semyendow: Issues for 2012 #4: What Should the Browser Become?

NCSA Mosaic logo.jpg It was NCSA Mosaic that introduced the world to the Web. Since that time, the browser has become the principal software-based element in all the world's digital communications and transactions. It is the harbinger of a very powerful new class of dynamic language interpreters, making JavaScript the unlikely, though undisputed, vehicle for conveying interactive functionality. And for some manufacturers, it is the center of an apps ecosystem unto itself.

So the browser is in no danger of disappearing. But as the Web expands into a delivery mechanism for all forms of applications and services, is a stand-alone, exclusive window into the Web, complete with bookmarks and toolbars and add-ons, truly the most sensible usage model for a system that may yet embrace all of computing? This is a question Mozilla began asking last summer, and whose answer remains inconclusive.

Sponsor

Either the new form factor is here, or it's not

As Mozilla Foundation Chair Mitchell Baker announced last July, "I think of apps as a new 'form-factor' for the Web.

"The browser is no longer the only way people access the Internet. People also use more focused 'apps' to do discrete tasks, and often feel a strong sense of attachment to the apps and the app model," Baker wrote. "This is an exciting addition. Mozilla should embrace some aspects of the current app model in addition to the browser model."

But Baker's announcement points to an evolving, wiki-like document elsewhere on Mozilla's Web site, the latest version of which seems far less willing to abandon the browser-centric Web as Baker's original inspiration. "The next generation of innovation on the Web," it reads, "will be anchored by a browser that is an honest broker committed to the interests of the individual user and developer, providing amazing experiences that match those offered by proprietary platforms; and user control and developer reach and freedom that is superior to proprietary platforms. As Firefox has transformed the browser landscape before, it must do so again."

111228 01 Future Firefox screenshot.jpg

To that end, Stephen Horlander, who rearranged the contents of the browser for Mozilla Firefox version 4 in 2010, has been busy rearranging them yet again for a future version, as yet unnumbered. Under consideration is an adaptation of menus to include captioned tiles, like the home screen on an iPhone.

While this goes on, the organization has renewed its ongoing deal with Google for the next three years, which had been the principal source of the free Firefox browser's revenue. The deal retains Google's place as the search engine of choice in Firefox's search box, while Google continues to develop the competing Chrome browser that doesn't even have a search box - its text search capabilities are shared with the address bar.

Though the notion of Google Chrome as an OS for laptops and netbooks hasn't yet caught on with consumers, it has succeeded thus far at advancing the vision of the browser as an environment for real-world work. It's like a guiding star for the product line, a clear direction. So one should keep in mind how clever Google has been with its product strategy when assessing its motives behind extending its default search agreement with Mozilla: an agreement which presumes that for at least the next three years, Firefox will maintain a prominent window adornment called a search box. This while Microsoft, in establishing its bold, new direction for Windows 8, begins publishing an entirely window-less build of Internet Explorer 10, exploiting minimalist design ideas ironically pioneered by Mozilla's Aza Raskin (who will, incidentally, have officially left Mozilla at the first of the year).

If Windows 8 succeeds in the PC and tablet markets in 2013, the browser that most users will find themselves using will, once again, be the one that users start by default. (As some users would say during the heyday of IE5, "Isn't the Internet the one with the blue "e?") On the other hand, if Windows 8 fails, then the healthy and growing tablet market will most likely be led by iOS, followed by Android - neither of which have proven fertile fields for Firefox to flourish.

Are Web apps not really the Web?

The rise of HTML5 has advanced the myth that applications run outside the browser are somehow "not the Web." Thus users find themselves defending the sanctity of the browser-based environment without really knowing why.

To be the Web, RSS pioneer Dave Winer suggested in December, one has to provide links. "I don't care how ugly it looks and how pretty your app is, if I can't link in and out of your world," Winer wrote, "it's not even close to a replacement for the Web." (This from the fellow who, ten years earlier, advocated an all-out defense of Java.)

Winer's argument holds some water in that no network is truly viable without connections. The Web was originally conceived as a network of contextual connections between related topics; it gave rise to a delivery system, HTTP, which became ubiquitous thanks to the popularity of the browser. That ubiquity opened up a path for a new class of distributed applications that was not beholden to any one company's or organization's proprietary strategy.

If that class of apps is not really the Web, though, as Winer suggests, then the browser truly does have a place in our future - just a limited one. Articles with hyperlinks are nice, but their utility is limited. The role that Mitchell Baker foresees of an "honest broker" of functionality sounds less like a newsreader and more like an operating system.

For the browser to truly evolve - to fulfill all of Mozilla's many goals and obligations simultaneously - it could conceivably become a kind of language platform. Think of an "inner iOS" or "inner Android" inside the operating system (I believe someone once coined the term "virtual machine"), within which the role of reading articles in a hyperlinked web falls to just one of many "apps." And here, Google search may be featured prominently in accordance with the agreement.

To that end, however, Google may already be further along with its advent of Dart, a kind of "JavaScript++" that incorporates the full scope and object support features of Java while maintaining compatibility.

However, if we're really considering HTTP as a kind of pipeline that enables end-to-end functionality, where the Web as we've come to know it is but one of many channels, then one could argue Microsoft is quite far along, at least with respect to its roadmap for Windows 8. The WinRT runtime would implant not a virtual machine subordinate to the OS, but a real machine that is a principal OS function, alongside its existing .NET Framework. Neither Google's nor Microsoft's development strategy is open to contribution from the community, like JavaScript. But the community has yet to reveal, or perhaps even coordinate, a counter-strategy - something which accelerates the evolution of the browser beyond yet another round of add-ons.

Can a browser be a reader?

HTML5 - the lingua franca du jour of the future Web - is already planning for a browser-less library, with its CSS Generated Content for Paged Media. This system envisions a Web without browsers alongside the Web with browsers, with the former being full of fine literature, long-form articles, and educational matter - a Web that speaks directly to the dreams of Dave Winer, full of links, references, and relationships.

But unlike a blog, and unlike the Web as it's currently realized in browsers, W3C envisions more of a reader. "There is nothing in Web specifications that prevent browsers from adding a page-based mode today," the draft reads. "However, most Web content is authored and styled with a continuous presentation in mind. This could change if it becomes possible to describe paged presentations in style sheets."

For anyone who presumes that conventional browsers are the Web, the W3C effectively proves with the very drafting of this module for HTML5 that they are not. It foresees a Web beyond the browser, in a new and perhaps richer context, full of new research and discovery tools that simply don't fit the old mold. While there are browser-based apps today for Amazon's Kindle and other literary contexts, the books in that environment are worlds unto themselves, without any links in or out with which Winer may escape.

For Mozilla or anyone else to get ahead of the game at this point, perhaps it may be agreeable to concede that the Web, such as it has been, is only a segment of a broader, more functional, more usable realm of information.

Discuss