What can push Linux to the next level?

I should preface this entire post with the following disclaimer:

This has to do strictly with Linux on the desktop and is nothing more than my personal opinions on the matter. I have no knowledge (of anything, really).

Good. With that out of the way, let’s jump right into it!

GNU/Linux has a very passionate following as a desktop operating system. There are a myriad of reasons why a person might feel this way, and that’s quite cool. Whether it be the developer wanting to have control of as much of the system as possible, the open source advocate choosing FOSS out of a sense of duty, or maybe a computer tweaker who just wants to be able to mess around with their OS.

Oh, it also might be your grandparents who were handed a machine with Linux installed so that their grandson might not have to continuously remove malware from their Windows box. You know, just maybe.

Needless to say, there are reasons.

However, desktop Linux has also not really caught on in the “mainstream” market. I don’t think anyone is really saying otherwise, but what would it take to push Linux to the next level and encourage mainstream uptake?

You need to start with the one thing you can control: software.

Right now things seem to be shaping up in a three-way battle for the desktop environment: Gnome 3 vs. KDE 4 vs. Unity. Now, that’s not entirely true because Unity is really (I think) a shell on top of Gnome, but I would say these are the BIG THREE at the moment mainly because Ubuntu has tossed their towel in with their Unity environment.

What needs to be done?

We need a time, now, after the huge changes that Gnome 3, KDE 4, and Unity have brought to just fix bugs, work out UI inconsistencies and do a ton of user testing to figure out what pitfalls there are for all three. I believe there is room for all three desktop environments, but they all need some cleanup and focus on the “polish” … which, by the way, is a terrible term for something like this, but I’ll go with it anyway.

I’ve thrown my hat in with Unity because I like a lot of what they have done and it works pretty well where I use it (mainly on widescreen monitors and laptops), but there are things that need to just be worked out. How windows react with the Unity toolbar can sometimes be … interesting. Unifying the scrollbars will be greatly appreciated and there has to be a better way for the menu bar, even if you need to rip off Apple’s ideas. Those are all design decisions, and things will continue to change in the future.

The default state of each environment needs to be attractive, but not gaudy. I think Mac OS X has proven that incremental changes to an OS can keep things fresh without needing to toss everything aside (I’m looking at you Windows Vista/7).

This is where we started.
Snow Leopard image
After 10 years, this is what we now have.

Very familiar and yet different at the same time. So much as been polished, changed, changed again, moved, added to, removed … you get the picture. With the changes brought about in Gnome 3, KDE 4, and Unity I think the base has been set for a long, incremental polishing of the user-facing interface to become something really really great.

I’ll go on record saying that I like what I see with Unity, but that I think all three are on the right path.

However, the software is just one part of it (or, maybe more as we will see). What is maybe even more frustrating is the seeming lack of OEM support for Linux.

It has been getting better on the hardware front, but there needs to be a huge push to get open source drivers (or at the least, comprehensive documentation) out there so that hardware support can get better.

I, for one, love the ThinkPad line of notebooks. I would like to see Lenovo make it a policy to only use hardware that is supported by Linux because I think that ThinkPads make excellent development machines. I understand that is probably not possible, but if a major player like Lenovo would make that push, maybe we would see component manufacturers get off of their duff and start really supporting open source operating systems.

I hope for the day when I can go to Lenovo’s site, choose any laptop I want based on my needs and be confident that I can download the latest Ubuntu/Fedora/etc., install it, and get to work.

Offering an install of an open source operating system would be great as well, even if Lenovo makes their own distribution to put on their machines. HP bought Palm and now there are rumors of webOS because put onto their traditionally Microsoft-only hardware platforms. If that starts to become a differentiator for OEMs, then maybe this will become a reality.

So, those are just some generic thoughts about what might need to happen for Linux to get pushed to the next level. As for me, I’ll continue to push it where it is viable. I already have my brother happy with his desktop because he can do anything he needs without the need to worry about malware at the moment (not 100% true, but more so than before for him).

I also think the education market is primed for a little disruption, but that is for another time.

8 replies on “What can push Linux to the next level?”

The rise of Linux in that way will require someone to become the big player in the Linux area – were it not for the fact Apple want to provide the full stack to the end user they would have the ability to shape Linux in to a ‘Windows beating’ system – its not so much the OS itself that needs to be tweaked its more the software stack on top – everything needs to be brought up to a certainly level.

Microsoft has done so well because they have always tried to release software software for their platform that is professional and user friendly, other than Apple there are no Linux providers who are providing software like MS do.

You’re right, of course, except that Apple doesn’t use Linux at the kernel level, they use (I believe) a derivative of BSD. Besides that tiny oversight you’ve hit the nail, which is why I stress polish.

The underlying architecture might be there for many things on Linux, but the user-facing stuff is just not there and needs to be for anything to take off. Maybe Canonical can become that, but along with having a heavy hand to guide things, the hardware support needs to be there from DAY ONE for major hardware updates.

Telling people they can just buy older hardware might work for some, but you need to also have the newest stuff available from DAY ONE in order to drive adoption. Apple gets around that, to an extent, by writing their own drivers and only dealing with a tiny subset of hardware. That makes their life and their customers’ lives easier.

As a guy in the MS World any difference between a BSD and a Linux system doesn’t matter at all – they’re both grouped together, the same way from a Linux guys POV the differences between Windows Server 2008R2 and Server 2003 don’t matter either – they’re both ‘just Windows’.

On the hardware you have a point running Linux on the server side would be difficult for me, we tend to buy the biggest baddest newest hardware for our servers – having to wait 18 months for a distro to support our hardware would be a show stopper. Being able to buy the latest generation of Dell server and deploy our software stack to it without fear of it not working is central to our Technology Services Department being able to respond to business needs.

Correct, since they actually are “just Windows” in the same way Fedora and Ubuntu are “just Linux”. I think that might be the better parallel in this case.

Funny thing about hardware support is that Microsoft is able to ride their prior monopoly to have that support. The hardware suppliers NEED to supply drivers for Windows on their own and Microsoft doesn’t really have to do much in that arena. On the other hand, Linux/BSD/etc. devs need to spend time writing drivers themselves because most hardware suppliers are just not that interested.

I think that’s a shame, but I’m slightly biased.

I think you’re just lazy ;). Dell offers Red Hat as a OS choice with their servers and lots of companies bigger than you run Linux on bleeding edge hardware.
(Runs and hides so Dan can’t hurt him.)

Also, developing PHP/MySQL on Windows servers sucks.

Only Bob can make a post about Linux an anti-Windows, OS X love fest.

Comments are closed.