent to. If we have to obscure the information being collected or otherwise "hide" Purple Robot, we find ourselves in a problematic position with respect to informed consent.
I want to encourage people to open PR and understand what's going on as a check on possible future abuse. (This is probably a philosophical discussion that should be fleshed out in more depth at a future time.)
Informed consent and transparency about PR does not mean that an end user has to see it doing what we said it would do. They just need to be informed of what PR does and what is being collected, which changes for every different use of PR. I understand your position on both of these items and do not necessarily disagree. But then I am going to back to the bug that led to this discussion and reopening that as a bug:
Users should not be taken to PR when responding NO to a prompt. I don't care how that is handled, either the ways I outlined above or by adding logic to the prompts.
Apologies if I sound pedantic about these issues from time to time, but I want to be crystal clear about separating the issues and being clear about what we're doing about them given that this entire conversation is happening in public by virtue of this being an open source project where development's done entirely publicly.
Part of the purpose of my response is responding directly to you, and part of it is describing to the anonymous reader / potential users where I stand on things as the principal developer of this software.
I don't think you're being pedantic. I'm being too colloquial and I'll make sure I tighten it up in the future.
No worries.
On that note - can you post the PR version you're seeing the dialog issue in? I'm reasonably sure that it's still in the most recent 1.5.1 unofficial build I distributed yesterday, but I'd like to get us in the habit of reporting the version number since not everyone's on the same build.
Yup that occurred in version 1.4.32 and 1.5.0
@nupmmarkbegale @lauriestark
This is an issue because:
We can't assume participants will not open PR. I know for a fact some participants have in the past. And given these participants had lower cognitive capabilities than the "general population", it is safe to assume participant WILL open PR.