Time to Redefine the Browser

5:30 PM -- There is no more important debate in wireless than whether wireless Web services will replace the thick-client/local-applications model that represents the logical evolution of the computer itself. I’m going to continue to argue that we want to move towards Web services for reasons of cost, management, and flexibility. While we might base further mobile platforms on real operating systems, and this appears to be the likely direction, all we really need to support is a big browser, a local Web server, some caching functionality, and maybe a little more. But there should be no need to port most applications to every mobile device -- that’s both expensive and impractical.

And this brings me to the most critical piece of local software, the browser. Now, “browser” is an increasingly inaccurate and inappropriate term. The modern browser is more analogous to the “intelligent terminal” of yore -- the DEC VT-100 and the IBM 3270 come to mind here. While these were hardware, of course, the important part was the API defined for each. This allowed applications to be run on any terminal that was compatible.

A similar challenge awaits us today. First, we should redefine the browser as an application interface. I’m still thinking about the right term for this. And we should finally standardize the browser itself, with an ANSI or other standard. Vendors should stop competing, by changing the API to the browser. I’m not a browser expert, but I have run across a grass-roots organization working in this area -- the Web Standards Project . I don’t know much about them, but I already wish them well.

— Craig Mathias is Principal Analyst at the Farpoint Group , an advisory firm specializing in wireless communications and mobile computing. Special to Unstrung

Sign In