Fragmentation is Why Linux Hasn’t Succeeded on Desktop: Linus Torvalds

[PondLife]

Marketing is the problem rather than the number of desktop environments, if a single format could be chosen to be the official version of Linux with the right backing that could make a difference. Maybe that will be an evolution of a chromebook as suggested in the interview. Other distros would benefit in the long run as more people were introduced to the open source world natural curiosity would get some to try the alternatives.

But I don't regard the current situation as a failure a 2 to 3% market share is still a lot of computers justifying its existence and I for one are quite happy with the current status quo, market share is not everything and I doubt the MX developers put in all the hard work they do in the hope of world domination (correct me if I am wrong!).

[Adrian]

There are simpler explanations for the current situation:
1. Most computers come with an OS. Few people even know what an OS is, even less think about replacing it.
2. People want to run programs not Operating Systems. Basically if somebody wants to run Word or Excel or Adobe Creative Cloud or whatever, they will expect the OS to handle those without fussing with things like Wine that might or might not work.

Confusion about which Linux and what Desktop environment to use come very last in the list of things that stop people from switching to Linux.

Also, I'm not so sure that having millions of desktop installation in this context can be seen as "hasn't succeeded". The market was already established, think about a different phone OS trying to replace iOS and Android... not really easy to do. Heard of Windows mobile? Did it fail because of fragmentation?

[GuiGuy]

I hope that Linus will discover MX one day.
Of course he would not want to use it himself, but his family would probably like it.

Any distro that tries to look like android or chromebook deserves to fail.

[rasat]

I agree to some extent with Linus Torvalds by comparing my change of WMs and Desktops since 1996. But I don't think the fragmentation was a "bad" thing, rather gave scope for innovation and creativity. Without this, distros today may have have run in IceWM. :)
Maybe true, Linux is now ready to move to a more uniformed usage of desktop.

[wulf]

There are simpler explanations for the current situation:
1. Most computers come with an OS. Few people even know what an OS is, even less think about replacing it.
2. People want to run programs not Operating Systems. Basically if somebody wants to run Word or Excel or Adobe Creative Cloud or whatever, they will expect the OS to handle those without fussing with things like Wine that might or might not work.

Confusion about which Linux and what Desktop environment to use come very last in the list of things that stop people from switching to Linux.

Also, I'm not so sure that having millions of desktop installation in this context can be seen as "hasn't succeeded". The market was already established, think about a different phone OS trying to replace iOS and Android... not really easy to do. Heard of Windows mobile? Did it fail because of fragmentation?

+1
Additionally, given that almost universally, the first contact children have with computing at school is likely to be a Windows OS of one form or another, I doubt that the majority even discover Linux until their teen years. Regarding the phone OS, one organisation is currently giving it a go. >
https://e.foundation/e-solutions/

[Richard]

Linux is wonderfully successful for me. MX is on top of the best for my needs.

Began learning with MuLinux about 1997. Mu fit on a few 3.5 floppies and was a good learning tool.
Bought SuSE Linux v7.2 mainly for the manuals. It worked fine but the WinModem was a drag.
Failed installing RedHat and Debian but Mandrake 7.2 finally let me get online with a serial modem.
From dual booting Mandrake in mid 2000 to Linux first in early 2001 it has been adequate for my needs.

Then Mepis 3 until the Ubuntu experiment which bothered my Debian sensibilities, I suppose.
Did a lot of searching, hopping and used a lot of different distros. Seemed to keep finding the one
that almost did everything I wanted but there was always something missing. Learned a lot over the years.

And then I found MX Linux in March of 2014. Like a dream come true!

Yes, it wasn't much to look at but it had some great utilities and a Snapshot tool! Had dreamed
of being able to create ISOs of my setup so I wouldn't have to fiddle so much with each new install.
Really need to script it some day instead of just make install notes.

I believe that Linux is great for those who are willing to learn it's ways. The LSB (Linux Standard Base)
never seemed to get off the ground. There are so many ways to build it that a unified LSB seems distant.

[handy]

Without the Bazaar, the Linux OS would not exist.

The Cathedral is the way that MS, Apple, & the BSD's (among others) do their thing.

Linux is what it is, & has all that it has due to the multifaceted chaos of the Bazaar method of development/support.

https://en.wikipedia.org/wiki/The_Cathe ... the_Bazaar

Linus knows this as well as anyone does.

[uncle mark]

People want to run programs not Operating Systems. Basically if somebody wants to run Word or Excel or Adobe Creative Cloud or whatever, they will expect the OS to handle those without fussing with things like Wine that might or might not work.

This. PhotoShop, QuickBooks, MSFT Office, and games are the killers. People want to run these programs, period. No substitutions.

Those who aren't married to those programs are perfectly fine with desktop Linux.

[j2mcgreg]

Actually, the linux killer I run into most often is income tax software.

[richb]

Actually, the linux killer I run into most often is income tax software.

I used to run TurboTax in a Windows VBox virtual machine in Linux. Now I just use the online version.