a proposal to add opt-out telemetry in fedora is being discussed on fedora forums

sherlockandghibli@lemm.ee to Linux@lemmy.ml – 117 points –
F40 Change Request: Privacy-preserving Telemetry for Fedora Workstation (System-Wide)
discussion.fedoraproject.org

the proposal is from a red-hat team member and is proposing addition of "privacy-respecting" telemetry. here's the link to the hyperkitty thread

71

Jesus bellydancing Christ people, read beyond the title of the post for once. This is good.

This will enable results that aren't biased by the few users who find the option to opt-in (when it's not presented front and center) or actively answer to surveys.

They are taking as much care as possible with their words to make it clear that THEY DO NOT WANT TO TRACK YOU OR ABUSE YOUR DATA, it's just what's necessary for studies to better understand the users' needs.

  • You will have the chance to disable it front and center during the setup process;
  • No data will be sent before an active confirmation from the user during the setup process;
  • You will be able to fetch the data at any point to check what's being collected
  • They are reaching out to the community for guidance on what type of data we find acceptable or not
  • You can have an active participation on the building of this tool

And specially this, for the OH GOD, ITS THE END OF THE LINOX DESKTOP among you:

IT'S A PROPOSAL, IT'S NOT BEING ACTIVELY DEVELOPED YET!

Not only there's no reason to overreact, this could be the start of something beautiful for the Linux desktop, where the users not only actively participate but actively control their data

dude, i could not have been more simplistic with the title, and added both the threads for the post. yes, people are gonna get mad, that's what a general thread on the internet is. if you want actual dev discussions, go to the linked threads. if i tried to be more like, "guys don't harass the devs, they're definitely not in the wrong here", you know what that would've led to, and i just tried to avoid that. there are plenty of projects that don't need or use telemetry for development, including linux (i think, i'm pretty sure). it's great that they're being honest and open about it, but this coming just after red hat issue is what prompted me to start these threads. so yeah, sorry if this is frustrating for you, but this thread is not a ragebait, and i apologize if this does seem like one...

No need to apologize, I'm not mad at you, I found this thread while checking if someone had already posted it before doing so myself.

That's the worst part, to be honest, it's not even ragebait yet people just take everything at face value and start to spread FUD.

And they say Lemmy will not become Reddit. Pfft. The culture of reading only the headline and immediately take the rage bait is already seeping in.

That culture is human. If there are going to be humans on Lemmy, then you're going to get the exact same pros and cons.

Right? People are reading this as BIG CORPO DOES BAD DECISION and ignore the fact that they can (and should) go to the forum to actively discuss the proposal with the people proposing it.

i thought the whole idea was to be the only one who is using your computer. it doesnt matter if they are taking data that cant be used to identify you personally, the road to hell is made with good intentions. and also, i agree that by using telemetry, great insights can help the devs develop a better Linux desktop, but linux is more about the freedom to be let alone if you want to be left alone.

and how exactly does having telemetry (that can be easily turned off at any time) impedes your freedom to be left alone?

easy for who? why not use it in the sense of 'easily turned on.' and then make it as easy as possible for people to turn it on because it is in the devs and sponsors' best interest, but making an easy-to-turn-off telemetry is a conflict of interest for them.

Telemetry should not exist in the first place. If it exists, it should completely be opt-in and self-contained in one single package that is not installed by default.

Everything else is hostile.

Agreed; telemetry should always be opt-in because it requires consent.

The problem with opt-in telemetry is that it messes with the scope of the research.

If you want to understand something about most users (and not just the ones that are active enough in the project to participate in opt-in) you need this, otherwise your results only tell the needs of this subset of your userbase.

The problem with telemetry isn't the telemetry itself, is how it is used, and the way the proposal is worded makes me very optimistic. They are trying 200% hard to make sure we understand that it will never be used in violation of the users' privacy.

The problem with opt-in telemetry is that it messes with the scope of the research.

Too bad. That does not make it okay to collect data without consent.

Not ever.

In other words, unbiased telemetry is not possible to do ethically. (Or to say it differently, ethical telemetry necessarily has bias.)

