please share your thoughts if i should have done things differently

Soulless LLM-ese. The content is needlessly stretched to occupy space, paragraphs, and tables. Any actual insights could have been condensed and conveyed with 80% less text.

The "Complete 72-Hour Raw CSV Dataset" CSV contains 13 records. That incomplete data is formatted incorrectly: rows 6 and 11 are missing payload_bytes, so their description shows in the wrong column.

The "How DeGoogled Are You" section is mostly a separate issue from the test described in the article, and it has no place being at the top. The article is about how Android phones are passively phoning home. For example, switching from Chrome to Brave won't reduce idle phone-home, as Android users can't uninstall Chrome, and the idle activity was not related to Chrome. The same goes for most of the other recommendations in that widget.

Seems like you used AI to steal the results from the same experiment done a decade ago to be honest. Heard this story in a YouTube video long ago.

Stock Pixel 8 (Default Settings)-> 348.4 req/hr

of which: Nearby Wi-Fi Routers (Location Triangulation) is 84.2 pings/hr

So a stock (idle!) pixel 8 is checking where you are more than once per second (without maps running, without asking for directions) and asking what you're doing almost five times a second.

Is there any slope at all? Like, does it ask more frequently in the beginning and less after an hour+ of idle, or is it fairly static?

Would be very curious to see what adding social media and voice assistants does to this, especially with a controls against a silent room and a room with a radio/movies/TV/pure music playing.

Does the voice assistant send more data in the conversational context? How much more? If it is indeed more, how much voice is theoretically being transmitted given the average words per bitrate?

It could also be interesting to see if the stock device transmits more in a conversational environment. Like, if the device is idle, why would it be transmitting more while sitting in an office instead of a quiet lab?

You could not have posted AI slop, and written your own thoughts as if you were an actual human being with something to say in their own voice.