“Canonical keeping up by taking control of LXD (monopoly!)” - it was revealed today that this was motivated by the lead of the LXD project leaving Canonical.
Want to discuss particular failings in the governance models of Ubuntu or Fedora? Fine. Let's talk about it. We can even discuss if that specific feature of the governance model is similar to an action Microsoft is/has taken.
Vague statements about two different governance models "turning into" a monopoly OS? There's no cogent baseline here to talk about. It's just clickbait and belly aching.
Garbage article. The author claims that Flatpak (which is decentralized) is an attempt at centralizing software since Flathub (not controlled by IBM/Red Hat) is popular. Also any telemetry is awful, even telemetry that is open and optional.
I'm going to be controversial and disagree with this. If you're being onboarded to an app, and there's a "enable telemetry?" button and it's clear and checked by default, that's not awful. If your data is sufficiently anonymized, that's the only way you're going to get actual user data, otherwise it's skewed to be those that go out of the way to enable it - in other words, not normal users.
No, data collection ought to always be based on informed consent. A default enabled checkbox that can easily be skipped over does not constitute informed consent.
I agree that opt-in telemetry however is useless.
Thus, the only ethical and useful option is forced choice; i.e. present the user with two unselected options where a decision between them must be made.
This is e.g. how Apple designed its usage data option for the Apple TV, AFAIK. Two unselected options, given equal weight and styling etc.
By that logic, it should be fine to have a big obvious "enable telemetry?" button that's disabled by default, right? Because otherwise you're just assuming that people will ignore it and leave things on the default setting, and claiming that it's okay to exploit that to enable telemetry.
First thing I did on my Ubuntu install was remove the Firefox snap, add the Mozilla ppa, and install Firefox from there via apt.
If Mozilla maintains its own ppa and they do all the hard working of getting everything working, I don’t understand why Canonical has to push a snap option instead.
I did give it a try, but only to get obnoxious popups about being forced to restart Firefox every other day for snap to update, and these popups would even bring my monitor out of dpms (and maybe my machine out of sleep?) to inform me of this. But even heeding these prompts, Firefox would still update in the background leaving me unable to open a new tab because IPC broke since the master and slave processes were now running different Firefox versions..
FYI, Mozilla also maintains the snap and the flathub images, so that part of the argument doesn't have much traction imho.
Can't speak to waking from a sleep state (even just screen off)... that said, having an evergreen browser is pretty important, given how bad 0-day exploits can be. This is also part of why Snaps instead of the repo/ppa for such things as GUI apps. Snap and Flatpak both have advantages and disadvantages over each other, technical and ethical.
That said, by using such an abstraction for common applications, it allows for the applications to be released ahead of core OS changes. It allows the delivery of both application features and security updates without risk to the core os running.
I know some really hate AppImage/Flatpak/Snaps etc. In the end, it's better in a lot of ways and there are distros for those that don't want to participate. It's still likely the best option for most apps for most people. It allows for developers of software to generate releases that can be consumed from any Linux distro without having to worry about specific downstream distros potentially breaking things, or worse, running unmaintained, years old, versions that are broken.
If you wanted to make some kind of argument about what's going on with RHEL, I'd hear you out... but Fedora is and remains a community project. While Red Hat certainly has a strong voice in the team meetings, development continues to occur in the open. And while it is true that the Fedora project is considering adding telemetry to Fedora 40, they're openly discussing what kinds of telemetry are ethical before even committing to adding any telemetry. It's a stretch to equate this immediately to Microsoft Windows.
> We believe an open source community can ethically collect limited aggregate data on how its software is used without involving big data companies or building creepy tracking profiles that are not in the best interests of users. Users will have the option to disable data upload before any data is sent for the first time. Our service will be operated by Fedora on Fedora infrastructure, and will not depend on Google Analytics or any other controversial third-party services. And in contrast to proprietary software operating systems, you can redirect the data collection to your own private metrics server instead of Fedora’s to see precisely what data is being collected from you, because the server components are open source too.[1]
While I'd personally love to see opt-in rather than opt-out telemetry, I believe in the value that telemetry can provide to projects hoping to improve their software/OS offerings. I'm glad to see open discussion on the subject of how to do so ethically.
>> While I'd personally love to see opt-in rather than opt-out telemetry
I think the vast majority of Linux users would not opt-in to any sort of telemetry. Some would consciously make a determination, but MS/Windows has trained the bulk of us to just say no to such things during any sort of install procedure. I personally would like telemetry to be done fully openly, with a running local text file showing in user-readable format exactly what has been shared in plain English.
> Fedora is and remains a community project. While Red Hat certainly has a strong voice in the team meetings, development continues to occur in the open.
https://docs.fedoraproject.org/en-US/council/members/ lists 7 people. Of these, 6 appear to be Red Hat employees (the other 1 doesn't list employment, so I'll assume it isn't RH). I'll agree that it's developed in the open, just like CentOS Stream, but I struggle to see how Fedora - RHEL's upstream, the continuation of Red Hat Linux, a project whose trademarks are owned by Red Hat, whose leadership is 6/7ths RH employees - is anything but a project by Red Hat.
This isn't accurate. Only ~30% of Fedora contributions come from Red Hatters, they specifically changed the council structure so much more appointments can be selected by the community rather than Red Hat:
"Historical Note
The previous previous governance structure (Fedora Board) had five members directly appointed by Red Hat and five elected at large. The current structure is more complicated but has a much greater proportion of members selected by the community by election or merit. In the previous board structure, the Fedora Project leader had a special veto power; in the current model, all voting members can block on issues, with a valid reason. The FPL does not have a special veto, but does have a limited power to “unstick” things if consensus genuinely can’t be reached and a decision needs to be made."
