Walt Mossberg and Kara Swisher put Bill Gates and Steve Ballmer up on stage first thing at the D6 conference, and by doing so, let them set the agenda for the operating system discussion here at the show.
Microsoft's duo didn't do a great job of capitalizing on that position. Rather, they left a lot of room for other companies to excite the audience with newer ideas. Two companies here are taking on that charge.
The first, Ghost, demo'd Wednesday. Its product is a "virtual computer," as the company calls it. Hosted at Amazon Web Services, it's designed to be everything you need from a computer, except it's not on your computer. You get a file system (with 5GB of storage for free), a media player, and links to some apps. For example, if you want to edit a word processing file, you can launch it into Thinkfree or Zoho.
The fact that Ghost doesn't come with a bunch of its own apps is the key to this product, and what makes it more like an actual OS than many other Web-based OS experiments I've seen. The design goal of Ghost is that it acts as the clearinghouse for all your Web app accounts, letting you shuffle data between them. For example, if you have a document in Zoho Writer and want to edit it in Google Docs, Ghost will make the transition automatic. If you want to drag a file from your Flickr account into your Ghost file store, and later to an e-mail, Ghost will do that, too.
The interface for all of this runs within a browser, and that's the only place it will work at first. There's no offline version, and one of the venture funders for the company said the team doesn't believe that online/offline synchronizing will work with typically forgetful users (although a company representative later told me they're considering talking to Sharpcast to offer sync capability). Ghost is being built to be the one true glue that holds all your online apps together.
Also, Ghost is cool since it's a Palestinian/Israeli collaboration.
The company plans to make money through affiliate deals for the services it links to.
Answering the obvious question--How do you change user behavior to get them to move their computing to the cloud?--the founders respond that people have already moved their e-mail behavior online. So it could happen to productivity as well.
Like many other Web-based operating systems, it's a compelling demo but a confusing marketing pitch. The "you don't need your own computer" line doesn't work all that well when everyone has their own laptop already. Ghost is a very long bet. It is not a product for today (and it's still in early alpha testing anyway), but it is one of the most interesting Webtops I've seen, and more OS-like than most.
The second company, Glide, will showcase its new service on Thursday. Glide is more of Web application suite than a Web OS. It's more mature than Ghost, and more comprehensible to the average user today.
The new "Glide OS 3.0" is a fairly complete Web-based desktop, with a word processor, a presentation app, a spreadsheet, e-mail, calendar, media players, and so on. I was critical of the apps in Glide's early versions, but they do keep getting better. Glide also has a rights management system built into its system, which lets users closely control what happens to their documents if they choose to share them.
Unlike Ghost, Glide has an offline version, and a sync engine that keeps your online and offline files in lockstep. Glide works on mobile devices, and the offline apps works on PCs, Macs, and Linux devices.
Glide does not offer real-time collaboration, like Google Docs does, though.
These two suites show two different approaches to building Web-based app platforms. Ghost is a glue app--really more like an operating system than a suite, since the idea is that you run your favorite apps through it, and use it to manage your file storage and sharing. Glide is an app suite--the Microsoft Office of Web apps.
Of the two projects, the Glide direction is more comprehensible. Users get that it's a competitor to paid productivity suites. Ghost is far too novel an idea for mass adoption; and I do worry that by the time users are ready for it, its strongest capabilities--data transfer and universal online storage--will be typical features of online apps. But it is, still, the more interesting idea and the bigger bet.
Meanwhile, Microsoft continues to gingerly explore online extensions to its core suite. And although I have been known to complain about the glacial evolution of the architecture of its application suite, in truth it's probably moving at close to the right pace to keep users onboard.
Microsoft could not do Glide today. It wouldn't work for its business. And it certainly couldn't do Ghost (although it's possible that Microsoft Live Mesh [stories] is an early move in this direction). But these new Web-based operating systems do show us where Microsoft must eventually move because that's where many of its users will be going.