Slashdot Mirror
Ask Slashdot: Whatever Happened To the 'Year of Linux on Desktop'?
An anonymous reader writes: Investors, enthusiasts, and Linux distro makers have for more than a decade projected that the upcoming year will be the year of Linux on the desktop platform. But we just can't seem to get to that year for some reason. Windows continues to dominate the consumer market. Apple's macOS X is quickly gaining ground among business customers and designers, and is already ahead of Linux. Do you see Linux getting a significant boost in the desktop market in the coming years?
What happened was, we'd already been using it for years so it sounded really stupid and it was only ever a joke where people laughed at anybody who had repeated the phrase.
It was already a great desktop, and it still is.
New users are not really useful to us, either. Please don't switch.
From 2016, 2015, 2014 . . .
Posted from my iPhone
I don't see Linux gaining a significant part of the desktop market in the foreseeable future. And, as an avid Linux user, I think that's a great thing.
I don't want Linux to get so popular. Getting that popular brings two really terrible things with it: more attention from hackers, and a more rapid degradation of the operating system as it tries harder to cater to everybody.
The desktop was replaced with the smartphone and considering that every android phone runs on a linux kernel its fair to say that Linux rules the world
...will be the year of Linux on the desktop. W8 4 it!
When was YOUR year of the Linux desktop?
Mine was 1998. I installed Redhat 5.2 and Linux has been on my desktop ever since.
My beliefs do not require that you agree with them.
Those are sort of on the desktop. Granted given that the general trend is people are using mobile devices more often than not, and your choices are a Linux kernel or Mach, we've already been there a while.
My 83-years old father uses Ubuntu. Granted, I was the one who installed it and he only uses Firefox.
I have used Linux as my primary desktop since ~1997. As a software developer it is a power platform. The shell is critical. However, as a conventional desktop it is just not competitive with Windows. And OSX isn't either. Both Linux and OSX are below 4% market share. Vertical integration is very weak. Windows has an identity management system that allows transparent filesharing, advanced group based access control, sophisticated business applications. Getting stuff like that to work on Linux is too difficult or simply not possible. So software venders focus on the Windows platform. And rightly so. I just tried and application that recently released a Beta for Linux and it was a total fail. I occasionally dabble in engineering related stuff and I have to have a Windows machine for all of the various programs for cad, PCB design, simulation. Yeah, programs like that exist for Linux but they're just not good. And I know people agree with me that the GNOME desktop has actually regressed. It used to be much more usable. But they dumbed it down for reasons that where not entirely clear. My guess would be that when new developers come along, they have a tendency to want to re-write everything from scratch. I'm not diametrically opposed to this strategy but you better come up with something that was at least as good as what you're dumping. And that didn't happen. There are other integration related issues as well. For example, for as long as I can recall there has always been a fight between X and the desktop over who should remember the positions of windows. X says applications should save that information and recall it when re-launching an app. Desktop people think it should be handled by lower level facilities. Now, whenever logout and back in, all of my terminal windows have to be re-launced and repositioned (I run 6-8 terms on 4-5 workspaces). That is something that actually used to work somewhat in GNOME. It worked in WindowMaker IIRC. The Linux desktop has been dumbed way down to the point where it's not nearly as useful as it used to be. At least not for people doing more than surfing the web and email. Might as well just get a Chomebook for that.
It's actually easier. I've put my in-laws on Linux (Mate as a desktop environment) and it's much easier for them. Windows 10 is terribly confusing (even for me). Mate has a 'start' menu not too different from Windows XP, using Linux Mint means that it practically updates itself. And they love the fact that they are much less vulnerable for malware.
Not to mention the retarded 'Windows is updating' message lasting forever even on a I7 with SSD and lots of memory when shutting down AND starting up. At most Linux wants a normal restart after it updated itself after it quietly updated in the background.
Elderly people that don't do much more than use it for online stuff are better off with Linux.
The only thing keeping a lot of people and especially companies on Windows is software that only runs on Windows. With more and more software being web based that is becoming less of a problem. Only the large volume of MS office documents will be a big hurdle for a long time IMO
---
Here is a blast from the past for you:
https://linux.slashdot.org/sto...
Do you use Android? Do you spend more time with your phone or tablet than on a workstation or laptop? Congratulations, you're in the year of the Linux desktop.
