6 ms·

Not getting this idea that Linux is freedom per se. Since the article was written, RedHat was acquired by IBM, and even before, Linux was largely controlled by

Not getting this idea that Linux is freedom per se. Since the article was written, RedHat was acquired by IBM, and even before, Linux was largely controlled by enterprise demands (systemd, namespaces/docker) and folks with their own agenda (glibc, gnome). Freedom is choice. Choice requires alternative implementations. Alternatives can and will be developed if there's a common spec, like POSIX, SUS, LSB, X, OpenGL used to be, all of which have atrophied. Linux doesn't inspire anyone to create new end-user apps (with the exception of Blender and Krita, which were started well over ten years ago). Instead, we're loosing the apps we had, and march towards dumb UIs and needlessly containerized apps a la snap/flatpack. It's of no use and only frustrating to lament and decry the past without also showing a perspective to get out of our situation, or clean up the mess for those who follow after us. F/OSS is clearly not the solution, as can be seen by the demise of the web going the way of all previous digital communication mechanisms: it has been foobar'd by the ad industry, like email, Usenet, XMPP before. Open, meaningful standards are the only solution. Not like "web standards" eg WHATWG and W3C. Who thought it was a good idea to hand standard development to an ad company in the first place? We now have a closed-loop web where the content provider and the user agent are from the same party. We had that with CompuServe et al before the web; there was no need for a craptastic CSS and JS detour to arrive at that situation.

JetAlone 4y ago
Over the last few years I've begun to understand that freedom in a too-real sense is the reward we experience for loving the truth, to the extent that we do.
mschuster91 4y ago
> Instead, we're loosing the apps we had, and march towards dumb UIs and needlessly containerized apps a la snap/flatpack. There is a very real need behind app containerization: distributions are exceedlingly slow to pick up updates in a time where web applications routinely run on a "merge request passed, 5 minutes later it's on production" model. On top of that, distributions routinely run into the problem that package A wants a new version of library B, but the other applications haven't been updated to use the new version yet, so everyone is stuck until all applications in the distribution have been updated. Also: Most people use LTS distributions and let's be real no one wants to be stuck with a 2 year old browser that only gets backported security patches. Distributions had the chance to move to a model of "the core is stable but the applications the user uses are flexibly updateable", and they blew it completely.
- tannhaeuser 4y ago
  The question you should be asking is why do we need all this if the desktop app "market" is contracting and I'm using just the same old apps on Linux I used 15 years ago. To the detriment of alt-OSs like FBSD and OBSD which I always loved. Why would I want frequent browser updates all the time when the web is 30 years old, and also contracting? As I type this (not making this up!) FF prompts me again to restart because updates - why? I'm sick of browsers updating more frequently than the very few sites I still go to; Moz should be producing an essential browser instead of participating in this drivel. It works just fine on my old 2016 XPS with Ubuntu ESR and infrequent updates. You can read my comments to see I've praised Ubuntu, Dell (XPS, specifically), and Lenovo in the past. But I'm on my third Dell (Latitude/Precision so no consumer crap) with fatal hardware defects ootb within 8 months that various customers found necessary to send me for work (to run stunts like Docker-on-Windows or something), while my Thinkpad's miniscule track pad with uneccessary mechanical parts is collecting dust, and mechanically blocking due to it. Right now I think I've got enough of it already and am going back to Apple which I've used on and off since around 2003.
  - jen20 4y ago
    A different question might be whether the native app market is contracting because operating system distributions handle updates so poorly.
- dTal 4y ago
  Setting aside the fact that rolling release distros are 1) a thing and 2) highly popular, I have to believe that these are problems that are within our capacity as engineers to solve without just saying "screw it, we'll give each program its own OS". In fact in some sense they are already solved, from a technical standpoint - you can install Debian, with its crufty out of date repos, and then download Firefox from its official website and run it right in your home directory. There you go, stable core and bleeding edge application. What's the problem?
  - AnIdiotOnTheNet 4y ago