And once again, it isn't "without consent", it just means that the default state of the checkbox is on. Users will still be presented with a confirm option before any data is sent.

In other words, unbiased telemetry is not possible to do ethically.

Say that to the opentelemetry and Plausible folks, who have been on the vanguard of doing exactly that for years now.

the default state of the checkbox is on.

That's a very strange thing to mistake for consent.

It's not mistaken for context, you're just missing the point. The switch is just part of it, the user would still have to consent to send their data before it is sent and the proposal proposes to have it detailing the data that is going to be sent and explaining the process.

Having it as a default guarantees it doesn't scare non-power users away from it. It's not about just having people clicking next and accepting it without consent.

Disagree, no matter the level of detail, having "yes" automatically selected is an assumption. What purpose would it have other than hoping people will just select the defaults and ignore it?

Having it as a default guarantees it doesn’t scare non-power users away from it. It’s not about just having people clicking next and accepting it without consent.

Scare away from what? Data collection? I mean even in that wording you are saying there is something to be scared of. It should be up the user. If you are saying "non-power users won't fully understand what is being collected and might get scared away if it isn't the default option" then that is even worse TBH. Preying on people not fully understanding what's going on.

It's not mistaken for consent, only if you dont switch it off the data is sent.

Inaction is not consent.

Wrong again, the "switch" only sets the setting itself, but before finishing the setup process the user would be provided with a confirmation detailing the data that would be sent and provided with the option to send it or not. The process would have user consent.

Having the default box being "on" is only for the purpose of hoping people click through without realizing.

There is literally no other argument here. "Consent" is: "Hey do you want this, yes or no?". Not "We are assuming yes unless you explicitly tell us otherwise".

a classic paradox-ish thing:

if the average person doesnt consent to fill a questionnaire, or an interview, how do you collect data about the average person. but then again, how do you know the average person doesnt want to fill a questionnaire? did you spread a questionnaire that had the question 'do you fill questionnaire?' in it

Usage data is important for developers to know how people use their software, so I'm okay with it. But given Red Hat's recent direction, I'm not sure I trust them to slowly increase the data being collected.

But I don't use Fedora and I've already moved off Red Hat/CentOS, so I don't have a horse in this race ¯\_(ツ)_/¯

Usage data is a crutch, effective developers can make good software with zero telemetry and did so for several decades.

Telemetry is important for desktop developers, you can negate it but it's a fact. Most users do not know what is GitLab

i genuinely do think most of the fedora users do know what gitlab is

I guess it's more accurate to say that people know what GitLab is but don't know/aren't willing to troubleshoot and submit quality feedback.

At work, many Linux user, they may know gitlab but not GNOME one.

Is that really a fact? (No.) Wow, it's crazy how every desktop program that doesn't use telemetry isn't any good, according to you.

They didn't say any software that doesn't have telemetry isn't any good.

His exact words:

Telemetry is important for desktop developers, you can negate it but it's a fact

I'm going to assume you know what the subject and object of that sentence are. Here's the thing about how language works on my planet: through the magic of a radical new concept called "context", we can accurately discern both meaning and normative statements from what people say and how they phrase it. In other words, "It's a fact that telemetry is important for desktop developers" is an ostensibly descriptive statement that also creates a normative statement in the same way that standing in the sun casts a shadow: it has to, it isn't optional. It's "Desktop developers who don't use telemetry are ignoring something that it's factual to say is important they not ignore". Please tell me you get it now, and that you don't need the rest spoonfed to you.

why? just don't ruin my beloved fedora please, if telemetry exists it should be opt in, not opt out

I totally agree. It's a shame, Fedora really is an awesome distro. Adding telemetry was to be expected since IBM is the parent company of Red hat and almost anything IBM touches turns to crap.

yeah might be time to distro hop i guess

Fedora is what finally got me to stop distro hopping 2-3 years ago. :(

fedora is my first experience with linux and i thought i made a good choice and i never had to mess around anymore but now i am seriously considering it :/

Same. It would be a real shame if we need to jump ship.

This. I always opt in BECAUSE it's opt in. Ask and ye shall receive and all that; but if you're gonna try to force my hand, well...no, I don't think I will.

