5 ms·

> Ubuntu with its Unity design first, then Snap, put desktop Linux back a decade. What a way to snatch defeat from the jaws of victory. A SUV. Strongly disagre

> Ubuntu with its Unity design first, then Snap, put desktop Linux back a decade. What a way to snatch defeat from the jaws of victory. A SUV. Strongly disagree. It seems to me that Ubuntu was( and still is) the main driver behind most of the improvements in the linux desktop space. The linux desktop experiences lagged being MacOS and windows for years. Even after ubuntu transitioned back to gnome, they still made a lot of improvements which lead to significantly better UX. The curse of linux desktop, is that the linux community seems to be allergic to any changes be it systemd,unity or snap packages.

linguae 2y ago
For better or worse, part of the story of the FOSS ecosystem is that there is no singular "Linux community." Rather, there are many communities with different interests, some of them overlapping, some of them complementary, and some of them competing. There are plenty of detractors to modern technologies such as systemd and Wayland, but there are also plenty of supporters (or else those technologies wouldn't have gotten adopted in the first place), and there are people who don't have a passionate opinion one way or the other as long as their software works. This lack of a singular community is simultaneously a blessing and a curse. It's a blessing due to the freedom FOSS software provides; not all software is going to fit everybody's needs, and FOSS allows the freedom to modify software tools. However, the curses are resources spread thinly across competing projects, as well as the rough, sometimes chaotic development cycles that sometimes occur in bazaar-style projects that lack Steve Jobs-esque "benevolent dictators" who have a clear vision for the product being developed. For example, the transition from X to Wayland would've probably gone smoother if there were a powerful entity that was able to enforce its will among Linux users and developers, but since there is no singular Linux community, the "bazaar" independently adopted Wayland at each project's own pace. This is the fundamental difference between Linux versus platforms whose vendors exert greater control, such as Windows and especially macOS.
- soulbadguy 2y ago
  > It's a blessing due to the freedom FOSS software provides; not all software is going to fit everybody's needs, and FOSS allows the freedom to modify software tools. In my view, the lack of singular vision and focus is not something intrinsic to FOSS. When i look like at projects like blender3d, Cpython, linux kernel,LLVM etc... I think a project can be fully OSS, with a degree of cohesion and focus conducive to faster interation and improvements. I would even go as far as saying that the most successfull projects tends to be that way. The linux desktop space just seems to be particularly fragmented. Seems to be mainly from the historical of a system born mainly from command line space, a weird attachment to a certain "old ways" of doing things etc... And a whole lot of strong personalities too busy bike shedding to see the forest from the tree. > For example, the transition from X to Wayland would've probably gone smoother if there were a powerful entity that was able to enforce its will among Linux users and developers, but since there is no singular Linux community, the "bazaar" independently adopted Wayland at each project's own pace. That's fair. A strong leader would have helped. But i think more than a strong leader, a shift in the community culture is more what's needed. But i am not holding my breath. People interested in desktop OS are either using MacOS or windows 11.
  - flakeoil 2y ago
