"What's this 'DEEE-bee-en' you write about?" Or will Linux ever (ever?) make its move on the desktop?
The entries in this blog flow through my Twitter and Facebook feeds, and once in awhile a friend of mine who has nothing to do with the open-source software world wonders what the hell I'm talking about.
Just this weekend, somebody asked me, "What's this 'DEEE-bee-en' you write about?" Would the question be any different with the word "Ubuntu" in there?
I don't think so. When it comes to desktop Linux (not to mention BSD, which is even further in the geek ghetto), nobody knows what we're talking about.
I'm as guilty as the next "journalist/blogger/fanboy" of preaching to the pulpit and writing about Linux and BSD geekery for Linux and BSD geeks.
I don't see a whole lot of writing out there, even about Ubuntu (which in my view is the highest-profile Linux distribution by far), that aims at the unconverted, the Windows or Mac user who doesn't even know there's a third (through 300th) way.
I tend to write about what I'm doing with free, open-source software and my particular stable of hardware. Writing this blog isn't exactly in my job description. I do it because I'm compelled to do it.
Like the great majority of writers I've encountered over the years through sites like LXer, I use this stuff every day to get actual work done. As some put it, I eat my own dog food.
And even though every year since 2000 has been declared "The year of the Linux desktop," it's just not happening.
Despite the bashing by many in the geek world, Ubuntu's parent company Canonical is doing more than just about any company in promoting Linux on the desktop.
We all may argue about how Canonical makes decisions (and tends to leverage "the community" when it's to the company's advantage, not so much at other times), why one default application or another is chosen, which proprietary bits are included in the distribution and how they're presented to the user, and on and on about which side of the window contains the buttons and what colors are used to "brand" the system.
Canonical/Ubuntu tries to keep the vast fanboy army happy while doing what it can to drag new users into the fold. I wonder if both goals are attainable with the same distribution.
As a "veteran" user of Linux and BSD, like many of you I've done dozens of installs, tried all of the major and quite a few of the "minor" distributions. I have opinions on which applications are best for a given use, how "fast" or "slow" one distribution (and its configuration) are compared to another. I care about software freedom but also need to get things done that are sometimes incompatible with a rigid hewing to said freedom.
Also like many of you, I also use Windows and Mac OS. I've even done a few Windows and Mac OS installations over the past few years. I've dropped Windows 7 on my laptop, then manually downloaded and installed a dozen drivers to make everything from the display to the sound chip and webcam work.
I didn't have to do that with Ubuntu, Debian or Fedora because Linux tends to handle hardware "better" in some ways, with a lot of drivers included in the kernel. Not that the manufacturers of my various hardware bits helped as much as they should (or at all in many cases).
Right now I'm running Debian Squeeze because my Lenovo laptop made in 2010 is pretty much unusable with the combination of ATI open-source video drivers and the recent kernels. Any potential Linux user with my hardware who somehow happened to try a live CD of any current distro would conclude that Linux is horribly broken.
Sound works, but it's not perfect, and manual trickery is involved to add some needed functionality at the expense of other less-needed (but nice to have) control of the sound chip.
The seemingly holy grail of Linux, suspend and resume, works better in some distributions than others, and a kernel update can make or break this feature.
After years of commenters telling me that my 2000-02-era equipment was too old to be well-supported in Linux, now I have 2010 hardware that's also not terribly well-supported.
If the hardware you happen to be running is working absolutely perfectly, good for you I say, but it doesn't negate the fact that the Linux experience isn't all barefoot romping in a thorn-free green meadow for everybody.
Yep, I'm duial-booting. I have Windows 7 on this laptop, although I probably spend 95 percent of my time in Debian Squeeze (with 2.6.32, the display works perfectly; kernel mode setting hasn't ruined Xorg for this ATI Mobility Radeon 4200 HD chip).
While Windows 7 is "better" than XP, it's not foolproof. There's a lot of cruft in there. I don't trust the NTFS filesystem, the Registry configuration concept, or the way Microsoft does core-OS updates. The myriad of ways other applications are updated (or not) doesn't work well.
