3 ms·

> "The common use-case is a data collector in the field fills out a form while offline. Whenever a connection is found, the data gets sent to server where it ca

> "The common use-case is a data collector in the field fills out a form while offline. Whenever a connection is found, the data gets sent to server where it can be seen and acted on." is there any user control over this? otherwise, this is an totalitarian's dream come true. i'd suggest thinking about this capability a little more in relation to "positive impact".

yanokwa 4y ago
The vast majority of users are staff who are hired to collect data for a non-profit. The vast majority of servers are self-hosted by those non-profits. It's very opt-in in that regard. Think of say a census worker counting households or farmer measuring plot yield. And even in those cases, yes, users have a lot of visibility in what is being collected and a fair bit of control. Sensitive data that is collected from participants that users interact with are typically governed by policies set by Institutional Review Boards or Ministries of Health.
- clairity 4y ago
  appreciate the forthrightness, but frankly, that's an unconvincing answer. participants should have full control over their own data rather than offloading trust to government agencies or corporate organizations that are potentially themselves corrupt. note, for instance, that many gov agencies blindly and willingly send PII to google without participant permission.
  - yanokwa 4y ago
    I don't disagree and we work hard to provide guidance on best practices and software to support implementation. Ultimately, in the contexts in which we work, the organizations are the data controllers and there are complex legal issues at play. For example, in some locales, clinical trial participants don't have the right to be forgotten.
snowwrestler 4y ago
I’m really confused by your objection. Can you be specific about what kind of user control you think is missing? People know they are voluntarily filling out a form and for whom. It just doesn’t go to a server immediately due to lack of signal. How is this any worse than voluntarily filling out a printed form on paper and putting it into a mailbox to get picked up later? A practice that has been socially accepted for decades?

yanokwa 4y ago

The vast majority of users are staff who are hired to collect data for a non-profit. The vast majority of servers are self-hosted by those non-profits. It's very opt-in in that regard. Think of say a census worker counting households or farmer measuring plot yield. And even in those cases, yes, users have a lot of visibility in what is being collected and a fair bit of control. Sensitive data that is collected from participants that users interact with are typically governed by policies set by Institutional Review Boards or Ministries of Health.

clairity 4y ago
appreciate the forthrightness, but frankly, that's an unconvincing answer. participants should have full control over their own data rather than offloading trust to government agencies or corporate organizations that are potentially themselves corrupt. note, for instance, that many gov agencies blindly and willingly send PII to google without participant permission.
- yanokwa 4y ago
  I don't disagree and we work hard to provide guidance on best practices and software to support implementation. Ultimately, in the contexts in which we work, the organizations are the data controllers and there are complex legal issues at play. For example, in some locales, clinical trial participants don't have the right to be forgotten.

snowwrestler 4y ago

I’m really confused by your objection. Can you be specific about what kind of user control you think is missing? People know they are voluntarily filling out a form and for whom. It just doesn’t go to a server immediately due to lack of signal. How is this any worse than voluntarily filling out a printed form on paper and putting it into a mailbox to get picked up later? A practice that has been socially accepted for decades?