Meanwhile I've got an XBox One S for movies and some games, several generations of other game consoles, a couple of Raspberry Pis running Raspbian but often used to emulate older console and desktop systems, a WebOS smart TV, a Linux smart TV, a couple of Chromecasts, a Windows desktop for games, a Linux desktop for personal non-game use, a Linux laptop for travel, a Mac desktop for company work that mostly connects to Linux systems and runs Linux VMs, a Mac laptop for company work that mostly connects to Linux machines or to my work desktop, two Android phones one each for work and personal use, and a non-Fire Kindle for reading without interruptions like I get on my other devices. My girlfriend has a Mac laptop, a Linux desktop, and an Android phone.
So... what's the question again?
Android has become even more insufferable than Windows, with annoying useless spyware/adware programs pre-loaded that most users have no way at all to remove, due to device makers not providing hardware drivers and software access permission.
Also (partly due to that?) Android has the charming feature of updating until the meager memory fills up, and then you,,,, ummm,,, what? Then you go buy a new one. Because the vast majority of people with a device in this state have no idea what to do with the thing after it keeps warning that it can't get updates. They assume that they shouldn't keep using it, but they don't know what to do when there's twelve programs named Google, Google+, Google.com, Google.service, Google.accounts, Google.user and so on. -And you can't delete ANY of them anyway, even if you did know what to get rid of.
Android is an OS that device makers wanted; its main feature is that it can be locked down against modification--and Google catered heavily to that desire. It's not the one that users wanted.
Android is a horrible example to use of Linux being popular. In fact, it shows the complete opposite: Linux can only become a widely used consumer OS kernel when users and developers have absolutely no idea it's there, and it's thoroughly hidden under many layers of abstraction.
Google could silently replace the Linux kernel with some other kernel, and Android users and developers would have no idea it had even happened. That just goes to show how irrelevant Linux is within the Android ecosystem. Yeah, it's present, but nobody cares that it's present.
We may actually see a kernel replacement along those lines happen, with Google Fuchsia being in the works.
Linux contributes almost nothing to Android's success. Android could have been just as much of a success if they had used the NetBSD kernel or some other kernel instead. The success of Android is in its application framework and its userland apps, which have nothing to do with Linux at all.
Android shows exactly what needs to happen if Linux does want to be successful on desktops and laptops. Almost all of the GNU utilities, X, Wayland, GNOME, GTK+, systemd, PulseAudio, and other open source software will need to be thrown out and replaced with a far more cohesive and sane userland stack.
The only way that Linux has advanced in usability is by copying Windows and Mac features.
> more rapid degradation of the operating system as it tries harder to cater to everybody.
That would have been a reasonable prediction 30 years ago. For the last couple decades, almost all supercomputers have used Linux, as have many embedded systems, most web servers, and now most phones / mobile devices use Linux, each with an appropriate UI on top. The fact is, Linux does suit a vast array of very different use cases, and that has worked out very well.
One reason that has worked well is new use cases, such as mobile and cloud. When there was suddenly a need for an operating system well-suited to run the hardware cloud hosts, Amazon and others choose the OS that had already been proven to be quite flexible, and made it even more flexible as they extended it's usefulness in that role. When the Android team needed as OS (not GUI shell) well-suited for advanced mobile devices, they chose Linux because it had been proven to be flexible. They made it even more flexible. So it's a cycle. The more different uses Linux is put to, the more flexible and modular it becomes, making it well suited to applications that don't even exist yet.
There is a lot of truth to that. The article mentions business desktop and consumer desktop.
Microsoft is still very popular on business desktops, of course. Windows on the desktop is NOT popular with consumers. Consumers have largely left the Windows desktop, moving to Android. Even if you leave out iPhone, people bought more Android devices last year than the total sales of Windows devices by both business and consumers combined. For consumers, Android and the mobile form factor are three to four times more popular than the Windows desktop.
Of course, with the UI on Android, some of the storage and processing is being done on the server - by Linux. What consumers use is a Linux-based product which communicates to other Linux-based systems.
While saying "Windows is most popular on the desktop" is technically true, it's a lot like saying "David Duke is popular with the KKK". True, but that doesn't mean that either Windows or David Duke are well-liked.
Eh?
Consumer have largely left the desktop, period.
Lots of people I know who used to slog around laptops or have a desktop at home now rely entirely on their pocket computer (whatever OS it is running) for everything.
As I see it, it's desktop-vs-pocket, not Windows vs Android: Android is not [intended to be] a desktop operating system.
Kid-proof tablet..
If your "non technical mother" was running a recent copy of windows and complaining about it crashing all the time. Make me wonder what porn sites your dear old mother was visiting.
No, strike that. It doesn't make me wonder. Images of senior donkey porn .com are now filling my head.
I read at +2. If your post doesn't reach that level I will not see or respond to it.
