You'd think opting out of sharing analytics data with app developers would also cover push notification metrics but it doesn't seem to be that way from a quick look over the official docs.
If anything we need more settings, and more warning popups to go along with them that list or link to information on some of the breakage users can expect. Inform the user and let them decide.
This is entirely false, you must be blocking the requests on the network level or something.
For one the getfirefox.com redirect includes tracking parameters in the destination URL, then there's Google Tag Manager, there is Sentry, some pages with their first party analytics solutions like Glean and more.
I installed uBlock Origin, with my Privacy settings set to normal/default, and there is nothing blocked (normally the plugin would show a long list of blocked analytics stuff). No clue, are you using Firefox as well? Not sure what the difference is with our experiences, but I see not a shred of Google or analytics-related network requests when I navigate to getfirefox.com . It DOES have a query param in the destination URL that suggests tracking, but it never does anything with it (???) https://www.mozilla.org/en-US/firefox/new/?redirect_source=g... ... I'm investigating further and I see that the page itself loads a self-hosted Google Tag Manager .js, but it looks like it has code that respects the "do not track" setting in my browser, but oddly I don't even have that enabled (woops)! Strange, no idea what's different for me. I'll try on another machine just to compare!
That's interesting, I'm indeed seeing some differences between browsers on those pages but those aren't exactly baseline setups. I'll have to do a more thorough comparison tomorrow and will reply with my findings.
Homepage: "Zero data collection. Quetta does not collect, store or share any data. Your data always belongs to you."
Google Play: "No data shared with third parties", "No data collected", "Zero data collection & sharing"
Privacy Policy: "we do process app launch information, which is used to determine if the software can run properly. This service is provided by the third-party service provider Google Analytics for Firebase."
I wish you all the best, the more competition the better, but you really shouldn't advertise zero data collection & sharing when it simply isn't true.
bikeshedding is more of a group phenomenon where trivial decisions outweigh more meaningful and difficult decisions due to the trivial decisions being easily understood.
If anybody's using it like that, it's totally lost its meaning. The whole concept is that sometimes to get result A you have to a total laundry list of seemingly unrelated tasks to clear the way.
This scene[1] from Malcolm in the Middle almost illustrates it perfectly except it's more Hal getting distracted on side quests that were spawned because he tried to change a light bulb...
It'd open the website in a new tab, discarding the old one. The new tab has its own isolated history with nothing in it. The back button would work fine but there'd be nothing to go back to :)
Even with all telemetry disabled (including the option "Allow Firefox to send technical and interaction data to Mozilla") and without anything showing on that page, the browser still sends out interaction data by means of so called "pings", URL parameters, etc.
Wow, a browser is making network connections, how dare it. Come on, these things are there so that if you are on "connected" to a network that require some registration, e.g. an airport, then you can actually be forwarded to that registration page.
Like what do you even think, Mozilla people are sitting on top of all that sweet
PING www.mozilla.org (3.161.119.172) 56(84) bytes of data.
64 bytes from server-3-161-119-172.vie50.r.cloudfront.net (3.161.119.172): icmp_seq=1 ttl=248 time=4.06 ms
64 bytes from server-3-161-119-172.vie50.r.cloudfront.net (3.161.119.172): icmp_seq=2 ttl=248 time=4.16 ms
64 bytes from server-3-161-119-172.vie50.r.cloudfront.net (3.161.119.172): icmp_seq=3 ttl=248 time=3.64 ms
It makes these connections even when all the things mentioned in the official "How to stop Firefox from making automatic connections"[1] are disabled. This includes the captive portal detection you allude to.
Regardless:
The browser asks me if I allow it to make these connections.
I tell it in unmistakable and irrevocable terms that I do not.
The browser makes these connections anyway.
The problem was never the connections or the contents of the data stream. You were aware of this before posting your comment, and yet you commented anyway. You are a troll.
Am I really the troll, or those who bring down the only remaining truly libre, open-source browser engine over stupid nitpicky bullshits, while not contributing anything positive to improve on the status?
The point is, if you provide an option to turn off telemetry, it should be off. No exception. In this case of an airport connection requiring a registration page intranetwork, this occurs via first request through their proxy with header redirect.
The point is, don’t lie about saying you respect privacy with half-baked options and telemetry for me but not for thee bs. If I turn off telemetry, you aren’t allowed to send telemetry. Of any kind. Not a crash log. Not an install token. Not a call home to see if maybe possibly your on an airport wifi and make-our-app-work-edge-case()
There is a distinct difference between telemetry and the captive portal requests Firefox makes. The captive portal checks have nothing to do with telemetry and contain no payload, which is why they are not covered by that setting.
Captive portal detection was only brought up by the non-constructive commenter and isn't part of the problem.
When the setting labelled "Allow Firefox to send technical and interaction data to Mozilla" is unchecked, the browser still sends out technical and interaction data through parameters added to various browser-provided/generated URLs (e.g. search results pages from built-in search providers, Firefox documentation links provided by the browser, DevTools compatibility panel links to MDN, etc.), by sending out pings when hovering over or clicking interactive browser-native interface elements, and in several other ways.
In summary: that setting is non-functional and should either be fixed, re-labelled, or removed entirely.
Not all places in the world share the same set of laws or jurisprudence. What's allowed or required and what isn't is entirely dependent on the jurisdiction(s) the website operators are beholden to.
Are you sure of that ? I knew that the competent jurisprudence or "forum" is that of the user, or everyone could circumvent the law using "offshore" companies to track people in regulated countries (i.e. EU). I was reading here on HN that , for that reason ( avoid EU Regulation ) sites like New York Times, blocked the availability of their contents in whole EU.
I was checking, see here, for example:
https://reutersinstitute.politics.ox.ac.uk/news/many-eu-visi...
Insofar as what the jurisdictions the website operators are beholden to or perhaps care about have to say about it, but otherwise I (not a laywer) share your perspective.
Cheers! It looks to be slightly more limited compared to Goodreads (e.g. no expanded series/universes, no different reading orders, etc.) but its the best alternative so far.
https://developer.apple.com/notifications/push-notifications...