The new wave of operating systems is, if not useless, at least pointless without the internet. Believe me, I've tried. I couldn't set up a Nexus 7 without connecting it to Wi-Fi, Mountain Lion isn't sold on DVD or even thumb drive, and Windows 8 without the internet is basically Windows 7 with a bizarre, pointless, fullscreen Start Menu. Every app wants a log-in, needs to be married to some cloud service, longs to share, hopes for updates, bathes your computer in push notifications.
This is a wonderful thing.
Talking to the internet was a glorified version of Palm HotSync
For too long, desktop operating systems have skated by on an outdated model of the world. You would configure calendar "syncing," you'd "upload" your documents, you'd "back up" your files — it's like talking to the internet was a glorified version of Palm HotSync. The internet native experience in your browser offered a radically different vision where you never needed to "sync" new listings on eBay, where you could "share" a document instead of "send" it, and where your calendar was simply your calendar, not some frozen-in-time, frozen-in-place simulacrum of one.
Once upon a time, I was scared that 2012 would look like Gmail 2005, or even worse: Hotmail 1998. But it turns out the Chrome OS vision of the future, where all we need is a browser, is also wrong. The consistency and familiarity of OS interfaces makes those nascent "web designs" look childish, and our browsers still struggle with simple concepts like drag-and-drop, responsiveness, native code, hardware acceleration, and multitasking. Microsoft just renamed "Hotmail" to "Outlook," finally shrugging off the vestiges of 1998 web app frustrations, at least in name. And when we're truly astonished at a web app, crying out "I can't believe this is running in a browser," we're talking about exceptions that prove the rule.
You could think of the desktop's move from "digital hub" to being another part of the cloud as a demotion, an end of an era, but I think the Desktop 2.0 isn't just the future of operating systems, but the future of the internet as well. With the desktop becoming an internet native — a direct client of the cloud instead of a reluctant collaborator — I think the browser paradigm will begin to look like a clunky proof-of-concept for many of our favorite services.
This has been happening for a while with Twitter, where native apps provide a vastly more convenient and powerful method of dealing with the onslaught — even as Twitter attempts to beef up its web app with exclusive, sexier "functionality." More recently, Google has bought Sparrow, a purveyor of desktop and mobile email applications, which I see as a clear admission by Google that trying to make a Gmail app that's simply a window to HTML5's version of events was a total failure.
And so, instead of going to weather-4-u.com for info, a local personal assistant cheerily reminds us to bring an umbrella. Instead of just talking to our friends over Facebook messaging, our friend's profile pictures have populated our IM software, contact lists compiled freely from a multitude of sources, URLs irrelevant. Every message bubbles up in a Growl notification long before we click over to emails-R-us.com. Instead of churning through an upload dialog for the twentieth time, we just drag a folder of pictures into Google Drive.
The Chrome OS vision of the future, where all we need is a browser, is wrong
Many of these features have mobile operating systems to thank. With small screens and molasses-slow browsers, web app purveyors were required to build APIs and apps that would allow the devices to join the conversation. But what we learned is that many services make more sense when the hardware is intrinsic, not just a "visitor," which is how we used to refer to "web users" arriving at "websites." The original iPhone's vision of "apps" being glorified bookmarks was wrong wrong wrong as well.
But I think we haven't even scratched the surface of what a desktop, complete with keyboard, mouse, 16GB of RAM, a hot new graphics card, and the low level code to run it all at a lightning pace, can do now that it's an intrinsic member of this internet conversation. What would Maya or Photoshop look like with Google Docs-style collaboration features? What if Siri or Google Now got a good look at your actual workflow — you know, the stuff you do for 8+ hours a day on a "real" computer — and figured out how to help? When everything is an API, are modular applications a la OpenDoc finally possible?
Desktop vs. web is no contest
I know for myself that now that I'm confined to a desktop operating system that's self-sufficient — a network of one — I'm much more productive, at least the definition of "productive" that involves "producing" things. Instead of Safari being my number one app, I'm spending most of my time in Logic, Audacity, Byword, and Notational Velocity. Because, outside of my music collection, there's nothing to passively consume on my computer; it's a content-creation machine. I know the consumption vs. creation argument gets thrown around a lot in the tablet vs. laptop comparison, but desktop vs. web is no contest.
Before I got a 56k modem, I used to use my PowerPC 6100 primarily to make stop motion animations. And then I had AIM and messages boards, and I was a filmmaker no longer. The browser version of the internet turned me into a "web surfer," a role where content creation is possible, but not a primary pursuit. My hope is that now that we can use the internet more effectively outside of the browser, that shift of power can reverse.
Yes, as Nilay pointed out in his Mountain Lion review, the PC is the hub no longer. iCloud, Dropbox, SkyDrive, Facebook, Twitter, Spotify, and Google Drive are. And as much as I complain, the web browser has provided the spark for each of these innovations in turn — its rapid prototyping, low cost of entry, and endless malleability have spawned perhaps the greatest technological boom in history.
But the browser isn't the endgame, and the desktop is no thin client. Maybe it's a "Thick Client"? A "Mega Spoke"? A "Cloud Captain"? I don't know what to call it, but I think we'll have years to figure out the terminology, because the desktop isn't going anywhere soon.