Even after considering the success seen with Linux on netbooks, there is really no question that it feels like something ominous is holding back desktop Linux from the masses.
In the past these hurdles have been attributed to familiarity, legacy Windows software and even a lack of a perceived collaborative effort from a business standpoint. Based on my own experience however, there may actually be issues that come up for new users before even hitting any of these hurdles.
In this article, Ill exploring some of the challenges that I feel are indisputable headaches, while also offering possible solutions.
Atheros, Intel and Ralink: each company provides decent Linux wireless for the most part, despite running into occasional bugs. Despite the level of support shown from each, it still frustrates me that wireless options like Broadcom wireless remains the gold standard in most notebook computers sold today.
Even with the hard work that has gone into providing support for these problematic wireless chipsets thanks to projects like b43, most people still find that they are relying on NDISWrapper and windows drivers anyway.
It's nearly 2010 and we are still fighting Broadcom wireless hassles just search Google for an abundance of examples. The work behind b43 is great, but I think its not really addressing the larger issue the lack of natively working Linux ready wi-fi devices offered pre-installed on most notebooks.
What makes matters worse is the amazing inconsistency I have seen with natively supported devices using chipsets for 802.11n. One USB Ralink wireless device I own, for example, works out of the box with Puppy Linux yet falls on its face in the latest release of Ubuntu. Linux should be a more uniform experience, yet sadly this is simply not always the case.
In my view, the solution to these problems is a simple one: Support wireless chipsets that support us as Linux users.
I don't really care so much about how firmware is licensed, I simply want my wireless to work. Then get something consistent going on with the version of the working drivers from one distro to another. Assuming the same kernel version is being used, perhaps a stronger collaborative effort is needed here?
Video cards in Linux
Compared to even just a few years ago, video card driver support for most Linux distributions has come a very long way. So far, in fact, that the installation of NVIDIA and ATI proprietary drivers is a brainless affair. Job well done there.
What makes this even more exciting is that I am still free to install these same drivers the old-fashioned way! So my freedom of choice on how I deal with this has not been inhibited at all.
Where things fall apart, however, is that not everyone is using Intel, ATI or NVIDIA for graphics. Some lower-end PCs and notebooks still use VIA graphics. And this sadly translates into using OpenChrome drivers.
In some cases, using OpenChrome works fine. But it's no secret that OpenChrome is about effective as the valiant efforts seen with B43 in the wireless realm. Users are still relying on the hard work of others to make a poorly supported chipset work "most" of the time. This doesn't make sense.
But since VIA wins on market share, this means many of us are stuck with that lousy chipset when purchasing low cost notebooks. With cost being a consideration here, I'm not sure how this can be overcome outside of saving a few more dollars and then purchasing the better notebook.
Restricted Codec Handling
Any Linux user who has been on a Linux desktop for at least a month likely understands why restricted codecs are not generally provided by default. It can fall into conflict with the US patent laws.
Windows and OS X provide freedom from this hassle out of the box. The arrangements have already been made behind the scenes, legally, to ensure these users have access to most restricted codecs with no extra effort required.
With desktop Linux distros on the other hand, everyone is on their own to figuring this out. Some distributions include restricted codecs by default while others take Fedora's approach in making them available...if you pay the license fees required before installing them.
Then we have Ubuntu. With this distro, we simply see them available for free download and not installed by default. So even though Canonical sells restricted codecs from their store legally, good luck knowing anything about it from the Ubuntu website I am not finding any clearly labeled links to this information.
Like so many others, I celebrate many aspects of diversity in the Linux realm. But codecs is one area that leaves me quite frustrated. Not due to perceived difficulty, mind you, rather the fact that many end users are left playing armchair attorneys in hopes of "doing the right thing" in the eyes of U.S. law.
VoIP is not always universal across platforms
Today's desktop Linux distros have no shortage of great VoIP clients to choose from. The most polished of them using SIP technology would have to be Ekiga.
Sadly, using SIP-compatible clients on OS X and Windows is hardly universal...especially if you want webcam support.
This leaves us with proprietary Skype as the only real viable cross platform solution meeting both audio and video needs for most people. And until recently, Skype on Linux was awful to use due to it not playing well with PulseAudio.
Thankfully the latest beta has resolved this. Even better, the Skype GUI is going to go open source in the near future. At least this is one area of Linux challenges where we are seeing some real progress taking place.
Just provide us with some consistency
Desktop Linux, at its core, is totally ready for the mainstream world. Properly managed, today's distros are in a stronger position than any of us could have possibly managed even just a few years ago. The combination of software choices and hardware detection is just jaw-dropping considering how much voluntarism goes into making Linux what it is today.
Yet despite this great news, there remains a dark side to be overcome. This smudge on the desktop Linux record is not something out of our hands. No, it's a simple matter of finding a stronger balance with consistency while remaining diverse for the end users.
My call to action is for the community to work with the strengths that already exist with the various distros out there. I'd love to see users step up and provide more up-to-date documentation for the distros they use each day.
I want to witness Canonical making a real effort to realize that here in the U.S., we would like visible access to restricted codecs from their store without having to dig for it. And of course, for Ubuntu users looking to purchase new notebooks to stop purchasing from big name vendors that offer only Broadcom chipsets and instead, spend a little more with smaller vendors that support Intel wireless options.
That's it, nothing Earth shattering. Just taking baby steps toward showing the world that we as Linux enthusiasts are not all cheapskate tightwads that will bend over backwards to avoid spending a dime to support extended functionality.
Want to listen to MP3s on your Linux box, then vote for it with your wallet. Fair as it might not seem, if you want the world to start taking us as desktop Linux users more seriously, then we need to lose the cheapskate image that has been holding us back. Even if it is cheaper because Microsoft and Apple pay for some of this for their users, we as Linux enthusiasts must take a stand.