The Manjaro Linux team are asking for testers for their new data collection tool called "Manjaro Data Donor", which they plan to have as opt-out and not opt-in.
As expected, they are catching some comments from users not particularly happy about their plan to have it hooked up as something you have to opt out of due to privacy concerns. Telemetry is often a bad word to a lot of people, but the reality is that when done correctly it can be truly essential for projects to know where to focus.
Manjaro developer Roman Gilg said in the announcement that currently they use Matomo, which is pinged via the Network Manager but it's unreliable for various reasons. So they created Manjaro Data Donor (MDD) which gives more information and should be more reliable overall, and they actually have full control over it.
Currently they just want Manjaro users to test it and give feedback, but eventually it will be hooked up as a systemd service:
In the next few days we’ll do some more testing and if results are positive, I plan on installing it on all Manjaro systems and adding a systemd service to submit the data automatically.
As a reminder: Right now you have to install MDD manually and there is no systemd service yet.
With this systemd service later in place, sending the hardware data with MDD will be opt-out because I believe, if you do opt-in, the data you gather will be so heavily skewed you can just leave it be.
Let me know what you think. I know telemetry is a contentious subject, but we need at least some data about how Manjaro is being used by so many people around the world in order to show that the project has a future and also to plan for that future.
Quoting: Mountain ManIn fairness, it is rock solid stable now. At least, more stable than Arch (In my own experience).Quoting: BlackBloodRumI ran Gentoo many years ago and found it to be rather labor intensive and too easy to break even with a routine update. I imagine things have changed in the past 15 years, but I also have no compelling reason to abandon Manjaro at the moment since it has always just worked for me. I suppose that's boring to people who like to constantly tweak and tinker, but these days, I really don't mind boring!Quoting: Mountain ManWith all this doom and gloom talk about the end of Manjaro, I have to ask, what's the best alternative?Gentoo.
(Of course every Linux distro seems to be surrounded by pronouncements of its impending demise, but it rarely comes to pass.)
I know that feels like a meme, but it's really not. This thing is rock solid stable like a mountain, and man switching to it was the best choice I ever made. I don't even get all those little "odd bugs" you normally get on other distros.
You compile all the packages yourself which means you can patch out forced telemetry yourself or disable it at compile time (I do this for KDE). Binary packages are available now, but you lose some customizability.
Anything done in a way you don't like can basically be changed, no questions asked.
You don't *have* to compile anymore though. (Though I customize too much to be able to avoid it.)
There are things like btrfs snapshots which you could use for quick rollbacks though if updates break it.
Personally, I've come to love the feel of being able to change absolutely anything, it's something no other distribution can provide.
It takes a bit more time to setup and slightly more time to maintain, but it's not terrible. It's maybe an hour or two a week of my time to update it and make changes I like.
Quoting: dpanterQuoting: CZiNTrPTIn my country real organ donorship is opt-out and that's right approach as wellWhich country is that? Sounds like human rights violation to me.
In the Netherlands this is the case as well. Why would it be a human rights violation? You're not alive anymore when they do it. You can also always opt-out and this is widely known. The problem is that most people forgot or didn't bother to opt-in, which was costing lives.
As for the telemetry, it doesn't have to be all bad. Telemetry is important for devs to improve a project and like someone earlier in the thread has said, most people that don't care won't opt-in and the ones that do will opt-out. If they make it a clear toggle which is on by default and the data is non-personal, I don't see the issue. We're talking about PC specs here, not address info.
Last edited by Adutchman on 6 November 2024 at 6:52 am UTC
Quoting: AdutchmanQuoting: dpanterQuoting: CZiNTrPTIn my country real organ donorship is opt-out and that's right approach as wellWhich country is that? Sounds like human rights violation to me.
In the Netherlands this is the case as well. Why would it be a human rights violation? You're not alive anymore when they do it. You can also always opt-out and this is widely known. The problem is that most people forgot or didn't bother to opt-in, which was costing lives.
As for the telemetry, it doesn't have to be all bad. Telemetry is important for devs to improve a project and like someone earlier in the thread has said, most people that don't care won't opt-in and the ones that do will opt-out. If they make it a clear toggle which is on by default and the data is non-personal, I don't see the issue. We're talking about PC specs here, not address info.
On your first point I agree.
On the second I disagree.
A. pc specs can be more personal than you think. HWID's are unique and personally identifying, BIOS release date is a good indicator of when you bought the device and with that of when you had a personal shift requiring more fancy hardware, IP is enough to determine at least your nation, accesibility settings inform about someone's medical condition, Mac adresses inform whose hardware you bought, your device name is often your name, etc.
B. One of the primary selling points of Linux distros is their lack of spying. Lots of people use it, so their data doesn't get send to other parties. These people already made their choice and "opt-ed out" and now they have to do it another time.
C. For debugging it's mostly useless feature request and crash reports help much better there, because it clearly communicates customer demand and what's missing.
It's mostly useful for UI development and prioritization.
Prioritization currently happens on: "how easy is it to find volunteers and/or funds for it" and I don't think devs would like to let that power slip.
It's really useful for UI development, but a better fix for that is to sucker users into doing their own development by providing a forum and desktop setting sharing tools.
Also I don't actually believe they've the data analysis talent available to interpret this data.
It's not really something Os developers are known for.
Edit:
For me the base requirement for any telemetry to be considered acceptable it needs to be opt-in and employ differential privacy.
Last edited by LoudTechie on 6 November 2024 at 8:16 am UTC
Quoting: KlaasQuoting: BrokattTruth is telemetry is really useful for developers and without it's way too easy to make the wrong assumptions.And it is even easier to make wrong assumptions due to telemetry. Calling it a double-edged sword even with consent is putting it mildly.
I agree. That is true for all statistics though.
Quoting: BrokattQuoting: KlaasQuoting: BrokattTruth is telemetry is really useful for developers and without it's way too easy to make the wrong assumptions.And it is even easier to make wrong assumptions due to telemetry. Calling it a double-edged sword even with consent is putting it mildly.
I agree. That is true for all statistics though.
Most open source projects don't use statistics(, because they've no data).
They use reproducible and peer reviewable anecdotes(issues and pull requests) and expert review.
What you present here is a false dilemma.
What you two are comparing is the advantages and disadvantages of using telemetry yes/no, but in this case that doesn't mean they would use different statistics, so although it's true for all statistics it's not "true for all statistics though."
Edit: to put it in the mortal words of Ronald Munroe that's a false dichthomy.
Last edited by LoudTechie on 6 November 2024 at 8:41 am UTC
Quoting: dpanter> In the EU, most countries use an opt-out system (also known as "presumed consent") for organ donation, meaning that individuals are considered potential donors unless they have explicitly stated otherwise. Approximately 20 of the 27 EU countries operate under some form of opt-out system. This means around 74% of EU countries have opt-out laws for organ donation.Quoting: CZiNTrPTIn my country real organ donorship is opt-out and that's right approach as wellWhich country is that? Sounds like human rights violation to me.
Get with the program!
See more from me