Telemetry Policy - Remaining Questions

[Nicolás Alvarez]

2017-09-13 19:20 GMT-03:00 Volker Krause <vkrause at kde.org>:
> (1) Should we allow opt-in tracking of unique identifiers?
>
> This was requested by Jaroslaw, as Kexi has this right now and the policy as
> written right now would thus conflict with it.

My view is that we need opt-out telemetry with unique identifiers to
get useful conclusions out, but from what I saw in the big thread I
know this is an unpopular opinion.

> (2) Should we require/allow/forbid publication of the raw data?
>
> Publication was suggested by Martin F. Practically, this would have to allow
> for a certain delay, we can't have public access to live data. Suitable
> licensing options of the data would probably be CC0 or CC-BY-SA.

FWIW, Mozilla does not publish raw data. Mozilla employees can run
custom analyses on raw data (for cases where the public aggregated
data isn't enough), but I think even them can't just look at
individual records.

(Then again, Mozilla raw data contains unique identifiers, which can
make their privacy requirements stricter)

I also wouldn't want to "require" publication of raw data for
practical/technical reasons: until we implement some minimal
pseudo-telemetry pings, we don't even know how many active users we
have, so we can't estimate what's the total volume of raw data we will
end up collecting. It may end up needing extra server resources to
store and aggregate. I wouldn't want to also make said giant raw data
publicly available. Maybe it's technically possible, but I don't want
to already require/promise that we will do it.

-- 
Nicolás