4 ms·

There shouldn't be any telemetry in free software, end of. There are other ways to estimate the number of active users, e.g. through downloads and browser finge

by drnick1 14d ago

There shouldn't be any telemetry in free software, end of. There are other ways to estimate the number of active users, e.g. through downloads and browser fingerprints. Mozilla does not need more money than it already has. Ladybird has shown that you don't need a huge team to build a browser from scratch, so clearly Mozilla isn't spending all the money it gets from Google for development.

sourthyme 14d ago
Does Firefox sell this data? That would be bad and should be able to turn off. But there's other uses for telemetry that can help free software.
- JohnFen 14d ago
  Like with any software, you can't ever really be sure if they share this data with other entities. That's part of why I don't let software phone home, FOSS or not.
- throwa356262 14d ago
  Why gather this much data if you are not monetizeing it? It has a cost and can also become a liability. You don't do it just for fun.
  - coldfloor 14d ago
    Because it's useful in aiding in development. "Telemetry" is not "spying" or "collecting personal information" or whatever else uninformed people like this blog author try to dramatize it as. Telemetry is anonymous crash reports, how many users are using vertical vs horizontal tabs, how many people engaged the AI killswitch, etc. If it's NOT anonymized/bucketized usage date, then it's NOT telemetry. People need to stop conflating the two. Ironically, it's the people who disable all telemetry out of ignorance and paranoia that end up crying every time a developer deprecates their favorite feature or they have an issue with a new release. If you don't vote you don't get to complain. Mozilla's PM's in particular have repeatedly published and cited telemetry data in their decision making, and have referenced crash reports in proactively filed bugs.
    - JohnFen 13d ago
      > "Telemetry" is not "spying" or "collecting personal information" or whatever else uninformed people like this blog author try to dramatize it as. Whether or not telemetry counts as spying depends on whether or not informed consent was obtained before data is sent. If any data is collected about me, my machines, or my use of my machines, and I have not explicitly been told about and agreed to it first, then I'm being spied on.
- jeltz 14d ago
  Historically thet have done so but I have no idea if thet currently do.
kibwen 14d ago
> There are other ways to estimate the number of active users The point of telemetry is not to estimate the number of active users, and focusing on "telemetry" is the wrong bogeyman. Things like crash reports are telemetry, and are perfectly acceptable as long as the user remains in control over whether or not to send them. Software should respect user privacy and user control.
- wtallis 14d ago
  At some point, the pervasive unethical behavior of having telemetry on by default instead of opt-in, not just in proprietary software but also in open-source software and even from Mozilla itself becomes overwhelming enough that "telemetry=bad" becomes a reasonable summary of the situation. We need a new word specifically for ethical opt-in data collection.
demibabs 14d ago
I don’t get this take. There’s good reasons to have telemetry in any software. Doesn’t really have to do with whether it’s free or not. Of course, it’s also good to have the option to disable it.
- ginko 14d ago
  Telemetry should be opt-in.
nickthegreek 14d ago
> There shouldn't be any telemetry in free software, end of. Don't install it if you dont want it, that is where you choice lies. Telemetry inclusion is up to the author. Good ones will make it opt-in if it is truly necessary to include. > Ladybird has shown that you don't need a huge team to build a browser from scratch Ladybird has done no such thing.
zamadatix 14d ago
This is like saying you shouldn't have a computer because you can do addition yourself. Okay, what about all of the other use cases? Besides, collecting stats based on downloads/updates/etc is still telemetry. I'm perfectly fine with telemetry in any software, free or paid. Just make it opt-in and don't turn into cookie notice style malicious compliance about that.
userbinator 14d ago
This, absolutely. I used to publish free software for download, and I had absolutely zero idea how it was used, nor did I care. The problem is that the surveillance mindset has infected developers these days so now you have a bunch of telemetry apologists and advocates influencing the narrative. Unfortunately they are themselves fine with being spied on... although the inevitable leak happens and hopefully some of them start to worry.

by drnick1 14d ago

sourthyme 14d ago

Does Firefox sell this data? That would be bad and should be able to turn off. But there's other uses for telemetry that can help free software.

JohnFen 14d ago
Like with any software, you can't ever really be sure if they share this data with other entities. That's part of why I don't let software phone home, FOSS or not.
throwa356262 14d ago
Why gather this much data if you are not monetizeing it? It has a cost and can also become a liability. You don't do it just for fun.
- coldfloor 14d ago
  Because it's useful in aiding in development. "Telemetry" is not "spying" or "collecting personal information" or whatever else uninformed people like this blog author try to dramatize it as. Telemetry is anonymous crash reports, how many users are using vertical vs horizontal tabs, how many people engaged the AI killswitch, etc. If it's NOT anonymized/bucketized usage date, then it's NOT telemetry. People need to stop conflating the two. Ironically, it's the people who disable all telemetry out of ignorance and paranoia that end up crying every time a developer deprecates their favorite feature or they have an issue with a new release. If you don't vote you don't get to complain. Mozilla's PM's in particular have repeatedly published and cited telemetry data in their decision making, and have referenced crash reports in proactively filed bugs.
  - JohnFen 13d ago
    > "Telemetry" is not "spying" or "collecting personal information" or whatever else uninformed people like this blog author try to dramatize it as. Whether or not telemetry counts as spying depends on whether or not informed consent was obtained before data is sent. If any data is collected about me, my machines, or my use of my machines, and I have not explicitly been told about and agreed to it first, then I'm being spied on.
jeltz 14d ago
Historically thet have done so but I have no idea if thet currently do.

kibwen 14d ago

> There are other ways to estimate the number of active users The point of telemetry is not to estimate the number of active users, and focusing on "telemetry" is the wrong bogeyman. Things like crash reports are telemetry, and are perfectly acceptable as long as the user remains in control over whether or not to send them. Software should respect user privacy and user control.

wtallis 14d ago
At some point, the pervasive unethical behavior of having telemetry on by default instead of opt-in, not just in proprietary software but also in open-source software and even from Mozilla itself becomes overwhelming enough that "telemetry=bad" becomes a reasonable summary of the situation. We need a new word specifically for ethical opt-in data collection.

demibabs 14d ago

I don’t get this take. There’s good reasons to have telemetry in any software. Doesn’t really have to do with whether it’s free or not. Of course, it’s also good to have the option to disable it.

ginko 14d ago
Telemetry should be opt-in.

nickthegreek 14d ago

> There shouldn't be any telemetry in free software, end of. Don't install it if you dont want it, that is where you choice lies. Telemetry inclusion is up to the author. Good ones will make it opt-in if it is truly necessary to include. > Ladybird has shown that you don't need a huge team to build a browser from scratch Ladybird has done no such thing.

zamadatix 14d ago

This is like saying you shouldn't have a computer because you can do addition yourself. Okay, what about all of the other use cases? Besides, collecting stats based on downloads/updates/etc is still telemetry. I'm perfectly fine with telemetry in any software, free or paid. Just make it opt-in and don't turn into cookie notice style malicious compliance about that.

userbinator 14d ago

This, absolutely. I used to publish free software for download, and I had absolutely zero idea how it was used, nor did I care. The problem is that the surveillance mindset has infected developers these days so now you have a bunch of telemetry apologists and advocates influencing the narrative. Unfortunately they are themselves fine with being spied on... although the inevitable leak happens and hopefully some of them start to worry.