This is a project intended to support unattended data collection in a way that protects the data being collected. The initial project is aimed at audio recording using a Raspberry Pi Zero (simple, low hanging fruit), with the intention of being extensible for other uses.
There are many hardcoded paths, including some security-relevant ones like the directory for the gpg repository / data. This is bad from a usability and portability perspective. It works (for now) because this project is being used on a bare, minimal, limited system that is consistent due to those features.
When the system is operated without WiFi or other network access, accurate time is unable to be maintained (no on-board clock maintained by battery for the Pi Zero). Something seems to be preventing the clock from resetting to the start of the unix epoch, possibly (likely?) something like fake-hwclock [https://manpages.debian.org/jessie/fake-hwclock/fake-hwclock.8.en.html].
This is better than not, BUT, there is still value in being able to determine the correct time.
Partly / largely as a result of issue #2, timestamps recorded with data is inaccurate. It will be valuable to be able to determine when recorded events occurred.
Given that someone is likely to sleep for 8 hours at a time, and reviewing 8 hours of data (audio recordings) in 10 minute increments is ... a lot (could take more than 8 hours), it is important to facilitate review.
Because recording is started using cron periodically (every 10 minutes on the 10th minute) and not otherwise, there is about 5 minutes between when the microphone is connected and recording begins.
The mitigation for this issue is that there is visual indication when the microphone begins recording (the LED is extinguished when the microphone is recording), but this does impact other issues.