Is the "killer app" argument dead?

The other day I was talking with a bunch of other tech heads about the ongoing Windows vs. Mac vs. Linux market share war and why, from a stats point of view, that Mac and Linux were still trailing far behind Windows.

"It all revolves around killer apps," pipes up one of the tech heads. "The problem with both Apple and Linux is that neither OS has a compelling killer app driving the user base."

The conversation then went off on to discuss apps such as Photoshop and AutoCAD and how with Windows you have a single unified platform that'll run almost any professional grade package that you can think of, while with Mac and Linux your choices are restricted.

Bull.

Now, I will accept that at least part of the Windows market share is down to pro-grade apps. Certainly Microsoft Office was a great driver of Windows, especially a few years ago where there was no free alternative (OpenOffice). And then there's how the gaming industry embraced Windows while giving other platforms the cold shoulder. That helped a lot. But the problem with the "killer app" argument is that in the last few years it has worn thin. Let's face it, if you piled up all the world's Photoshop and AutoCAD users (and by users I mean people who have bought a license for it, rather than having pirated it) then you have a pretty small user base. Your average user isn't likely to have Photoshop installed, and is less likely to have AutoCAD or LightWave. What's more, many home PCs nowadays don't even ship with a cut-down version of Office or even Works. Even gaming has lost significant ground to cheap, powerful and easy-to-use gaming consoles.

What's the single application that you're most likely to see open on a consumer PC nowadays? It's the web browser. I'm staggered by how much people get done through their web browser nowadays. Email, organizing and displaying photos, staying in touch with others, banking, casual gaming and a lot more. While there are some users (take me for example) who feel that each of the above tasks needs a specific application installed, there are an increasing number of people who consider the web browser to be the killer app installed on their system which gives them access to a vast (and constantly increasing) number of web services that are replacing the installed application. Most of these web services are free and most happily cater to the needs of a good 80% of the users.

Let me offer a case in point. Photoshop. I have Photoshop CS3 installed on my main system. It's a hell of a program, but the truth is that it's much more of a photo editing program than I will ever need. I know I should learn what each and every button does, but given that I've been a Photoshop user for quite a few years now and I haven't, it's a fair bet to assume that I'm not going to. Truth is, it's a far bigger application than I really need. Same goes with Office. I have the full Office 2007 installed (Ultimate edition I think it was called) and I don't think I've ever fired up Access, Publisher, InfoPath or Groove, and I've had PowerPoint open no more than a couple of times (I hate PowerPoints ...). Yes, I do use Word, Excel and Outlook a lot (my life is contained within Outlook) but I know I could free up a lot of disk space and simplify things a lot by uninstalling all the stuff I don't use. But I don't, instead choosing to keep it on there because "it might come in useful one day ..." Oh, and I installed all of Office because I paid for it.

I know that I could be doing a lot more work through the browser than I do (for example, I'm typing this blog post in BlogJet rather than using the online interface ... I've been doing things this way for years and I'm loathed to change) but even I'm feeling the influence that Firefox is having on standardizing the web browsing experience across Windows, Mac and Linux.

Thoughts? Is the browser your killer app or do you rely on other killer apps?