A few months ago, I had trouble with Firefox on Android, so I started looking again in the settings; something you really rarely do in a browser. Finding a few things like data collection, usage data, marketing data, and "occasional studies" being all enabled by default sure reminded me that Mozilla isn't what it used to be.
Any product manager needs data about how a product is used to make the product better. Of course they need to test if moving a button to a different place leads to an easier to understand setting screen; or if moving extensions into a separate menu means fewer people find the malicious extension and turn it off.
I’ll be the first person to say that Mozilla is bigger than it needs to be and their org size isn’t justified by their results. But to think collecting data automatically makes them suspect seems to me lazy. It’s what they do with the data that counts.
You're conveniently missing the point that there is an actually labeledtelemarketing partner that is opt-out. That's not user habit collection. You're also missing that "random future studies" should not be auto-enabled by default either. Finally, the topic of this particular post is about categorizing search queries, which as far as they describe it isn't something your browser should care about.
The only thing that may be legitimate is, as you say, actual UX and feature usage. But for that to be done properly, you have to ask and make it opt-in, as with any data collection scheme. It's actually a requirement in some places.
The point is, people give shit to chrome because "evil google collects your habits data and monetize them", while people like you are a-ok with Firefox openly sending data to a third-party marketing partner on opt-out conditions and, as demonstrated by today's post, adding more collection that have absolutely nothing to do with the behavior of the browser and all to do with user habits.
If you go through my comment history you’ll find me saying, multiple times, that Mozilla has worked itself into this problem, by adding far more people than they need. The browser would be healthier, I suspect, if there was a 50-strong, open-collective backed, dev team working on just the browser. At the minute the org is enormous and they now need to find a way to pay for that enormous org.
The conversion metric is not whether or not a screen is easier to understand and malicious extensions are off. The conversion metric is whether or not you subscribe to one of their services.
That's all essential in modern development and to see how to target what people are actually using. Though I don't like it either and it should be a simple single button opt out, or even an opt in.
Why would they need to tie that telemetry to people in order to use it to inform their own development (as it states as the purpose, and is the purpose of all their telemetry as far as I know).
To know what features people are using, how fast it's running, know what hardware and where it's being used, and to try to investigate crashing issues? Telemetry doesn't only mean knowing where you live or who you're banging.
To know what features people are using, how fast it’s running, know what hardware and where it’s being used, and to try to investigate crashing issues?
None of those things are what's being discussed here, or what GP asked about. As stated in the article, this is about categorizing people's searches.
Because that allows them to sell the default search engine spot for more; the more you know about an audience the more it’s worth, even this high up the food chain.
Lets say you live in a tribe. Everyone eats the same shit. Everyone does the same work. Everyone feels the same way. Why is it necessary to pinpoint an individual?
Any specimen from the batch is going to tell you what you what you want to know. #deanonymity
To improve Firefox based on your needs, understanding how users interact with essential functions like search is key. We’re ramping up our efforts to enhance search experience by developing new features like Firefox Suggest, which provides recommended online content that corresponds to queries. To make sure that features like this work well, we need better insights on overall search activity – all without trading off on our commitment to user privacy. Our goal is to understand what types of searches are happening so that we can prioritize the correct features by use case.
This sadly is in line with Mozilla's increasingly bad privacy defaults. Users who care have moved on to more reasonable configurd forks at this point (e.g. Librewolf).
For anyone wondering, it's controlled by the existing top-level Send Technical And Interaction Data toggle in the privacy menu that's been there for ages, so most users who care about privacy have probably already opted out.
If it's optional it should be disabled by default. 99% of people aren't going to even know this is a setting or something that's going on behind the scenes
Telemetry is important for prioritizing feature development and support for the silent majority of users that don't disable it and then complain about ALSA support being dropped.
There are ways to prioritise feature development that don't involve telemetry and I'd even say they are better than telemetry. For instance, surveys, user interviews, usability tests and that sort of thing.
Being real here? Anyone that can't see the damn button for it during initial setup isn't going to give a damn.
Best practices? No. Opt in only should be the default. But that's still about choice, not whether or not telemetry is inherently a bad thing. But if someone is too damn lazy to look at the settings of a program when they first use it, that's pretty damn stupid. But, hey, people in general are stupid.
There's no "initial button". Installing Firefox on mobile you'll have technical data collection, marketing (with a third party) data collection, and "random studies" enabled without a clue. As someone that is very wary of this, I can assure you that at no point I was asked anything about sending data to "Adjust" (marketing partner), Mozilla, or allowing random, unknown at the time, studies.
why, when it can actually be helpful? Using chatgpt for searches is already much faster than denying 10 cookie banners. Some more privacy focused AI would be great, if firefox does that.
I cannot stress this enough: LLMs are not, have never been, and quite likely never will be search engines. You may as well ask your a auto-complete questions.
mozilla.org
Hot