    > In my view, the lack of singular vision and focus is not something intrinsic to FOSS. When i look like at projects like blender3d, Cpython, linux kernel,LLVM etc... I think one of the differences here is that the Desktop is more of an aesthetics and ergonomic thing compared to your examples above. It's harder for people to agree on and when there are subjective feelings around it instead of objective, there is a need for a central decision point which takes quick decisions.
    - soulbadguy 2y ago
      To a certain extend i agree. I can see how we can see the world of easthetism less objective and therefore harder to drive consensus around. But i don't think the problem with linux desktop is the easthetics. Agreeing on the objectives and fundational technical decisions. For example, kernel scheduling policy influencing latency and desktop responsiveness, X vs wayland and the impact on things like touch screen etc...
sevg 2y ago
Canonical tried and failed many times to be the driver behind improvements to Linux and the Linux desktop. They want control and influence across the stack, but so far their biggest bets have mostly been failures. They pushed Upstart. It fell to systemd. They pushed Unity. It fell to GNOME. They pushed Mir. It fell to Wayland. They pushed bazaar. It fell to git. And so on. Also worth noting that they've never been big contributors to the Linux kernel itself. Top contributors include Intel, AMD, Red Hat, Google, Meta, Linaro, SUSE. Even Oracle. Canonical? Nowhere to be seen.
- soulbadguy 2y ago
  Damn if you do, damn if you don't... Canonical approaches has always been to develop missing pieces in the desktop eco-system. And use anything external when possible. Unity started more than a year before gnome 3, and lasted a good 10 years. Calling it a failure seem exagerated. Same for upstart,mir and bazaar. At the moment of their inceptions, the landscape was very different.
  - Kwpolska 2y ago
    Wayland is older than Mir, and Mir was a failure that never really shipped. Bazaar is only one week older than Git.
    - seba_dos1 2y ago
      Even if it was one week younger, it wouldn't really change anything in the point made by GP. Mir, however, was a mistake from the very beginning indeed.
- grumpyprole 2y ago
  I don't think the "failures" are necessarily always technical, e.g. upstart versus systemd or GNOME versus Unity. To me it looks like Red Hat has possibly too much control and influence.
- pauby 2y ago
  You forgot the disaster that is Snap.
marcus_holmes 2y ago
I would agree, but snap is a mess. I have core snaps that are out of date, can't get rid of them, can't update them, can't fix them. I think I have to uninstall the whole service and reinstall to fix it. Meh, I'll just change distro and fix it that way. Luckily I have all the control and all the choices. I like systemd, but I do understand the objections to it. Luckily if you don't want to run systemd, you don't have to. I don't like Unity, which is fine because there are tons of other options. The article author loves MacOS's look and feel. I don't. So I'm screwed because MacOS only comes in one flavour. I don't think the article gives enough coverage to this. It's not just "tinkering" with Linux. We have real choices, and it's the only one of the three where that's the case. That's way more important (to me) than a lot of the other things they talk about.
- soulbadguy 2y ago
  # We have real choices, and it's the only one of the three where that's the case. Choice is not a substitute for quality. It doesn't really matter if i have the choice of 23 sedans if what i need is truck. In many ways, i feel like the choices in the linus desktop ecosystem are an illusion. Repackaking and slicing of the same software stack because someone doesn't like a color theme. And the price we paying for this illusion is the fragmentation and lack of cohesive experience afforded to other major OS. # That's way more important (to me) than a lot of the other things they talk about. That's fair. But i think the point is that you might be different than the vast majority of desktop user, for which a OS is just a mean to an end.

