4 ms·

I believe you, but you also have to understand this is largely a preference thing. Which is okay: It's all just whatever gets the job done in the end! I'd say

I believe you, but you also have to understand this is largely a preference thing. Which is okay: It's all just whatever gets the job done in the end! I'd say I have the opposite experience: I've had to use the Windows Desktop for extended periods before, and found it, as you put it, "completely backwards way of doing just about everything I want to do with a computer". I even went through all sorts of lengths to "fix" it with alternative WMs, launchers, etc, but as no matter how much I tinkered with it, I couldn't even get fairly "vanilla" features (eg multidesktop support) sufficiently working without constant papercuts and limitations. Of course, things may have improved since I last used it, so I'm sure Windows Desktop circa 2020 has much improved, but a few years back it seemed very broken. Ultimately, it's just preference, and what workflows you grow used to.

AnIdiotOnTheNet 5y ago
Sure, if you like the way it works then that's great for you, but I think it is worth noting that despite the stated desire of many in the Linux Desktop community to see the fabled "Year of the Linux Desktop", it still languishes in a distant 3rd place. I personally think that's because it has deeply rooted systemic problems that it has refused to address in any meaningful way for the past 20 years, which is incredibly frustrating for those of us who'd like a real alternative to Windows for what we do.
- michaelpb 5y ago
  True, but if the goal is to no longer languish in a distant 3rd place, then the power-user features that we find lacking on Lin/Win are a red-herring, since we constitute a tiny fraction of computer users. From my experience, I don't think Lin/Win DEs are actually so bad so as to affect mass adaption. I've introduced dozens of new students to Linux, and have had largely positive results, and also generally found the Windows -> Ubuntu switch easier than Windows -> macOS (assuming all are pre-installed) And that last assumption is truly the hard part. There is no software meritocracy. There is no "if you build it, they will come". It's sales all the way down. Until this changes, whatever is outputted from FAANG/MS market research firms will be what's popular, end of story.
  - AnIdiotOnTheNet 5y ago
    To be frank, I think this line of reasoning is something Linux Desktop evangelists tell themselves so they can feel better about their abject failure to gain any significant desktop market share. For starters, the separation of the world into "power users" and "normal users" is, to me, a sign of flawed thinking. The goal of a personal computing system should be to allow a complete novice to onramp into a "power user" as their understanding of the system grows. Discoverability and consistency are absolutely key to this and Linux DEs have always been terrible at both (and other Desktops are getting worse too). There are many more problems I could go on about, like walled gardens like Apple's App store or Linux's package repos creating another boundary between users and creators that need not exist, for example. Point being, separating the audience into "people who get shit, like me" and "people who just want X" is an excuse for having a bad system. Instead of admitting and attempting to address the issue, you instead get to say "well, normal users don't need that and power users should be able to deal with the bullshit".
    - michaelpb 5y ago
      My point is that I'm not sure even how I would "address" these issues that you are insisting Linux Desktop evangelist "stop excusing". That is, 1) I'm not getting paid to work on Linux DEs or have the time (right now) to volunteer, and 2) I don't think you could get 10 people in a room to agree how a DE should behave. Like, what would I change to what? I've consumed plenty of blog posts complaining about UI inconsistencies on Windows, macOS, or Linux, but these all still seem like preferences with no single "correct" answer as to how a GUI should be. Even taking the one concrete example you provided in this thread, Apple's App Store is incredibly popular -- in fact in the US it has a 56% marketshare [1] -- so why would a Linux OS moving toward what you prefer would increase adaption? I really see no actual evidence of this; as in the example you give is literally the most popular way to do it. That said, I don't even think it matters either way. I've grown cynical enough by now to realize that humans are a lot more complicated, markets more monopolistic, and purchasing habits are better explained by market researchers than developers (or even HCI researchers!) [1] https://gs.statcounter.com/os-market-share/mobile/united-states-of-america https://gs.statcounter.com/os-market-share/mobile/united-sta...

AnIdiotOnTheNet 5y ago

Sure, if you like the way it works then that's great for you, but I think it is worth noting that despite the stated desire of many in the Linux Desktop community to see the fabled "Year of the Linux Desktop", it still languishes in a distant 3rd place. I personally think that's because it has deeply rooted systemic problems that it has refused to address in any meaningful way for the past 20 years, which is incredibly frustrating for those of us who'd like a real alternative to Windows for what we do.

michaelpb 5y ago
True, but if the goal is to no longer languish in a distant 3rd place, then the power-user features that we find lacking on Lin/Win are a red-herring, since we constitute a tiny fraction of computer users. From my experience, I don't think Lin/Win DEs are actually so bad so as to affect mass adaption. I've introduced dozens of new students to Linux, and have had largely positive results, and also generally found the Windows -> Ubuntu switch easier than Windows -> macOS (assuming all are pre-installed) And that last assumption is truly the hard part. There is no software meritocracy. There is no "if you build it, they will come". It's sales all the way down. Until this changes, whatever is outputted from FAANG/MS market research firms will be what's popular, end of story.
- AnIdiotOnTheNet 5y ago
  To be frank, I think this line of reasoning is something Linux Desktop evangelists tell themselves so they can feel better about their abject failure to gain any significant desktop market share. For starters, the separation of the world into "power users" and "normal users" is, to me, a sign of flawed thinking. The goal of a personal computing system should be to allow a complete novice to onramp into a "power user" as their understanding of the system grows. Discoverability and consistency are absolutely key to this and Linux DEs have always been terrible at both (and other Desktops are getting worse too). There are many more problems I could go on about, like walled gardens like Apple's App store or Linux's package repos creating another boundary between users and creators that need not exist, for example. Point being, separating the audience into "people who get shit, like me" and "people who just want X" is an excuse for having a bad system. Instead of admitting and attempting to address the issue, you instead get to say "well, normal users don't need that and power users should be able to deal with the bullshit".
  - michaelpb 5y ago
    My point is that I'm not sure even how I would "address" these issues that you are insisting Linux Desktop evangelist "stop excusing". That is, 1) I'm not getting paid to work on Linux DEs or have the time (right now) to volunteer, and 2) I don't think you could get 10 people in a room to agree how a DE should behave. Like, what would I change to what? I've consumed plenty of blog posts complaining about UI inconsistencies on Windows, macOS, or Linux, but these all still seem like preferences with no single "correct" answer as to how a GUI should be. Even taking the one concrete example you provided in this thread, Apple's App Store is incredibly popular -- in fact in the US it has a 56% marketshare [1] -- so why would a Linux OS moving toward what you prefer would increase adaption? I really see no actual evidence of this; as in the example you give is literally the most popular way to do it. That said, I don't even think it matters either way. I've grown cynical enough by now to realize that humans are a lot more complicated, markets more monopolistic, and purchasing habits are better explained by market researchers than developers (or even HCI researchers!) [1] https://gs.statcounter.com/os-market-share/mobile/united-states-of-america https://gs.statcounter.com/os-market-share/mobile/united-sta...