And while the touchpad experience in Windows 7 is better than in Linux, it's not perfect (and it should be; and maybe in Linux I can figure out the configuration voodoo to make it so).
But the video display works. And it should continue to work. Sound works. It even mutes the speakers when I plug in headphones, and I don't have to mess with configuration files to have what should be the default behavior actually happen. And it survives a suspend/resume.
Like many of my fellow bloggers/journalists, I take the easy road and write about Linux and BSD for those who are already running them. In large part I do this because even for the "veteran" user, getting everything to work right is in no way a given. I share what I learn and hope others will do the same. The more desperate newbies who come across these entries in a Google search, the better. (But I can and should be doing more to extend my reach beyond the fanboy minions of which I'm a card-carrying, decoder-ring-wearing, slogan-bearing-T-shirt-bedecked member).
The fact that most Linux distributions (and to some extent the BSD projects) work as well as they do with hardware both old and new is a tribute to the thousands of hackers (paid and not, corporate and otherwise) who work on everything from the kernel to the video drivers, the various sound systems to the userland utilities.
But what Linux needs (and Ubuntu stands a chance of actually getting) is cooperation and actual contributions of code from the manufacturers of the hardware in our computers. This is how Windows works. If Dell/HP/Lenovo/Toshiba/Acer/Asus/Sony/Gateway and the vendors that make the various bits that go in their machines would take responsibility for the Linux compatibility of their devices and actually produced usable (and preferably open) drivers ... and THEN began shipping them with one flavor or another of Linux, then we'd be getting somewhere.
I know that Canonical is working on this. While Dell showed promise some years ago yet squandered it by pulling back from its Ubuntu strategy, and Asus created a whole category of hardware in the form of the netbook with a Linux OS yet dragged the whole line into the Windows camp, the potential for vendor-supported, Linux-loving hardware is definitely there.
And combining this manufacturer-driven compatibility with the proper marketing that Dell steadfastly avoided (automatic updates! no cost! all your applications free and united under one package manager!), the rest of the world that has never heard of Debian, Ubuntu, Slackware and Suse would have the true alternative to proprietary operating systems (pay now! pay later! pay for antivirus! pay for office software! no central software updating!).
We debate freedom endlessly. I'd be more accepting of binary blobs in the kernel if they didn't burn me on occasion, but I recognized that having hardware work — and work well — is more important (and sometimes the open solution does this, sometimes not).
Users should be educated on what software freedom means, why restricted audio and video formats are bad, and what the alternatives are now and in the future. But without freedom to make other choices, the free, open-source operating system world will continue to dwell in fanboy country, never enticing (or even attempting to entice) the other 99 percent of computer users out there, many of whom would like a choice in how they run the machines in their work and home lives.
Android in the mobile space appears to be a success. A smartphone alternative to the restricted Apple iPhone was and is needed. Not only do phone users want a choice, but the manufacturers who aren't Apple and the carriers who aren't AT&T want and need product they can produce and distribute.
Would Android have happened — and grabbed market share so quickly — without the largesse of Google behind it?
And like it or hate it, Google Chrome OS on branded hardware might achieve some success — and it doesn't look like Windows will be able to replace what Google has put together (although I bet Microsoft is working on their own Chrome OS-like devices).
For the Linux desktop, all we need is a Dell or two (or more) that are committed to offering Linux in a fair and equal way — with the pocketbook and stomach to actually let their customers know what they'll get by choosing a Linux distribution over Windows (or Mac OS).
I concede that restricted drivers, proprietary codecs and more hand-holding/less RTFM are part of this deal.
Hardware has to work. Multimedia has to work. Plain language has to replace geek-speak. Actual software-hardware partnerships need to be created and nurtured. I hate to boil it down to one thing, but I am doing just that: Linux needs manufacturer-supported preloads with lots of marketing, education and general fanfare.
Companies large and small need to be involved.
Can we do it? Say it with me, "Yes we can."
Will we do it? I have no idea.
17 Comments
Leave a comment
About this blog
I work for a large retailer. Almost all linux PC's get returned with the reason "It will not run Windows software". Sorry but most useers simply don't want it.
