Is there a long-delayed echo in here? Yesterday's announcement of Google's Chrome OS observed that "The operating systems that browsers run on were designed in an era where there was no web...For application developers, the web is the platform."

Pardon me if I'm reminded of something that appeared six years ago in eWEEK:

If I asked you to name the Internet's dominant operating system, you'd probably nominate Linux, Windows or possibly Solaris. My answer would be none of the above. Increasingly, our most value-adding interface layer is Google—and our industry's annals of operating system wars and browser wars are looking ever more like ancient history.

My point, in that May 2003 column, was that the major functions of a personal computer operating system are increasingly valued in the context of the Web at least as much as they are in the context of a single machine.

• As far back as CP/M or DOS, we've wanted to have access to data without concern for the details of the device that stores it; today, we want that abstraction to include the location as well as the type of hardware.
• As far back as the original Macintosh OS, we've wanted to be able to call up a piece of content without concern for which application created it; today, that abstraction resides in our browsers and their portfolios of plug-ins as much as it does in our local libraries of thick-client applications.

If you set out to design an operating system today, I'm sure you'd treat access to the assets of the cloud as a first-class function -- not as a secondary feature of some optional application called a "browser," suggesting that the user is casually exploring cyberspace in hope of running across something interesting. Further, you'd design for a model in which the cloud is at least as interesting for what it does as for what it knows: a world of active content, not just static pages or even Web 2.0's interactive communities.

The Web we have now is a place where people can package and offer their expertise in process, as well as their knowledge of facts. Making that expertise accessible, without the nuisance and complexity of the old model of thick-client computing, is the mission that's shared by cloud-committed companies like Google and salesforce.com.

Chrome OS has the potential to be an existence proof of a fundamentally better way. I look forward to seeing how it turns out.