1. You're proposing a mechanism for collecting data, and a strategy for extracting more data than you currently do. You have not figured out the type of data that you will finally need, only a set of things that you currently envision. Naturally, the data that you will collect in the future will be more than what you currently envision. There is built-in mission creep that is dangerous.
2. What you currently envision is not fleshed out as especially useful. You only believe it is useful. The pain point of biased data is red herring. Your concern is more about not enough data.
3. You have found a technology which you believe will allow you to collect a lot of data anonymously. But none of you seem to understand the technology very well. It seems like a shiny toy that you are eager to go to town with. I am not sure this is the right attitude.
4. You're proposing to use your users in lieu of proper testers, or to save time. There are many ways to properly test software and to save time. Have they been explored? There used to be a time when beta software was a thing. Prompt the users to become testers for your beta software. If users don't want to be testers then don't collect data from them. How much data do you actually need anyway? Have you fully utilized your existing data?
Over all, I see this as a nice-to-have luxury, not some life-and-death situation, and subverting the goodwill of users is not worth it, IMHO.
> There used to be a time when beta software was a thing. Prompt the users to become testers for your beta software.
Firefox already has opt-in telemetry, and Firefox already has a beta channel. It's unclear to me how it would help to tie telemetry to the beta channel; that would just make the existing problems (not enough data, and biased data) even worse, since there are probably far more users willing to share telemetry data than to use beta software.
This is incorrect, at least in theory. RAPPOR is designed to protect the user's data even if an attacker can see all of their individual responses over time. Of course, there could be implementation issues...
For the case of RAPPOR (and for what Apple is doing), you do not need to trust the aggregator with your data. These algorithms operate in the "local" model of differential privacy, where all privatization occurs on the users' local machines before being sent to the aggregator.
1. You're proposing a mechanism for collecting data, and a strategy for extracting more data than you currently do. You have not figured out the type of data that you will finally need, only a set of things that you currently envision. Naturally, the data that you will collect in the future will be more than what you currently envision. There is built-in mission creep that is dangerous.
2. What you currently envision is not fleshed out as especially useful. You only believe it is useful. The pain point of biased data is red herring. Your concern is more about not enough data.
3. You have found a technology which you believe will allow you to collect a lot of data anonymously. But none of you seem to understand the technology very well. It seems like a shiny toy that you are eager to go to town with. I am not sure this is the right attitude.
4. You're proposing to use your users in lieu of proper testers, or to save time. There are many ways to properly test software and to save time. Have they been explored? There used to be a time when beta software was a thing. Prompt the users to become testers for your beta software. If users don't want to be testers then don't collect data from them. How much data do you actually need anyway? Have you fully utilized your existing data?
Over all, I see this as a nice-to-have luxury, not some life-and-death situation, and subverting the goodwill of users is not worth it, IMHO.