linguae 2y ago

For better or worse, part of the story of the FOSS ecosystem is that there is no singular "Linux community." Rather, there are many communities with different interests, some of them overlapping, some of them complementary, and some of them competing. There are plenty of detractors to modern technologies such as systemd and Wayland, but there are also plenty of supporters (or else those technologies wouldn't have gotten adopted in the first place), and there are people who don't have a passionate opinion one way or the other as long as their software works. This lack of a singular community is simultaneously a blessing and a curse. It's a blessing due to the freedom FOSS software provides; not all software is going to fit everybody's needs, and FOSS allows the freedom to modify software tools. However, the curses are resources spread thinly across competing projects, as well as the rough, sometimes chaotic development cycles that sometimes occur in bazaar-style projects that lack Steve Jobs-esque "benevolent dictators" who have a clear vision for the product being developed. For example, the transition from X to Wayland would've probably gone smoother if there were a powerful entity that was able to enforce its will among Linux users and developers, but since there is no singular Linux community, the "bazaar" independently adopted Wayland at each project's own pace. This is the fundamental difference between Linux versus platforms whose vendors exert greater control, such as Windows and especially macOS.

soulbadguy 2y ago
> It's a blessing due to the freedom FOSS software provides; not all software is going to fit everybody's needs, and FOSS allows the freedom to modify software tools. In my view, the lack of singular vision and focus is not something intrinsic to FOSS. When i look like at projects like blender3d, Cpython, linux kernel,LLVM etc... I think a project can be fully OSS, with a degree of cohesion and focus conducive to faster interation and improvements. I would even go as far as saying that the most successfull projects tends to be that way. The linux desktop space just seems to be particularly fragmented. Seems to be mainly from the historical of a system born mainly from command line space, a weird attachment to a certain "old ways" of doing things etc... And a whole lot of strong personalities too busy bike shedding to see the forest from the tree. > For example, the transition from X to Wayland would've probably gone smoother if there were a powerful entity that was able to enforce its will among Linux users and developers, but since there is no singular Linux community, the "bazaar" independently adopted Wayland at each project's own pace. That's fair. A strong leader would have helped. But i think more than a strong leader, a shift in the community culture is more what's needed. But i am not holding my breath. People interested in desktop OS are either using MacOS or windows 11.
- flakeoil 2y ago
  > In my view, the lack of singular vision and focus is not something intrinsic to FOSS. When i look like at projects like blender3d, Cpython, linux kernel,LLVM etc... I think one of the differences here is that the Desktop is more of an aesthetics and ergonomic thing compared to your examples above. It's harder for people to agree on and when there are subjective feelings around it instead of objective, there is a need for a central decision point which takes quick decisions.
  - soulbadguy 2y ago
    To a certain extend i agree. I can see how we can see the world of easthetism less objective and therefore harder to drive consensus around. But i don't think the problem with linux desktop is the easthetics. Agreeing on the objectives and fundational technical decisions. For example, kernel scheduling policy influencing latency and desktop responsiveness, X vs wayland and the impact on things like touch screen etc...

sevg 2y ago

Canonical tried and failed many times to be the driver behind improvements to Linux and the Linux desktop. They want control and influence across the stack, but so far their biggest bets have mostly been failures. They pushed Upstart. It fell to systemd. They pushed Unity. It fell to GNOME. They pushed Mir. It fell to Wayland. They pushed bazaar. It fell to git. And so on. Also worth noting that they've never been big contributors to the Linux kernel itself. Top contributors include Intel, AMD, Red Hat, Google, Meta, Linaro, SUSE. Even Oracle. Canonical? Nowhere to be seen.

soulbadguy 2y ago
Damn if you do, damn if you don't... Canonical approaches has always been to develop missing pieces in the desktop eco-system. And use anything external when possible. Unity started more than a year before gnome 3, and lasted a good 10 years. Calling it a failure seem exagerated. Same for upstart,mir and bazaar. At the moment of their inceptions, the landscape was very different.
- Kwpolska 2y ago
  Wayland is older than Mir, and Mir was a failure that never really shipped. Bazaar is only one week older than Git.
  - seba_dos1 2y ago
    Even if it was one week younger, it wouldn't really change anything in the point made by GP. Mir, however, was a mistake from the very beginning indeed.
grumpyprole 2y ago
I don't think the "failures" are necessarily always technical, e.g. upstart versus systemd or GNOME versus Unity. To me it looks like Red Hat has possibly too much control and influence.
pauby 2y ago
You forgot the disaster that is Snap.

marcus_holmes 2y ago

I would agree, but snap is a mess. I have core snaps that are out of date, can't get rid of them, can't update them, can't fix them. I think I have to uninstall the whole service and reinstall to fix it. Meh, I'll just change distro and fix it that way. Luckily I have all the control and all the choices. I like systemd, but I do understand the objections to it. Luckily if you don't want to run systemd, you don't have to. I don't like Unity, which is fine because there are tons of other options. The article author loves MacOS's look and feel. I don't. So I'm screwed because MacOS only comes in one flavour. I don't think the article gives enough coverage to this. It's not just "tinkering" with Linux. We have real choices, and it's the only one of the three where that's the case. That's way more important (to me) than a lot of the other things they talk about.

soulbadguy 2y ago
# We have real choices, and it's the only one of the three where that's the case. Choice is not a substitute for quality. It doesn't really matter if i have the choice of 23 sedans if what i need is truck. In many ways, i feel like the choices in the linus desktop ecosystem are an illusion. Repackaking and slicing of the same software stack because someone doesn't like a color theme. And the price we paying for this illusion is the fragmentation and lack of cohesive experience afforded to other major OS. # That's way more important (to me) than a lot of the other things they talk about. That's fair. But i think the point is that you might be different than the vast majority of desktop user, for which a OS is just a mean to an end.