06-07-2018 12:33
- Mark as New
- Bookmark
- Subscribe
- Permalink
- Report this post

06-07-2018 12:33
- Mark as New
- Bookmark
- Subscribe
- Permalink
- Report this post
Greetings! For the past 5 years I've been writing web apps with Fitbit integration as a Research Software Engineer for research studies at my university, and I've generally simplified the account generation process by using Gmail's well known "myaccount+ignored@gmail.com" hack. Basically Gmail will ignore any characters between a plus '+' sign and the @ sign in an email address, so we create Fitbit accounts like this:
mystudy+001@gmail.com
mystudy+002@gmail.com
etc.
This lets the researchers log into a single account to manage the registration process. To date this has worked fine for us, and Fitbit.com seems to happily permit this.
However, I learned earlier from a major Fitbit integrator that when they used the gmail hack recently the Fitbit host flagged this type of account as suspect in some way and shut all matching accounts down.
I would like to learn if there's any way we might avoid similar problems. Could you explain when and how you detect that a new account is suspect? Do you currently have or do you have plans to create a flag that would permit this type of account generation?
I know we're coloring outside the lines by using Fitbits in research studies and that we should probably be using something like ActiGraph to retain greatest control over user data, but it's hard to ignore the cost difference. 🙂
Any thoughts?
08-10-2020 23:08
- Mark as New
- Bookmark
- Subscribe
- Permalink
- Report this post

08-10-2020 23:08
- Mark as New
- Bookmark
- Subscribe
- Permalink
- Report this post
Hi Kevin:
I am writing a server side application for a research study at MD Anderson Cancer Center. I am trying to solve the same problem you describe in your post. Did you ever find an answer? I would be most interested in hearing about your experiences writing software for research studies.
Charles

