Over the past few months, I've been progressively migrating to what I call the "fat browser" model for my day-to-day computing tasks. If it ain't running in a browser, I don't want to use it, or at least use less of it.
My experience so far? Pretty good in most areas, just a few legacy areas to sort through.
And I think that -- one we move beyond virtualizing legacy desktops -- more people will come to appreciate the fat browser model as their preferred client experience.
What Is A Fat Browser?
For me, it's best exemplified by something like an iPad, or a netbook, or perhaps a smartphone.
Your browser is your primary UI. There's always good connectivity, whether it be wifi or 3G/4G. There are a few local applets and local information stores on the device if needed -- nothing important, though.
I differentiate between a fat browser and a thin client in that I've got a smattering of local device capabilities that help me get things done: some file storage, simple productivity apps, etc. as needed.
There are all sorts of on-line storage options for the stuff I care about. All sorts of cool web apps for finding stuff, and sharing it with others.
And yes, my iPad experience kind of brought it all into focus for me. For now, the iPad is the model of what I'd like my fat browser to be -- nice UI, good connectivity, good battery life, decent apps, comes in small (iPhone) and large (iPad) sizes, etc.
I'm assuming that we'll see more variations on the iPad theme going forward, hence my suspicion that most of us are heading for a fat browser world for the majority of our personal computing tasks.
And a touch interface can be amazingly productive for certain tasks -- like playing Plants vs. Zombies :-)
Where Are The Limitations?
Clearly, that sort of thing can work for consuming content, but what about generating content?
Well, most of my content generation is text-based. Give me a decent keyboard, I'm good. In addition, I have to manipulate powerpoints occasionally. Between Google Apps and Keynote on the iPad, I'm generally good.
Now, if I have to assemble a graphically-intensive presentations from scratch, I'd prefer a traditional desktop. Ditto for working with graphics, animation, playing with code, etc. -- local computing is still preferred.
But I don't do a lot of those tasks these days.
If there's no network (sometimes the case), I have enough local storage and capabilities to be productive. I can read web pages that I've queued up. Type up something or other. Even mangle presentations for that next meeting if needed.
Not exactly a full set of capabilities, but enough to be productive until I get connectivity again.
And, finally, there are a few legacy corporate IT applications in my world that don't like modern browsers. I queue that stuff up for when I'm at a traditional desktop. None of what needs to be done there is especially urgent or time-critical.
Where Does That Leave Enterprise IT?
My guess is that enterprise IT will -- over time -- migrate towards preferring a controlled browser experience for their knowledge workers. There will be control over who can access it, what they can do, and measures to keep certain kinds of information from leaking out.
If I -- as a knowledge worker -- feel that's too limiting, I'll simply fire up another browser instance that's more to my personal tastes. Running two browsers side-by-side (or in another tab) isn't too taxing to most of us. Add in a little cut-and-paste, and I think I'm good to go.
In some ways, this reflects on broader theme of IT provisioning user experiences, rather than provisioning physical hardware. IT groups have to provide access to a set of IT services to their users -- that never goes away for most people -- but the value has clearly shifted to the service, and way from the hardware it runs on.
How About You?
Take a moment, and look at what you do in your work and blended personal/work life. Could you see yourself running primarily on a fat browser model before too long?
I'd be interested in hearing your thoughts ...

I agree with the gist of your post. Although HTML5 will presumably improve things, I'm skeptical that trying to do everything through a browser is just too much of a force fit. OTOH, that doesn't mean we have to throw up our hands and go with a full-blown general purpose operating system model. An app store approach (which is at least akin to what you're calling a fat browser) seems to be a nice in-between approach and, yes, the iPad has helped further crystallize this for me.
(That said, I'd like to see further advances in transparently synchronizing state. Something like an iPad's state is at least arguably more device- than identity-centric than it should be. One of the reasons is that iTunes isn't nearly as "cloudy" as it should be.)
Posted by: Gordon Haff | June 02, 2010 at 12:02 PM
I recall a story told at a conference years ago about some company doing "man on the street" type tests with people using PCs and GUI based OS. The story goes that an older man looks at everything then picks up the mouse and speaks commands into it. The people running the tests thought it was hilarious and retold the story forever. Moral of the story is no one asked the question "Why not?"
I hear lots of people discounting the iPad as a toy or a big iPod but I see people trying to use it in different environments. My doctor told me today that her husband, also a doctor, is trying to use it in his practice and that other doctors are wanting to try.
I think many segments of our world are ready for the "fat browser". Some will find a way to deliver while many others will fade away still laughing at the iPad users.
Posted by: David Hurst | June 03, 2010 at 10:29 PM