There is a lot of doom and gloom out there about the future of the Web Browser. The Web is dead.". "The browser wars are over and the browser lost.
I see this as very short-term thinking. Netscape went several years without shipping a product. That was three years where innovation on the Web stalled. Now they are still productizing the code they generated during that period. So the evolution of the browser has stalled. That is no cause for panic. A browser-based future is just as inevitable today as it was when Netscape went public, for all of the same reasons.
First, let's start with a definition of a browser. The defining characteristics of a browser-based application are:
These Four Features are irresistable. And they are core to two areas of innovation on the Web today: blogs (anarchic fora for personal expression) and portals (regimented corporate application interface hubs). The popularity of blogs and portals shows that even with the recent slowdown of browser innovation, the Web Still Matters and innovation still happens.
Individuals and corporations only give the Four Features up reluctantly. They must be tempted by some competing feature. For instance they will sacrifice the features for a much richer user interface. But with the advent of XForms, XUL and SVG, browsers could offer a much richer user interface. Or they will sacrifice them for better perceived performance. But with client-side JavaScript, XML and XSLT, Web applications can minimize their network latency easily.
In other words, the choice between Web features and desktop app features is entirely a false dichotomy. Using Web techniques, we can build a Universal Interface Virtual Machine (UIVM). It could give us the best of both worlds. What is holding the UIVM back?
We are in the funny situation where the leading browser vendor is really not a fan of browsers. The last thing an operating system vendor wants to support is a product that makes the underlying operating system irrelevant. But it is instructive to recall how Microsoft came to get the leading market share in the browser world. Pre-installs was not sufficient. When IE 3.0 was the pre-install, everyone downloaded Netscape. Microsoft beat Netscape by producing a better browser: faster, more compatible, more standards compliant.
Microsoft had to legitimize the browser (and especially the CSS standard which might have otherwise been ignored) in order to grab the market share that would allow them to slow or halt the development of the browser. But they did not succeed and probably Bill Gates is smart enough to know that they can't, in the end, succeed.
Sure, Netscape shot themselves in the foot with a total rewrite. But that rewrite is done now. The level of innovation in the Mozilla world is incredible. As far as the UIVM is concerned, we are back in the Mosaic 0.9 days. We have the outlines (XML, XSLT, CSS, XHTML, SVG, XForms, ECMAscript, perhaps XUL). It seems clear to me that this set of technologies can close many of the gaps between the browser and mainsteam GUI apps. Given that the desktop GUIs progress extremely slowly, it is only a matter of time until the browser catches up.
Once the browser GUI catches up, the virtues of zero-install, URL-addressability, standards-basis and client agnosticism will kick in. It doesn't matter whether it takes one year or ten.
Microsoft will respond as they did in the early days of Netscape, with a better browser. They have only two responses: ignore the new platform and let someone else own it or be the best even if it wreaks havoc to the rest of your strategy. Microsoft could no more stop the Universal Interface Virtual Machine than IBM could stop the rise of hte PC. Somebody is always out there innovating, taking advantage of Moore's law and even the monopoly platform vendor cannot stop them.
They can't cut off Netscape's air supply again: half of 0 revenue is still 0 revenue. They could strong-arm AOL into jettisoning Mozilla, but then it would just migrate to Sun, or IBM or the Apache Group or ... Mozilla under the careful leadership of the Apache Group would be a nightmare for Microsoft.
And anyhow it doesn't matter if Mozilla died. There are other browsers. The key point is that as long as there are people clamouring for the Four Features, there will be developers improving the state of the art and incrementally reaching towards the UIVM.
HTML rendition created using stylesheets by Wendell Piez of Mulberry Technologies.