To dveditz's points: > Are you tracking users, or campaigns?
Me personally? Neither : ) But to my knowledge from helping with the implementation of attribution collection in Firefox Telemetry the answer is campaigns (and other things on that order)[1]. Basically `utm_` params. > do you track cross-site? Again, not me personally : ) But we may have partners driving us traffic, and we may host the installer on non-first-party eTLD+1s for, I dunno, CDN reasons? (not sure we do for Desktop, but we don't host our own packages for Android and iOS as you'd imagine. App Stores. Ick.). > Are you likely to use fbclid= or other known tracker to do so? I doubt that very much. To my knowledge we're interested in campaign/experiment/branch-level efficacy measurement, not user-level tracking. Ultimately, I'm bringing this topic of "will this hurt marketing attribution" up on behalf of folks not on dev-platform who might have their marketing spend analyses thrown were this to go awry. But it sounds like their analyses are safe. To jhoffman: > I'd like to learn more about how attribution for downloads works (and I can reach out to you separately and update this thread) but I doubt that it utilizes a high-entropy token to join user identities across sites, as will be disallowed in our anti-tracking policy. All data collection in mozilla projects is documented in public as required by Data Collection Review[2] so the info should all be "available" (still working on making it discoverable). Firefox Telemetry's (for Firefox Desktop) is the `attribution` section of the Environment[1] and is basically `utm_` params that get saved to disk on install and reported to Telemetry on run. Might be platform-specific wrinkles in here, definitely talk to the Desktop Integrations team about further technical detail. For mobile we probably have to mention Adjust and I'm the wrong person to talk to about that (and sadly I'm not sure who _is_ the right person to talk to about that). :chutten [1]: https://firefox-source-docs.mozilla.org/toolkit/components/telemetry/data/environment.html see "attribution" [2] https://wiki.mozilla.org/Data_Collection On Tue, Oct 12, 2021 at 1:00 PM Johann Hofmann <[email protected]> wrote: > (Re: Are we breaking attributed downloads?) > > I'd like to learn more about how attribution for downloads works (and I > can reach out to you separately and update this thread) but I doubt that it > utilizes a high-entropy token to join user identities across sites, as will > be disallowed in our anti-tracking policy. > > (Re: Future iterations of this protection) > > There is a lot more work to be done until we can move towards a > not-quite-listed-based version, though I think that this is a desirable end > goal. We'll have to figure out how to identify tracking identifiers and, at > the same time, allow for the various legit uses of URL parameters that look > very much like navigational tracking. The Privacy CG is currently > discussing this challenge as part of a standardization effort on > navigational tracking mitigations in > https://github.com/privacycg/nav-tracking-mitigations. We hope to > contribute to and benefit from this effort over time. > > On Tue, Oct 12, 2021 at 6:04 PM Daniel Veditz <[email protected]> wrote: > >> On Tue, Oct 12, 2021 at 8:42 AM Chris H-C <[email protected]> wrote: >> >>> Might this (or a future not-quite-list-based version) have an effect on >>> attributed downloads? I know we use attribution in product downloads to >>> determine the efficacy of marketing campaigns (as a non-exhaustive example). >>> >> >> Are you tracking users, or campaigns? >> >> Tim did say same-site navigation wouldn't be affected (any *.mozilla.org >> to another *.mozilla.org, for example) -- do you track cross-site? Are >> you likely to use fbclid= or other known tracker to do so? The intent is to >> stop broad cross-web re-identification, an end-run around cookie tracking >> protection. Firefox downloads or similar specific use-cases aren't going to >> end up on that radar. >> >> For a rough idea of scope, Brave's list (which can often get away with >> breaking more sites than we can) can be found at >> https://github.com/brave/brave-core/blob/master/browser/net/brave_site_hacks_network_delegate_helper.cc#L31-L58 >> >> -Dan Veditz >> >> -- >> You received this message because you are subscribed to the Google Groups >> "[email protected]" group. >> To unsubscribe from this group and stop receiving emails from it, send an >> email to [email protected]. >> To view this discussion on the web visit >> https://groups.google.com/a/mozilla.org/d/msgid/dev-platform/CADYDTCAPZO6jUNbYnzhmO2YS1DOordR7H6yE0Amoi4Lsq7-FxQ%40mail.gmail.com >> <https://groups.google.com/a/mozilla.org/d/msgid/dev-platform/CADYDTCAPZO6jUNbYnzhmO2YS1DOordR7H6yE0Amoi4Lsq7-FxQ%40mail.gmail.com?utm_medium=email&utm_source=footer> >> . >> > -- You received this message because you are subscribed to the Google Groups "[email protected]" group. To unsubscribe from this group and stop receiving emails from it, send an email to [email protected]. To view this discussion on the web visit https://groups.google.com/a/mozilla.org/d/msgid/dev-platform/CAMPhgK_0zHBrkt0ogf_j%2BEYgk5AEcdQO_QNkKFWm%2BW-2c4ctww%40mail.gmail.com.
