Résumé

What Browsers do (that they shouldn't)

The browser wars are no fun. Looking at a site like del.icio.us or the software features of Google’s search and web histories, it’s safe to say that features once relegated strictly to browsers are being taken up by web apps. In the same way that I want to predict everything else, I would like to give an idea of the direction browsing is going.

At home, I prefer Mozilla Firefox and browser tabs. But at work, we run IE6 (soon to be upgraded to 7), and have to settle with learning different applications and interfaces. In the course of a given week, I may use ten different computer workstations between home, school, and work, each with its own Favorites(Bookmarks), Histories, Cookies, Autofills, and all the other stuff stored in a typical browser. Some of that stuff is mine, while the other users contribute to the same lists. Not really an efficient model, is it? If my assumption of the white-collar workplace is correct, then I am certainly not alone in the number of computers I use.

If all my information could be stored securely on the web and accessed from anywhere, then I would save a lot of time and effort when it comes to logging in to sites, autofilling forms, and generally the simple tasks of browsing.

I see a future of the web where the browser is much less involved in the overall web experience. A web-based browser-within-a-browser (likely using html frames and an ajax backing) would store and manipulate all of the data commonly stored locally by the browser — www.palary.org is an example of such a browser, but without some of the features I’m looking for.

Such a web-based browser could store Cookies, Bookmarks (to and from del.icio.us, Google Bookmarks, or otherwise), History (once again, Google services could be integrated), Greasemonkey scripts, Bookmarklets (oh how I want a digg button everywhere I go), and a host of other things that you might be accustomed to only accessing locally from a single workstation.

These browsers will pop up all over the web, and a new browser war will start, all while the actual browser programs are relied on for less and less functionality. Microsoft and the collaborators of Mozilla will instead focus on standards-based rendering of documents, and providing services that are most useful from a single, local workstation (like local files, usw.) The address bar and other interface elements would only be useful once per session (and the web-based browser could be styled to the user’s preference).

The alternative to this vision is that all the browser companies and developers work to develop a standard for all the non-local user-specific data that can be transferred between browsers and computers, preferably securely via the web. Since it would be easier to make a browser-within-a-browser, I’d say this latter solution is much less likely.

For further reading, I suggest http://fortuito.us/2007/04/living_online_with_web_apps which indicates that you can run your life on the web, except for the pesky browser functions — Google Browser sync readily handles this, but who has install and customization privileges on all the computers they use? Browser Sync, likewise, does not handle all the browser functions that would be most useful to all users. Plus, I inherently want the option to not give Google all of my data.