If there were actual Linux-equipped PCs in the retail channel, I might just believe you.
But there aren't, so I don't.
As I write above, relentless education about what Linux on the desktop can and does offer is a big component of this three-legged stool.
Steven
I suppose it depends what you mean by "in the retail channel".
Mainstream? Remember that Walmart has GOS pre-installed computers for a while, although there was opinion that they were underpowered.
We had the Asus excursion with the Xandros Eee PC.
Dell?
Does it extend to specialised outlets i.e. NOT mainstream a quick Google of: linux installed computers
http://www.linux.org/vendor/system/index.html
http://lxer.com/module/db/index.php?dbn=14
http://www.linuxhq.com/vendors/systems.html
There seems to be a mental block for many people which means "computers == Windows".
One reason why Android has been a success is that people don't expect smartphones to look and act like Windows. It turned out people did expect netbooks to look and act like Windows.
It may be that tablets/slates will provide another opportunity for Linux, because they are another form-factor that people perhaps don't expect to be "just like Windows".
An additional layer of frustration is that when Linux does something differently from Windows, that's because it's for geeks. When Apple does exactly the same thing, it's because it's a premium product supremely well-crafted for its users. See: virtual desktops; software repositories vs "App Store for OSX".
It would be nice if people were educated in what software freedom means, it should happen but unfortunately, in my experience, nobody cares.
"Freedom" is more about choice than about money.People should have a choice to completely configure their machine as they want(provided they have the know-how).That is the driving idea of open-source software.Marketing OSS should not be taboo, nor should demanding a market price for a commercial open-source software.
People are willing to give away money and embrace marketing if the software they are using gives them more power and control.The linux kernel opens up a world of possibilities to the person who wishes to harness it.Sadly,open source GNU software hasn't lived up to the same excellence.Most of it does not work as well as proprietary software produced by commercial vendors.That the kernel remains invisible and the GNU-licensed applications are all that the new user interacts with leads to disappointment.VLC media player and Mozilla Firefox seem to be(personally) the only ones that actually provided something substantially better than contemporaries,while remaining open-source.With every other application, you had to sacrifice bits and pieces, and that led to gradual disappointment among many-many linux enthusiasts like me.
Chrome OS, however, promises a different realm.The web is essentially open, and the quality remains essentially independent of platform.The true performance and security enhancements that linux offers can be combined in a lucrative package through such an offering, and for a change we can actually hope that the linux product will be equal to commercial ones in term of software/app quality, and better in terms of performance, security and customisability.
Linux may never have offered something to desktop users because,well, the good part essentially remained invisible and the mediocre part came to the fore.With the advent of the cloud, there is a good chance this will change.Google is a smart player for sure!
Agreed, agreed, agreed...
But I think the author and many (most?) linux/FOSS writers and supporters miss one critical point...
Specifically, if an open source OS (ie: linux) is to succeed, there MUST BE A WAY for 3rd party software (and hardware, for that matter, but mostly software) creators to MAKE MONEY!!!
The success of the iPhone and Android can highly be attributed to their successful implementation of their App Stores. The App Stores serve at least 2 important functions:
1) A means for a 3rd party app creator to make money!
2) A failsafe method for a user to browse and install 3rd party software (no tarballs, no make/install, no checking dependcies, etc).
Without the linux ecosystem developing a method/policy to allow and support the above type of 3rd party software (for users to use and creators to make money), then linux is doomed to an eternal fate as a fringe operation system on the desktop.
Tell me if I'm wrong.
D.
First of all, not every (or most) open-source developers work for nothing. Many are employed by companies, with some or all of their time going toward open-source projects important to that particular company.
Second, I don't think it follows that the hardware industry needs to worry about the proprietary software industry. I think we can all agree that even if Linux were to get some significant uptake on desktops, there would still be plenty of Windows and Mac OS sales out there.
Third, a great majority of servers run on Linux and other free, open-source OSes, and I don't hear anybody crying about developers being out of work. Most Linux kernel developers, MySQL and Apache developers are paid for their work. Many open-source projects are "owned" by companies that sell services or value-added bits.
