[pwdisswordfish2]

There is an even easier analysis. If you were designing your own headphones just for your own use, would you have them collect personal data? If your answer is no, then choosing headphones that do not collect data is a logical choice.

The author cites some idea of "trading" ongoing collection of personal data^1 for features but I can't see how that applies here, assuming the user has already paid for the product, e.g., he has already paid for the headphpones.

1. This does not appear to be a one-time, voluntary submission of data by the purchaser. For example, submitting one's name and a product serial number in order to register for a warranty.

[MereInterest]

I'd put it a bit differently. If I were designing headphones for my own use, would I have them collect personal data and then give that data to somebody else. I like collecting statistics about myself, because they are interesting. For example, tracking exactly when I go to sleep, start my commute (pre-corona, that is), go on errands. But that is data that is for my use, and not something that I would trust in somebody else's hands.

For headphones, I could see tracking the usage by day, and by time of day. I could see tracking the average volume, the dynamic range, the frequency ranges used. How often do I spend listening to music, to conference presentations, to movies? Those would be fun data to have, but not something that I'd want under the control of a different party for privacy reasons.

[pwdisswordfish2]

Would you store the personal data in "the cloud" (i.e., remotely accessible shared computer)

[MereInterest]

Usually depends on who controls that remotely accessible shared computer. If it is under my control, or under the control of somebody I personally trust, then yes. If it is under the control of somebody else, then no.

Partly, that is because I've been coming to the conclusion that the privacy promises of a company are rather meaningless. In the case of bankruptcy, collected data is treated as an asset, rather than as a liability to be disposed of properly. In the case of acquisition, the database gets transferred to a parent company. For example, with Google buying Fitbit, they will have have access to all data collected by Fitbit. Though (as of yesterday), Google is stating that they don't intend to use this for advertising, I don't trust that not to be "accidentally" merged with google tracking ids in the future.

[pwdisswordfish2]

Indeed, I do not mean to suggest that shared computing is necessarily a net negative. Sharing a remotely accessible computer among trusted friends/family has always been an interesting idea to me because it would allow such easy communication. The way I see it, this is the core idea that underlies "e-mail" on UNIX. A group of people sharing accounts on a computer. UNIX allows easy messaging between accounts. Another group doing the same thing on another computer. Connect the two computers over a telephone line and now we have messaging to people in another group. Today, if you and I can both connect to the same computer, where we both have accounts, that is still an easy way to do messaging, in my opinion. Computers are no longer prohibitively expensive. There is no requirement that the computer belong to a third party. The ongoing problem however is that there needs to be someone who can administer the computer. It is not easy enough for anyone to do. It seems the "sysadmin" is still needed as much today as in the past.