    Consider that FlatPak is popular for this case and distributing the way you describe isn't for a good reason: having to deal with all the many and varied dependencies and configurations on any given distro is a giant pain in the ass. People have been screaming this at Linux Desktop evangelists for 20 years about this and mostly fallen on deaf ears, so it is no wonder Linux Desktop is still in a distant 3rd place in the desktop market.
microtonal 4y ago
Linux doesn't inspire anyone to create new end-user apps (with the exception of Blender and Krita, which were started well over ten years ago). Maybe it's because the community is full of people who see any change as an attack on how we used to do things, breaking the UNIX philosophy, etc.? This closed-mindedness leads to thinking like: Instead, we're loosing the apps we had, and march towards dumb UIs and needlessly containerized apps a la snap/flatpack. There is a lot of innovation in this space. OSTree (which is also used by Flatpak), replaces the old Unix filesystem model by something that is exciting and much better (atomic upgrades, rollbacks, signed system images, etc.). Flatpak itself finally makes it much easier to deploy desktop applications, which is long-needed to extend Linux beyond the traditional UNIX crowd.
- tannhaeuser 4y ago
  > see any change as an attack on how we used to do things, breaking the UNIX philosophy, etc. Yeah let's throw portability and any semblance of order away b/c of ... what exactly? > replaces the old Unix filesystem model by something that is exciting :shudder: Count me in on the closed-minded I guess ;)
  - microtonal 4y ago
    Yeah let's throw portability and any semblance of order away b/c of ... what exactly? There are only two Unix-like systems left that matter: macOS and Linux. No need to hold them back by pretending we are still in the UNIX wars and retaining compatibility with some UNIX system that nobody really uses anymore.
    - pjmlp 4y ago
      Linux distribution wars begat UNIX wars.
      - simiones 4y ago
        I think it's the other way around: the older UNIX wars begat the newer Linux distro wars.
everdrive 4y ago
>Linux doesn't inspire anyone to create new end-user apps I'm not really convinced there are more worthwhile apps on Windows / Mac, except for the usual big players. (office, photoshop, etc) I'm sure there are some edge cases here, but for normal users it's not clear this is an issue. Most just use a web browser. They might not even make much use of a file manager. I'm not suggesting that the application situation is perfect on Linux, but it just seems to be bad everywhere. It seems like simple, single-purpose apps are disappearing, and are being replaced by a constantly-changing sea of noise. Is App "X" a good app? Maybe it was a few years ago but the developer tried to monetize it and ruined the app with ads / extraneous features / countless UI redesigns, etc. It's gotten to the point where the static nature of Linux apps is a feature. Rhythmbox is far from a perfect music player, but one of the things it's got over it's major competitors is that it's lightweight and hasn't changed in years. I don't need to constantly re-learn something basic just because an application developer wanted to make his mark and shake up the UI or features of the application. It's flawed, but I can rely on it because it doesn't change very often.
- simonh 4y ago
  There's a ton of commercial apps for the Mac and they seem to be doing just fine. Until more recent mainstream adoption, Macs were traditionally most used by professionals, people like photographers, artists, designers, musicians, publishers, etc so there's always been a healthy ecosystem of software for these people. They're still there, although I expect a lot of the growth of the Mac user base has been home users.
GuB-42 4y ago
> Who thought it was a good idea to hand standard development to an ad company in the first place? It is actually not that bad an idea. Previously, software vendors were in charge, it didn't turn out well, you know, the whole Embrace Extend Extinguish thing. Hardware vendors are no better, they all want to keep competition away, especially if they have market dominance, and (lack of) standardization is a way to achieve that. Other standard bodies include bureaucratic entities, like committees and governments. They have the right incentives, but they have a hard time keeping up with the latest developments, the standards become obsolete in practice and people stop following them. Communities are a mixed bag, RFCs are highly successful, but without big money behind it, it often goes nowhere, too easy to ignore. Ad companies have none of these problems. They are all for interoperability, they want to have their ads seen by everyone. They don't have time to wait, so they keep current. And they have enough power and money to make standards standard. The modern web is far from perfect, but it is way better than it was in the IE days.
- account42 4y ago
  > Previously, software vendors were in charge, it didn't turn out well, you know, the whole Embrace Extend Extinguish thing. Firefox had a consistent upward trend until Chrome came along. I think things were working out just fine without the ad overlord having direct control.