Note that the community can and does select Red Hatters willingly if they are the best fit for the position. It doesn't change the fact that the community is still in full control of the project, Red Hat just sponsors it. Fedora could go off in whatever direction they want, and assuming the community supports it, Red Hat can't do anything about this, many community members have made this very clear over the years, and they also lay it out in the docs:
https://docs.fedoraproject.org/en-US/quick-docs/fedora-and-r...
That's (one of the reasons) why Red Hat forks Fedora and then makes whatever changes they need to make.
19 comments
[ 3.5 ms ] story [ 40.8 ms ] threadWant to discuss particular failings in the governance models of Ubuntu or Fedora? Fine. Let's talk about it. We can even discuss if that specific feature of the governance model is similar to an action Microsoft is/has taken.
Vague statements about two different governance models "turning into" a monopoly OS? There's no cogent baseline here to talk about. It's just clickbait and belly aching.
I agree that opt-in telemetry however is useless.
Thus, the only ethical and useful option is forced choice; i.e. present the user with two unselected options where a decision between them must be made.
This is e.g. how Apple designed its usage data option for the Apple TV, AFAIK. Two unselected options, given equal weight and styling etc.
- - -
First thing I did on my Ubuntu install was remove the Firefox snap, add the Mozilla ppa, and install Firefox from there via apt.
If Mozilla maintains its own ppa and they do all the hard working of getting everything working, I don’t understand why Canonical has to push a snap option instead.
I did give it a try, but only to get obnoxious popups about being forced to restart Firefox every other day for snap to update, and these popups would even bring my monitor out of dpms (and maybe my machine out of sleep?) to inform me of this. But even heeding these prompts, Firefox would still update in the background leaving me unable to open a new tab because IPC broke since the master and slave processes were now running different Firefox versions..
https://news.ycombinator.com/item?id=31237822
Can't speak to waking from a sleep state (even just screen off)... that said, having an evergreen browser is pretty important, given how bad 0-day exploits can be. This is also part of why Snaps instead of the repo/ppa for such things as GUI apps. Snap and Flatpak both have advantages and disadvantages over each other, technical and ethical.
That said, by using such an abstraction for common applications, it allows for the applications to be released ahead of core OS changes. It allows the delivery of both application features and security updates without risk to the core os running.
I know some really hate AppImage/Flatpak/Snaps etc. In the end, it's better in a lot of ways and there are distros for those that don't want to participate. It's still likely the best option for most apps for most people. It allows for developers of software to generate releases that can be consumed from any Linux distro without having to worry about specific downstream distros potentially breaking things, or worse, running unmaintained, years old, versions that are broken.
> We believe an open source community can ethically collect limited aggregate data on how its software is used without involving big data companies or building creepy tracking profiles that are not in the best interests of users. Users will have the option to disable data upload before any data is sent for the first time. Our service will be operated by Fedora on Fedora infrastructure, and will not depend on Google Analytics or any other controversial third-party services. And in contrast to proprietary software operating systems, you can redirect the data collection to your own private metrics server instead of Fedora’s to see precisely what data is being collected from you, because the server components are open source too.[1]
While I'd personally love to see opt-in rather than opt-out telemetry, I believe in the value that telemetry can provide to projects hoping to improve their software/OS offerings. I'm glad to see open discussion on the subject of how to do so ethically.
[1] https://discussion.fedoraproject.org/t/f40-change-request-pr...
I think the vast majority of Linux users would not opt-in to any sort of telemetry. Some would consciously make a determination, but MS/Windows has trained the bulk of us to just say no to such things during any sort of install procedure. I personally would like telemetry to be done fully openly, with a running local text file showing in user-readable format exactly what has been shared in plain English.
https://docs.fedoraproject.org/en-US/council/members/ lists 7 people. Of these, 6 appear to be Red Hat employees (the other 1 doesn't list employment, so I'll assume it isn't RH). I'll agree that it's developed in the open, just like CentOS Stream, but I struggle to see how Fedora - RHEL's upstream, the continuation of Red Hat Linux, a project whose trademarks are owned by Red Hat, whose leadership is 6/7ths RH employees - is anything but a project by Red Hat.
"Historical Note The previous previous governance structure (Fedora Board) had five members directly appointed by Red Hat and five elected at large. The current structure is more complicated but has a much greater proportion of members selected by the community by election or merit. In the previous board structure, the Fedora Project leader had a special veto power; in the current model, all voting members can block on issues, with a valid reason. The FPL does not have a special veto, but does have a limited power to “unstick” things if consensus genuinely can’t be reached and a decision needs to be made."
https://docs.fedoraproject.org/en-US/council/#_historical_no...
Note that the community can and does select Red Hatters willingly if they are the best fit for the position. It doesn't change the fact that the community is still in full control of the project, Red Hat just sponsors it. Fedora could go off in whatever direction they want, and assuming the community supports it, Red Hat can't do anything about this, many community members have made this very clear over the years, and they also lay it out in the docs: https://docs.fedoraproject.org/en-US/quick-docs/fedora-and-r...
That's (one of the reasons) why Red Hat forks Fedora and then makes whatever changes they need to make.