Why would you prefer an IBM maintained distro?

Who said anything about "preference" to an IBM distro? I used Linux mint, Ubuntu, manjaro, arch, opensuse in the past, so there's definitely no preference to IBM.

No, this is not "ruining your beautiful Fedora", this is a proposal that both us users and the developers can pitch in to develop telemtry in a way that does not abuse its users.

And about the opt-in/opt-out nature of it, feel free to join the follow-up discussion about it on the forum:

https://discussion.fedoraproject.org/t/opt-in-opt-out-a-breakout-topic-for-the-f40-change-request-on-privacy-preserving-telemetry-for-fedora-workstation/85395

i get the importance of telemetry and maybe i worded my inital reply a bit harshly. but you probably will agree that this really does not paint a good picture of red hat with the whole rhel controversy as well

Highly disagree. Although I'm not content with RH's decisions during the last few months (especially when it comes to their layoffs), this one shouldn't even be a controversy at all. It's just a proposal for a community project. There will be no RedHat threats or higher-up decisions without the consent of the community. That's the main difference between Fedora (which does have deep ties with RedHat but is through and through a community project) and RedHat (the downstream entreprise).

I'd think it should be opt in, but they're not even willing to entertain the idea of having neither selected and have the user choose on the basis they'd typically select to opt out. Perhaps that means you're in the wrong.

1 more...

A pop-up with a message "This website uses cookies to function. The compliance people asked us to tell you."

Really? How am I supposed to trust these guys with my data, when they show me a pop-up like that? I guess that this data is anonymous only because otherwise the telemetry would need to be opt-out… oh wait, they explicitly said that:

That said, Fedora Legal has determined that if we collect any personally-identifiable data, the entire metrics system must be opt-in. Since we are only interested in opt-out metrics due to the low value of opt-in metrics, we must accordingly never collect any personally-identifiable data.

That's one of the worst cookie banners I've ever seen

It's completely useless. If they don't collect personal data, then they don't have to show a banner. Bad faith compliance with GDPR (including "the GDPR made us show this banner") should be punished more heavily.

If this gets implemented, I’ll certainly be switching distros. That opt-out telemetry is even a serious proposal is potentially enough for me to switch, as it indicates where the Fedora team’s heads are. I’ve been using and recommending Fedora for over a decade, but this is crosses a line.

Please read the thing properly. It is nothing more than a proposal currently, which is being actively discussed with the community to make sure that users (including you) are satisfied with the result if it is implemented, or to make sure it is not implemented at all.

Reading through the post it looks like the project leads (Fedora council members) are arguing in favour of "opt-out" and the larger community is arguing in favour or either opt-in or a middle ground where the user has to select an option with no default.

Honestly it seems like the Fedora team is arguing that there are only two options: opt-out, or nothing at all. This isn't true and people are commenting with more reasonable alternatives.

I know its not in development yet, but if the Fedora council members are saying "opt-out or nothing", not a good look TBH given this initial community response.

For as long as there are other distros to use, it will always be opt-in; shame, I liked Fedora.

Dammit. Just this year I settled on Fedora as my distro of choice. Back to Arch I guess. These fools are ruining Linux and I ain't happy about it.

gnome-initial-setup will default to displaying the toggle as enabled, even though the underlying setting will initially be disabled. (The underlying setting will not actually be enabled until the user finishes the privacy page, to ensure users have the opportunity to disable the setting before any data is uploaded.)

I see what they're saying here and how they're trying to give users options for users to opt-out before "accidentally" sharing data.

We are not interested in opt-in metrics. To make this a little more confusing, metrics collection is actually separate from uploading. Collection is always initially enabled, while uploading is always initially disabled. The graphical toggle enables or disables both at the same time.

Given all this I have even lower confidence that the opt-out will be bug free, especially over time. If the settings are separate then why just one toggle? If it were separate, I might want to collect then inspect the data and maybe even choose to share (or use the data myself some other way).

Few users would opt in...

Well yes, if it's all my way or the highway. I understand that this is a tough problem to solve and a tougher one to message correctly. Hopefully someone will figure it out one day.

Nice, I hope this helps improve the Fedora experience :)