It’s funny that when the story is about their political allies, that data becomes much less concerning:

> “It’s really, really hard to assign even what side of the street you’re on when you’re using this kind of data,” said Paul Schmitt, a research scientist and professor at the University of Southern California.

https://www.nytimes.com/2022/05/29/us/politics/2000-mules-tr...

Looking at the preceding paragraphs, I'm not sure I understand what point you're trying to make:

> Mr. Phillips and Ms. Engelbrecht’s case is largely built on cellphone data. A report created by the group includes an appendix that claims to list “IMEI” numbers of the tracked devices — 15-digit codes unique to each cellphone. But each entry on the list is a 20-character string of numbers and letters followed by a lot of x’s. Mr. Phillips said new IDs had been created “to obfuscate the numbers.”

>"The same report says the group “purchased 25 terabytes of cellphone signal data emitted by devices” in the Milwaukee area in a two-week period before the 2020 election. They claim to have isolated 107 unique devices that made “20 or more visits to drop boxes” and “multiple visits to nongovernmental organizations” that were involved in get out the vote efforts.

>A number of researchers have said that while cellphone data is fairly precise, it cannot determine if someone is depositing ballots in a drop box or just passing by the area.

>“It’s really, really hard to assign even what side of the street you’re on when you’re using this kind of data,” said Paul Schmitt, a research scientist and professor at the University of Southern California.

The parent posted a NYT article about cell phone data being used to inferring an individual activity based on their location. Recently, the NYT is implying that the data isn’t really all that accurate and can’t be used to infer an individuals activity.

The data is probably just phone self-reported GPS coordinates, and we know how accurate or inaccurate those can be: https://www.gps.gov/systems/gps/performance/accuracy/#how-ac...

Both are articles are perfectly consistent with being able to locate a person to within at best a 16 feet radius or so, and less accurate when in a built-up area. Hence the quote about not being able to reliably tell which side of the road someone's phone pinged on.

It’s also the same data used by police to identify devices (and by extension their owners) that were near the scene of a crime. Phones don’t use “just” GPS for calculating position.

My point is one article uses the information to (rightfully) strike fear/discomfort in the reader. The other article uses it to dismiss an investigation that is inconsistent with the paper’s narrative.