Skip to main content

Linux desktop market musings

Posted by evanx on May 17, 2008 at 3:12 AM PDT

One site i visit every day is which is a Linux news site. Today that took me to a
blog entry "Five Reasons Red Hat Should Ignore Consumer Linux Desktops"
by The Var Guy, who i typically enjoy. I posted a comment in agreement, and when i do that, sometimes i like to repost it as my own blog entry, with some imbellishments, as i've done below, even though this is a Java blogging site and not a Linux one per se. However, my only other blog is decidedly non-techie.

So on the issue of Red Hat Linux desktops, we know that some of Red Hat's enterprise clients are migrating desktops to Linux eg. for call centers et cetera, and so Red Hat have their Enterprise Desktop for them.

Consumers are not going to want to pay for support, and at the same time want a lot of support to get their goodies working, from wifi to cameras, iPods et al - certainly not worth the hassle because let's face it, when you can't specify the hardware, Linux is a royal pain in the arse compared to Windows.

I think Canonical is investing in the consumer desktop primarily for personal passionate reasons of Mark Shuttleworth, and secondarily as a way to get into the market of enterprise desktops and servers with monthly support subscriptions, in order to sell the company because then it'll be getting too corporate and boring for the Benevolent Leader.

So the consumer desktop is Mark Shuttleworth's entry strategy, and the enterprise is his personal exit strategy. Red Hat is in the enterprise already, so no need for them to market consumer desktops as an entry strategy like Canonical.

Having said that, the "consumer desktop" is probably not Ubuntu's demographic, as much as the "technical desktop" ie. the Linux PCs of enthusiasts, system & network engineers, and software developers.

The risk then for Red Hat (and Solaris) and is that these engineers and developers become such passionate Ubuntu advocates, and/or that
Ubuntu skills are most readily available and accessible, to encourage management to choose Ubuntu over RedHat in some cases. Probably not for the NYSE, but maybe in some small to medium enterprises that might otherwise have choosen Red Hat.

But anyway what Red Hat looses on the swings, it gains on the roundabouts via Canonical's contribution to Linux desktop development, eg. GNOME et al, testing and what-not to accelerate upstream bug-fixing.

Do you think Red Hat et al are, or should be worried about Ubuntu's meteoric rise, and how they might or should respond?

Related Topics >>


Hard question! :-) My point is that the desktop challenge has always been a relevant point in the Linux world, so it's right that somebody tries to pursue it. You're right, matching the hardware is often a PITA and I know something about it (I've been trying to do presentations with my MacBook Pro and Linux for several months and I've failed because I can't get a video signal to the external port, even though with Hardy Heron I've done some substantial progress). But looking back at only two years ago, Ubuntu's progress has been great. I've expressed many times my unsatisfaction with Mac OS X, but at the moment I don't have any alternative, so I hope that the Ubuntu thing will go on. Next release in 2009 could be interesting.

This success is clearly pushing Ubuntu at the point that I've read somewhere that the name "Ubuntu" might even become more popular than Linux (I find myself using mostly the former, indeed). Clearly this is a branding success and competitors might be worried, but this is happening in the end-users community; in the enterprise world, RedHat is well known and BTW it's much more than the operating system, being indeed almost a full stack of products (up to the application server). So, in the end, I don't think RedHat should be worried at all.

fabriziogiudici, the next Ubuntu release should be in October of 2008, not 2009.

And hopefully people will use "Ubuntu" more than "Linux", because Ubuntu really is the operating system. Something that works on "Ubuntu" may not necessarily work on "Linux" in general.

I think you have an excellent point--developers will recommend what they use. Microsoft knows this--until recently, their dev tools were best of breed and dirt cheap. I've seen a lot of projects using MS tools just because the folks in the trenches were familiar with them. Those things started at some departmental level, they percolated up, and before you knew it, MS IIS and SQL Server were everywhere.

Now Ubuntu is getting that message by making it dead easy to get going with Linux. And of course Apple talks the talk, without quite walking the walk, at least when it comes to Java .

The message for Sun is that you've got to have the hearts and minds of the developer in the trenches. Don't just do the top-down big-iron thing. Make it easy to bang out a web app with Netbeans and JSF+JPA, and to deploy and manage it. If it's easy, that's what people will do. It is getting better, but it isn't easy enough.

Same for Red Hat. Top down, without the hearts and minds of the developers, only works for so long.

Recently I bought core 2 duo sony vaio CR36 laptop with 32 bit Vista home premium pre-installed. As its 64 bit machine I dumped 32 Vista and installed 64 bit Ubuntu 8.04 Desktop. Excluding two well known issues(Motion Eye webcam driver and Sleep/Hibernate not working. These proprietary driver issues should be any Linux Desktop distro's high priority as far as end user experience/productivity is concerned.) system is working great with LAMP setup. Recently I showed ubuntu 8.04 to couple of what we say 'normal' users and demo-ed the video/audio, firefox and openoffice applications (primary PC usage apps for any lay person) on it. With the advance desktop effects on,the lot was pretty impressed and was eager to try it on their own home/office machines currently running windows. This was not convince able just couple of years back with all major linux distros mainly focusing on server market with geeky mindsets. IMHO popularity and market penetration of both the desktop and server linux distros will provide greater value to end users as result of increasing developer community and vendor support. In one word its about increasing the pie while sharing it.