August 14 2008 / by Mielle Sullivan
Category: Technology Year: General Rating: 6 Hot
“Welcome to the future, at least one possible future anyway,” announces Mozilla Labs. Along with designers from Adaptive Path, Mozilla has released Aurora—a proposal for the visual and design components of what could be the future of, not only web browsing, but of the computing experience in general. In three dramatized videos, users retrieve, manipulate and utilize data with remarkable ease. Devices and computers communicate fluidly with the web and each other, pulling up relevant data quickly to help make plans. They even identify objects in the real world. At times it is hard to tell where the computer ends and the web begins. But is this really the future of computing? How can this all be made possible?
The Aurora concept browser differs from web browsers of today in three obvious ways. First, it incorporates all applications not just those that are connected to the web and thus replaces the desktop. Second, it attempts to make the experience primarily visual rather than textual. Finally, it takes full advantage of what the Semantic Web will hopefully have to offer.
After a few minutes of watching the concept video, you realize that Aurora bears little resemblance to today’s web browsers. For one thing, there is no distinction between applications and websites and there is no time when the web is accessed. Rather, the whole environment is constantly interacting with the web. Strictly speaking, the Aurora concept browser is not a web browser. It is a graphical user interface which anticipates that the web will be THE application and resource of future computing. All applications a computer may have, if they are not connected to the web, will serve only to enhance and facilitate the web experience. In other words, in the future, your desktop, your operating system, all your programs, and your web browser will merge into one user interface that is built around and inside the web.
make computing more “natural,” Aurora proposes to visually represent
every aspect of the computing experience. Since the beginning of
computing, particularly consumer computing, designers and engineers
have sought to make interacting with a computer more like interactions
in the physical world. Microsoft Windows and Apple Mac OS owe much of their
success to the graphical user interfaces (or GUIs) they used which
eliminated the need to learn command codes. Representing programs,
documents etc. as objects on a virtual two dimensional plane, proved a
much easier way for users to understand, interact with and manipulate
data. Today, our experience with web browsers is still largely text
based. If we want to go to a website we have visited before, we have to
remember the URL or find it in our
bookmarks. If we want to find a new website we type keywords into a
search engine to pull up websites that use the words (but not
necessarily the concepts) we are looking for.
everything (websites, applications, people) are represented as objects.
Even the use history of these objects is represented in a Z access of
time extending away from the user in the three-dimensional virtual
environment. “When we stop using a page, or interacting with a person,”
the concept video states, “they begin to drift steadily away from us.”
Thus, the more time has passed since we have used an object, the
smaller and farther away it appears.
All objects can be
dragged and dropped anywhere in the environment, even elements within
websites—such as data tables—can he moved and manipulated like objects.
Objects with strong associations are grouped together in clusters. A
given object can be a part of several different clusters at once. For
instance, let’s say if you are planning a trip to Japan. All the
websites relating to the travel arrangements, your Japanese language
learning program, and your calendar would be grouped together in a
cluster. If want to access the website for your hotel in Tokyo, you
simply move your cursor over the cluster of objects relating to your
Japan trip, find the object representing the website of your hotel in
Tokyo and click on it. Of course, you could still search for anything
using a text search bar, as demonstrated in the first video.
would make all this ease of data manipulation possible? Well, beyond
the design tools presented in Aurora— a fully realized Semantic Web is
necessary. The dream of the Aurora Concept Browser demonstrated in the
videos is essentially dramatic enactments (with a few nifty visual
elements thrown in) of the Semantic Web dream put forth by web creator
Tim Berners-Lee in his 2001 article for Scientific American.
In that article, Berners-Lee described wireless web “agents” which
interact seamlessly with the web, each other and other devices to help
people make plans and decisions without having to sort through bunches
of irrelevant information returned by search engines, or perform
multiple searches. The agents do all the drudgery, leaving the people
only to approve the final recommendations or refine the parameters of
the task. The same situation occurs in the second and third videos for
Aurora. All of this quick, uninterrupted communication is enabled by
the same Semantic Web concepts that allow data to be dragged, dropped,
manipulated and grouped together meaningfully as they were in the first
Inherent in the dream of the Semantic Web,
is the idea that computers will be able to distinguish between types of
data and identify people, places etc. as the unique things they are
with distinguishing relationships and properties. If a computer can
indeed make these distinctions, then a whole new level of
interoperability will be achieved. Devices and their programs will be
able to understand the web, people and one another which more ease.
Aurora proposes one possible graphic user interface not just for your
whole computer but for all your mobile devices as well—for the whole
So how will the Semantic Web realize its dream? Well, no one is completely sure yet, but there are a number of theories and practices
already being put in place in different ways. Most likely it will mean
implementing different, hopefully complimentary, forms of meta-data
that will tell a computer what a piece of data is and what its
relationships are. RDF
is the most logically sound and universally useful form of meta-data;
it can communicate hierarchies and relationships, but it is also the
most complex and right now requires the most work to input. Microformats are simpler, but contain less information. A top-down approach for inserting some kind of meta-data are programs such as Reuters Calais. These programs go through documents and essentially insert tags for known people,
places or things it finds. It is difficult to say how well any of these
methods will work or exactly how useful the end product will be
because, at present, the amount of data on the web that uses meta-data
is comparatively tiny—though it is growing.
meta-data increasing rapidly, we are still a long way from effortless
dream of Tim Berners-Lee and Aurora. Mozilla envisions Aurora as being
open source like Firefox, but creating such an interface would be a
massive project and one wonders if the open source community is really
best equipped to handle such a task. On the other hand, building and
building-upon such an interface is probably best achieved through a lot
of collaboration and such collaboration would mean a great deal of
transparency anyway. The open source versus proprietary debate may get
its best showdown to date if anything like Aurora is attempted. There
are also, of course, security concerns. With all of this communication
between devices and the web and computers, we’d better be sure the
requests for data are coming from the right place. Universal standards
and protocols also need to be established. The WC3
is already working to establish such protocols, but true universality
has not arrived. None of these are small stumbling blocks towards a
fully realized Semantic Web.
Even Berners-Lee thinks, perhaps, his original vision of the semantic web was a bit too “sci-fi” . Probably the real-life interface and experience of the Semantic Web will not be quite so fluid and easy as the Aurora concept browser videos display. But with so many advantages to achieving some level of semantization and so many different companies and organizations working towards that aim, I think a large degree of semantization will be achieved in the next few years, and will continue to grow beyond. The end result could be something much more important than a cool, easy to use interface. Some of the thinkers and participators in the semantization process like Nova Spivak (Radar Networks, Twine) and Kevin Kelly (Senior Maverick at Wired magazine) believe that once semantization is achieved the web will be a giant database and the inter-connectivity and the millions of users of such a database could function like a giant mind regardless of its interface. A revolutionary future for computing.