JetAlone 4y ago

Over the last few years I've begun to understand that freedom in a too-real sense is the reward we experience for loving the truth, to the extent that we do.

mschuster91 4y ago

> Instead, we're loosing the apps we had, and march towards dumb UIs and needlessly containerized apps a la snap/flatpack. There is a very real need behind app containerization: distributions are exceedlingly slow to pick up updates in a time where web applications routinely run on a "merge request passed, 5 minutes later it's on production" model. On top of that, distributions routinely run into the problem that package A wants a new version of library B, but the other applications haven't been updated to use the new version yet, so everyone is stuck until all applications in the distribution have been updated. Also: Most people use LTS distributions and let's be real no one wants to be stuck with a 2 year old browser that only gets backported security patches. Distributions had the chance to move to a model of "the core is stable but the applications the user uses are flexibly updateable", and they blew it completely.

tannhaeuser 4y ago
The question you should be asking is why do we need all this if the desktop app "market" is contracting and I'm using just the same old apps on Linux I used 15 years ago. To the detriment of alt-OSs like FBSD and OBSD which I always loved. Why would I want frequent browser updates all the time when the web is 30 years old, and also contracting? As I type this (not making this up!) FF prompts me again to restart because updates - why? I'm sick of browsers updating more frequently than the very few sites I still go to; Moz should be producing an essential browser instead of participating in this drivel. It works just fine on my old 2016 XPS with Ubuntu ESR and infrequent updates. You can read my comments to see I've praised Ubuntu, Dell (XPS, specifically), and Lenovo in the past. But I'm on my third Dell (Latitude/Precision so no consumer crap) with fatal hardware defects ootb within 8 months that various customers found necessary to send me for work (to run stunts like Docker-on-Windows or something), while my Thinkpad's miniscule track pad with uneccessary mechanical parts is collecting dust, and mechanically blocking due to it. Right now I think I've got enough of it already and am going back to Apple which I've used on and off since around 2003.
- jen20 4y ago
  A different question might be whether the native app market is contracting because operating system distributions handle updates so poorly.
dTal 4y ago
Setting aside the fact that rolling release distros are 1) a thing and 2) highly popular, I have to believe that these are problems that are within our capacity as engineers to solve without just saying "screw it, we'll give each program its own OS". In fact in some sense they are already solved, from a technical standpoint - you can install Debian, with its crufty out of date repos, and then download Firefox from its official website and run it right in your home directory. There you go, stable core and bleeding edge application. What's the problem?
- AnIdiotOnTheNet 4y ago
  Consider that FlatPak is popular for this case and distributing the way you describe isn't for a good reason: having to deal with all the many and varied dependencies and configurations on any given distro is a giant pain in the ass. People have been screaming this at Linux Desktop evangelists for 20 years about this and mostly fallen on deaf ears, so it is no wonder Linux Desktop is still in a distant 3rd place in the desktop market.

microtonal 4y ago

Linux doesn't inspire anyone to create new end-user apps (with the exception of Blender and Krita, which were started well over ten years ago). Maybe it's because the community is full of people who see any change as an attack on how we used to do things, breaking the UNIX philosophy, etc.? This closed-mindedness leads to thinking like: Instead, we're loosing the apps we had, and march towards dumb UIs and needlessly containerized apps a la snap/flatpack. There is a lot of innovation in this space. OSTree (which is also used by Flatpak), replaces the old Unix filesystem model by something that is exciting and much better (atomic upgrades, rollbacks, signed system images, etc.). Flatpak itself finally makes it much easier to deploy desktop applications, which is long-needed to extend Linux beyond the traditional UNIX crowd.

tannhaeuser 4y ago
> see any change as an attack on how we used to do things, breaking the UNIX philosophy, etc. Yeah let's throw portability and any semblance of order away b/c of ... what exactly? > replaces the old Unix filesystem model by something that is exciting :shudder: Count me in on the closed-minded I guess ;)
- microtonal 4y ago
  Yeah let's throw portability and any semblance of order away b/c of ... what exactly? There are only two Unix-like systems left that matter: macOS and Linux. No need to hold them back by pretending we are still in the UNIX wars and retaining compatibility with some UNIX system that nobody really uses anymore.
  - pjmlp 4y ago
    Linux distribution wars begat UNIX wars.
    - simiones 4y ago
      I think it's the other way around: the older UNIX wars begat the newer Linux distro wars.

everdrive 4y ago

>Linux doesn't inspire anyone to create new end-user apps I'm not really convinced there are more worthwhile apps on Windows / Mac, except for the usual big players. (office, photoshop, etc) I'm sure there are some edge cases here, but for normal users it's not clear this is an issue. Most just use a web browser. They might not even make much use of a file manager. I'm not suggesting that the application situation is perfect on Linux, but it just seems to be bad everywhere. It seems like simple, single-purpose apps are disappearing, and are being replaced by a constantly-changing sea of noise. Is App "X" a good app? Maybe it was a few years ago but the developer tried to monetize it and ruined the app with ads / extraneous features / countless UI redesigns, etc. It's gotten to the point where the static nature of Linux apps is a feature. Rhythmbox is far from a perfect music player, but one of the things it's got over it's major competitors is that it's lightweight and hasn't changed in years. I don't need to constantly re-learn something basic just because an application developer wanted to make his mark and shake up the UI or features of the application. It's flawed, but I can rely on it because it doesn't change very often.

simonh 4y ago
There's a ton of commercial apps for the Mac and they seem to be doing just fine. Until more recent mainstream adoption, Macs were traditionally most used by professionals, people like photographers, artists, designers, musicians, publishers, etc so there's always been a healthy ecosystem of software for these people. They're still there, although I expect a lot of the growth of the Mac user base has been home users.

GuB-42 4y ago

> Who thought it was a good idea to hand standard development to an ad company in the first place? It is actually not that bad an idea. Previously, software vendors were in charge, it didn't turn out well, you know, the whole Embrace Extend Extinguish thing. Hardware vendors are no better, they all want to keep competition away, especially if they have market dominance, and (lack of) standardization is a way to achieve that. Other standard bodies include bureaucratic entities, like committees and governments. They have the right incentives, but they have a hard time keeping up with the latest developments, the standards become obsolete in practice and people stop following them. Communities are a mixed bag, RFCs are highly successful, but without big money behind it, it often goes nowhere, too easy to ignore. Ad companies have none of these problems. They are all for interoperability, they want to have their ads seen by everyone. They don't have time to wait, so they keep current. And they have enough power and money to make standards standard. The modern web is far from perfect, but it is way better than it was in the IE days.

account42 4y ago
> Previously, software vendors were in charge, it didn't turn out well, you know, the whole Embrace Extend Extinguish thing. Firefox had a consistent upward trend until Chrome came along. I think things were working out just fine without the ad overlord having direct control.