Fourth, there's nothimg stopping software vendors from coding paid applications for Linux. If customers want them, there will be a market for it. If Adobe released a binary-only version of Photoshop for Linux, it would sell.
Fifth, the package-management utilities in almost every Linux distribution and BSD project are way more evolved than what Windows uses. You never have to unpack a tarball, do manual depenedency checking or run make unless you really, really want to.
Quote: Fourth, there's nothing stopping software vendors from coding paid applications for Linux. If customers want them, there will be a market for it. If Adobe released a binary-only version of Photoshop for Linux, it would sell. Endquote
Yes, it would sell, but only very few copies relatively speaking. And you also have the problem that often times software works on one distribution but not on another. There is a 3rd party German office suite that apparently is really good (professional quality - my uncle's recommendation) and comes in win, mac, and lin flavors, yet it will not install on my linux box. (PCLOS) Also, epson scanner utility iscan will not install on my lin-box. So, as many people have often said, the variations due to the plethora of distros, although empowering in many ways, can be a bane to the linux ecosystem due to issues like these.
Fifth, the package-management utilities in almost every Linux distribution and BSD project are way more evolved than what Windows uses. You never have to unpack a tarball, do manual depenedency checking or run make unless you really, really want to.
You are preaching to the converted. However, no matter what distro I have used (and I have used 50 or so), there are invariably software packages that are outside of the distro's package management system. Also, how does a 3rd party software creator charge $50 or whatever price for software that is freely available through Synaptics or apt-get or whatever.
I'll stand by my initial statement: Until the linux ecosystem gets a profit-enabling user-oriented standard "app-store" that will work amongst all distros (or at least most distros), linux will always remain as a fringe OS on the desktop.
Do you really think android or iphone would have // could have been so successful without their app stores??
What is the "geek ghetto"?
It's where the geeks are.
I thought it was pronounced DEEE-bee-en. Is it Debbie In? Or Duh Bean? Or what?
Since Debian is an amalgam of the names Deborah/Debbie and Ian, I pronounce it:
DEH-bee-en
DEBIAN GNU/LINUX.
i use also mint,which is great.
proud to be a geek.
I really enjoyed testing the Linux Mint Debian Edition. All it lacks that I want is the ability to encrypt partitions and/or volumes in the install.
It looks now like I'll be riding Debian Squeeze for awhile. It's working great and the great majority of applications I use are working very well (and since they're going to stay the same for the duration of the release, they'd better be).
With any "difficult" hardware (and I seem to have nothing but), once something works it's generally a bad idea to abandon it for the shiny and new. I do it but generally regret it.
Debian has been very, very good to me, as they say in baseball.
So I tried a few apps from a friend's iPhone/iPad.
Guess what -- they aren't available for any other platform.
The same apps have been available (in other forms and by other names) for Linux for about 20 years, but no one who uses the Apple products knows about them and thinks they were created for the iPhone/iPad.
The point is that marketing is the difference. What I learned is that Apple rewrote the existing apps and marketed them for the iPhone. Did they hire the original developer for this? Maybe yes, maybe no.
Ubuntu now has an app store. I'm certain that folks can buy apps there that are otherwise free (under other names/branding there), too. And they will, depending on the marketing.
The public will always pay for instant gratification, even if they can do something themselves for free. That's what marketing and earning money is all about.
Your argument has nothing to do with the intrinsics of the various OS's. It has to do with marketing.
"You get what you pay for..."
That's the reason i hear many times from peoples in my region. They prefer to pay a big company (choose one of two) to have a feeling of security from a big corporate company behind an OS.
To have many thousands of developers looking after some hundreds of distributions / drivers / applications seems too random for them.
Even when talking about companies behind Linux distributions (Red Hat, Novell, Canonical) - when they never heard about them, it can't be replaced with what they're used to - Window or MacOS. It's a too big risk worth to be taken.
Even while "we" are on the right direction towards the "Year of the Desktop Linux", there is still much work left.
I won't call this year the year of the Desktop Linux, maybe not the next year either, but the near future will make it become reality, with all the projects in the works right now (Bretzn, AppStream, Unity, etc.).