Here’s the Link to the Slashdot Article (I didn’t actually RTFA, at least the one that’s linked to from Slashdot)
http://linux.slashdot.org/article.pl?sid=09/04/02/1317246&from=rss
I’m writing here to give my two cents on this post, because it was too late for me to post it on /. .. as my post would have been sent to the abyss to never be seen again.
Just to give my self a little cred before you start reading and flame me half way through, I prefer FreeBSD for my servers because I like how easy it is to setup an empty shell of a box very quickly and add what I want to it just as quick. I like the ports and how easy it is to upgrade them and how everything builds the way I like them to be built. I have a few Linux servers, and I like them, but I prefer FreeBSD, but my response isn’t about servers, it’s about the barier to entry to the Linux destop.
Anyway, my main gripe on Linux at least as far as desktop applications are concerned, is the applications that are built for it. People don’t really use the operating system for the operating system sake, they use it for what’s available. For instance, my favorite editor in the world is UltraEdit, I love that editor, but the only problem, it only runs on Windows (at least the last time I checked). Now there are plenty of replacements or alternatives to my favorite editor such as JEdit and many others that have similar features, but if I were forced to drop all my Windows boxes and slap Linux on them, JEdit is the editor I would probably switch to. I like the editor, and it has a butt load of plugins, but it really is a pain in the ass to set up the way I want it to be setup, and most the plugins seem to be written just to match the features from other editors (not just UltraEdit) as opposed to really creating something innovative. Well, I hope if the JEdit guys read this, they don’t take it the wrong way, because I really do like JEdit. Another problem is, training other people here how to use it, how to install plugins, how to configure it and so on and so forth, where as the strong Windows apps have a much lighter barier to entry and don’t take me anywhere near as long to train people how to use them.
Now, UltraEdit was just an example, but the list goes on and on, such as the Photoshop vs Gimp… getting our graphic designers to even look at Gimp is trying to pull a caffeine wired mountain lions teeth. Alternative to quick books? I’m sure there is one, but I don’t know about it. One thing I’ll give linux is Open Office, it works just as well as a solid replacement for the big evil I will not mention (because I’ve pulled more hair out because of that damn program than any other program ever… EVER!)
Now if Linux is going to gain ground, it’s application writers are going to make sure and package more features into the core product, which inevitably means linking to more libraries which is ok if the package manager is good. From what I’ve seen, Debian and Ubuntu (Same package Manager?) are very good at this, so that’s one piece of the puzzle… but if the package manager isn’t good or stuff has to be built from source, could be devastating to n00bs.
Also, the application writers are going to have to do more than just package more features into the core product, they are going to have to build features in that nobody else has, and they are going to have to continually do it because the others out there will steal their ideas and do steal their ideas, and not only that they might even try to patent them!!!!! 
The patenting and stealing is not really an issue with Linux itself though…
The biggest reason why people don’t adopt Linux from what I can see though, is that the distributions need to find more ways to monetize and market their distributions. Without marketing, there’s no way to let the general public know about it, other than bundling it on cheaper computers. I have yet to see an ad on mainstream television about Linux, and until that happens, it won’t gain significant ground as a viable desktop option.