Go read Qt's commercial terms,
Maybe YOU should read them. You need a commercial license if you want to produce closed-source proprietary products. You can still sell your product / offer support, etc., without a commercial license, you just have to provide source.
And Qt is not the only game in town.
The year of the Linux desktop has past mostly unnoticed. And the OS is nothing more than a terminal for services. Just got a Chromebook for 130€ to try out this cloud thing. (I'm a 20 year Linux user and my other portable is a MB Air from 2011). The Chromebook concept is amazing. Dirt cheap, boots in seconds, runs for hours on a single charge with a very small battery (ARM system) and is totally idiot safe, usable but the other 99.999% of the population who aren't computer experts like us. Two-factor auth setup with two mouseclicks.
Given, I have to do *everything* with cloud services now (IDE, CI, Testing, Documents, Storage, etc.) and everything is hooked to accounts in the cloud. But as you know, that's not just disadvantage but also comes with huge advantages. Having Travis and Codeanywhere do the setup work for me lets me focus on coding. If the Chromebook gets stolen, I'll disable it remotely and pick up where I left somewhere else. I don't have to think twice about syncing my Smartphone with the stuff I did on the cBook.
Note that this stuff can be used by some kid in the third world aswell. Which is exactly how Google intended it to be.
You have to hand it to Google, when it comes to enablement, they are lightyears ahead of everybody else, including Apple.
Bottom line: The Linux Desktop is long since here and it will take the world in a storm. It's called Chrome OS.
We suffer more in our imagination than in reality. - Seneca
Linux has no exposure in the consumer market. Most people with Android phones don't even realize they have Linux in their pocket.
Consequently, Linux isn't available from PC vendors. They don't think there's a market for it, there's no OS vendor willing/able to make it worth their while, Microsoft aggressively forces OEMs to choose between Windows and anything else, and OEMs know anyone looking for such a machine won't tolerate the bloatware they love to include (which doesn't exist anyway).
Then there's hardware support issues, mainly Video. The Linux desktop needs a breach point into the consumer market, the most likely candidate is a Linux gaming console (looking at you, Steam).
Are you kidding me? On-screen keyboards flip out if you're a fast typist. I can have my physical keyboard sounding like it's a musical instrument with minimal trouble, but I try typing those kinds of speeds on an on-screen keyboard and suddenly letters drop out and autocowreckt joins in by guessing (very wrongly) what letters I 'obviously meant' to have hit. It gets worse if any amount of technical terms turn up.
You need either a monolithic architecture or a performance hit to do certain common user-based operations (what if someone plugs in a USB stick? what if they unplug it while it's being read and something else is waiting? what if they plug in multiple mice? what if they unplug monitors and rearrange them? what if they due to some horrible form of OCD need to unplug their mouse and plug it back into the same port 10 times before starting gimp? what if they kill the machine with a hard reboot after a partial soft reboot when transferring files or with hard drive locks? etc.) There are lots of user-initiated operations which simply don't apply to a server because servers are made to sit without direct interaction most of the time, save for the times when someone who knows what they are doing is using them.
The notion of lots of little things doing 1 job extraordinarily well works great for servers, it doesn't work great for a user-centered experience without a common pipeline of some form (which implies overhead of standardization because sometimes certain operations should be prioiritized massively over others, such as anything UI-related) which itself implies overhead in terms of performance if it isn't monolithic.
Meanwhile monolithic architectures are bad within open source, almost universally, because you end up with things like systemd: a handful of devs with conflicting vision (or at least conflicting with everyone else) who are the only ones who know it well enough to really touch it without fucking everything up. Corporate software doesn't suffer that particular issue because the vision is inherently derived from a hierarchical structure and they have the resources to support something so massive in spite of the developers not actually wanting to build that specific thing.
Philosophies are just that, when you start conflating them with practical works you run into issues, as the Unix philosophy does when extended to a desktop environment.
Qt is licensed under the LGPL.
If you dynamically link to the Qt libraries, you can sell your closed-source proprietary products without having to pay for a commercial license or share your source.
If you statically link to the Qt libraries, then you are required to either pay for a commercial license or share your source.
Go read Qt's commercial terms,
Maybe YOU should read them. You need a commercial license if you want to produce closed-source proprietary products. You can still sell your product / offer support, etc., without a commercial license, you just have to provide source.
And Qt is not the only game in town.
Which is not true. I worked for a small company selling close source Qt applications. We started out using the LGPL version of Qt 4.3 and once we made enough revenue we switched to the commercial license so that we could use some of their close source libraries. We used v4.3 through v5.5 and the quality and support was excellent. Going from 4.x to 5.x was painless so claims that the product is poorly supported are bunk.