Peripheral compatibility, software quality and community issues on the Linux desktop.
It seems like it wasn't that long ago when Windows was an exclusive part of my computing life. Ever so slowly, I began to move away from Windows XP into some of the popular Linux distributions of the time.
I found myself falling in love with a specific Linux distribution made popular by its ability to "just work" without a ton of configuration. At the time, this held a great appeal to me. After all, I had other things to do throughout my day besides having to configure everything on my desktop PC by hand.
Flash forward to today, my schedule is basically the same. Therefore, I tend to lean toward desktop-friendly distributions that help me get my work done with as little hassle as possible. And much to my satisfaction, today's modern Linux experience is beneficial to me in this space.
There are plenty of great Linux distributions to choose from that give me a first-rate, simple to “make work” solutions without spending an entire weekend setting things up.
In this article, I’ll focus on some of the hits and misses I've seen with the Linux desktop over the years. Considering my experiences from when I started using the Linux desktop up through today, I've managed to gain significant insights on what's working and which minor issues need some polish.
Linux peripheral detection
For every complaint I've ever heard about hardware compatibility with Linux, I can counter with at least 50 examples of hardware detection that would blow the doors off of what a Windows user experiences.
Just to give you an example: I recently did some testing with a HP printer, drawer full of wireless dongles, a Bluetooth dongle, among a few other things. The HP all-in-one printer was detected immediately and setup with zero intervention from me. Same with the Bluetooth dongle and most of my wireless dongles, as well. Even the wireless dongles that were not ready to go out of the box were easily made usable with Ubuntu's restricted driver tool.
Then I took the same devices and tried the same thing on an installation of 32-bit Windows 7. Everything I plugged into it failed. Three of the wireless dongles even failed after installing the proper drivers.
It's tempting to blame the makers of the operating system for this problem. Oddly, though, the truth is that it's the manufacturers of the peripherals that are at fault here. It's not Microsoft's responsibility to offer the drivers for older hardware.
But as you might have guessed, manufacturers have no incentive to offer drivers for older products. Why? Simply because there's no money in it. This is neither wrong or right, it simply "is" the reality of the situation.
Bounce back over to the Linux desktop, you'll find that the community came together to do everything possible to offer peripheral support going back as far as needed. It's a cooperative effort between developers from all walks of life that make this little miracle possible.
Comparing this to the spotty support of just a few years ago, it's tempting to suggest that Linux wins on the peripherals front. The truth is, however, that Linux wins in "community and cooperation." The reason I make that distinction is that a manufacturer can always release a new wireless chipset tomorrow, using something that is unknown to the community and only Windows compatible.
This translates into a mixed bowl of failure and success for Linux on the desktop. Despite the valiant effort of countless developers, wireless support for Linux is hung up on a single, avoidable issue that isn't likely to ever be remedied. At least the community did their part, I guess.
Is community enough?
There's no question that one of the strongest assets that desktop Linux has to offer new users is the community that supports it. Sadly, this asset isn't enough to address the problem of companies that release hardware without providing all the needed details for full Linux compatibility.
The community does an amazing job at making sure compatibility is top-notch for desktop Linux enthusiasts. But the fact is the latest wireless devices can present problems despite these well-intentioned efforts.
Even considering the big names in wireless chipsets like Broadcom that embrace Linux, it's still a tough sell with newer Linux users when their wireless connection doesn't work. Worse is when there are drivers released for known-to-work wireless options, and yet users are still forced to install a tweaked version of the same driver.
Considering it works, why isn't this addressed out of the box with distributions targeting newer users? I could speculate further here, but suffice to say corporation-backed Linux distributions like Ubuntu still ignore my suggestions.
By providing distribution branded 802.11n wireless dongles an aging problem would be solved overnight. Just select the chipset that you prefer, then support it properly. It's not rocket science, it simply requires some clout.
Both Atheros and Ralink offer solid solutions here that might be worthwhile places to start. I'd bet you that new users would buy these dongles if they knew their chipsets took center stage with each new Linux distribution release!
Unfortunately, this isn't likely to happen anytime soon. Maybe some day I can get it through to distribution maintainers that this is both a fix and an income source. However, until this happens, we'll continue to go in circles.
Considering the issues with wireless above, I'm forced to give desktop Linux mixed marks for wireless support. Great wireless stack, decent community involvement, yet we're staring at a big fat failure from distribution maintainers point of view. If they'd stop trying to support "everything under the sun," life would be easier for everyone.
Linux software examined
Many Windows users feel that software for the Linux platform isn't cutting it. They claim that they would miss their proprietary solutions from Microsoft, Adobe and Apple.
Despite this outlook, I've found most Linux software exceeds my own needs daily. LibreOffice works just fine to fulfill my office suite requirements, as does using software for just about anything one might be doing with Windows. I use Linux software to do audio/video editing work and photo manipulation, along with other related activities. Additionally, the available applications for the Linux desktop have been very easy for me to work with. I also have a strong appreciation that I'm not being locked into silly licensing headaches.
For anyone who believes that Linux software is lacking features, I would point to applications like Ekiga or Gnome Do. Both are examples in simplicity and software that works well.
I'd also toss in newcomers like Kazam and Pandora for Rhythmbox to show us how something we take for granted, can become a whole lot better.
Over the years, I've watched as software for the Linux platform has both "caught up" and in some areas, outclassed what I've found for Windows and OS X. Obviously this isn't the case with every software category, but developers for the Linux desktop have a lot to be proud of in my opinion. If you compare the software of a mere five years ago with what we have today you'll notice some pretty impressive innovation, everything considered.
The good, the bad and the ugly
I don't think there is any question that the Linux desktop is ready for mainstream use. The key is making sure that it’s providing the option for a mainstream experience on distributions for newer users. In order for growth to continue, I think we need to address the misses as well as celebrating our hits.
What I mean by this is we have a platform that offers needed peripheral support, software solutions and a safe computing environment. Unfortunately, when you're working with Linux that isn't pre-installed, the results can vary from users to user. This means that everything works fine most of the time, like with proprietary operating systems. But unlike proprietary operating systems, issues such as wireless support tend to be a huge buzz kill for potential platform adopters.
My wish is for companies that embrace the Linux desktop to consider the advantages of not waiting for OEMs to pick up the slack for them. Instead, how about we take what's working already, while empowering DIY'ers to use branded USB wifi dongles instead of playing musical chipsets hoping to strike gold?
Sure, there's still going to be the occasional video card or sound card that provides some challenges, but overall I think we'd see adoption explode if wireless was taken more seriously by the companies supporting today's distributions. The developers did their job by providing working wireless options.
Now how about we offer USB pluggable solutions, so newbies can experience some of the positives that Linux on the desktop has to offer? While the platform may be hit and miss, generally speaking, I'd suggest that Linux is a hit out of the park overall.