General appearance of Web Forager Click on the image to enlarge it Source. Atlas of Cyberspaces. See the article by Card et al for more information and images

In the previous issue we spoke about the information foraging theory. One of the first exercises of applying this theory in practice, although only in an experimental way, was WebForager, software that allows you to structure the foraging workspace in the Web. 

It’s well known that the limiting factor when surfing Internet in search of information is time. It’s also known, from empirical studies, the tendency for users to interact repeatedly with small sets of information, something usually referred to as "locality of reference".

For this reason, according to Stuart Card, George G. Robertson, and William York* creators of the Web Forager, users tend to structure their information workspace in a way that somehow optimises the cost structure of access to information. 

In its most elementary form the user has a certain amount of “proximal” information that he/she can find at a very low cost, for example the papers we have in our desktop. Another, larger, set of information is also used at an intermediate cost of retrieval, like the one you store in the drawers of your desk and, finally, large spaces of “distal” information where the cost of finding what we are looking for is much higher, like in the general archive or the library of our company.

The Web, nevertheless, doesn’t exhibit an information structure that easily supports the creation of these “natural” foraging spaces. To overcome this limitation Card proposes two movements: 

• To overcome the web page as interaction unit in favour of an entity with a higher level of integration. This environment is called WebBook™
  
  
• To move from a workspace where you have only one entity at the same time to another with multiple entities including, among others, the ones of the former movement. This workspace gets the name of WebForager™

WebBook™, as its name indicates, uses the book metaphor in order to aggregate web pages into a structure that is visualised in 3D with the appearance of a book. Each page is a web page that is shown the traditional way, but for the fact that you can flip it to see the next or the previous page. Links to other pages of the WebBook are coded in a different colour than those that point to external pages (red and blue respectively).

Should we click a link to another page of the same book these begin to flip until you get to the selected page. If they are located in another WebBook, the current one is closed and the one containing the referenced page is open. If the page lies outside the workspace it is presented as a standalone page floating in the WebForager environment. To flip a page you just click on the present page. The closer the page to the margin the more pages are flipped at once. This allows you to move easily from the beginning to the end of the book.

Web Book.  To the left you can see the appearance of it when open. To the right one of them is shown closed. Notice the tabs that in their lower part, that allow us to perform operations on the book or leave our own bookmarks. Click on the images to enlarge them Source. Parts of the general image available at Atlas of Cyberspaces.

The pages of the WebBook can hold user placed bookmarks, as if it were a real world book. It’s also possible to flick through the pages to get a quick idea of the information they contain. A WebBook can be stored on a shelf or bookcase of WebForager. When you ask for it again it opens at the last page visited. 

WebBook also supports a focus + context technique in which you can see all the pages of the book at the same time, although the one that is in our focus has a larger size than the others.

WebBooks are built dynamically so that the algorithm can read the bookmarks of your favourite browser and build the corresponding book. It is also possible to follow all the links of a particular page building a book with them. You can also create topic books about a specific subject or a WebBook with the pages that come out of a search engine enquiry. 

According to Card the information is the same but its accessibility and availability is much greater.

WebForager is a 3D space where you can place WebBooks, shelves and bookcases and/or just individual web pages, as if it were a real desktop. Individual pages can be grabbed and incorporated into WebBooks at the user’s will.

This space is organised hierarchically into three levels that respond to different interaction intensities.

• Attention Focus. The page or open book with which the interaction is immediate, without the need for any intermediate action.
  
  
• Immediate storage space. The “air” and desktop where you can leave the pages or entire books we are using floating in the air or leaning on the desktop. To access them you have to close the focus and select the desired element. Books and pages can be left at different heights and distances.
  
  
• Tertiary storage space. Shelves and bookcases that can contain large amounts of books or pages that are seen as book covers.

Web Forager. Hierarchical environment where you can find WebBooks, standalone web pages and bookcases with more WebBooks. Notice the three levels of interaction, the activo book (attention focus) the air and desktop (immediate storage space) and the bookcases (tertiary storage space). Click on the image to enlarge it Source. Atlas of Cyberspaces. See the article by Card et al for more information and images

As we can see this structure pursues the information cost structure referred to at the beginning of the article.

WebForager and WebBook appeared in 1996 in the Xerox PARC labs as a first approach to the information foraging theory. Three-dimensional metaphors like the one we have seen in the book or other similar ones haven’t yielded, up to now, better results than other more abstract metaphors. Possibly due to the fact that doing the same things in the virtual world that we do in the real world maybe doesn’t give more added-value to our insight.

In the next issue we’ll see some examples of this.

* See also the article “The WebBook and the WebForager: An Information Workspace for the World Wide Web” by Stuart K. Card, George G. Robertson, and William York.

Links of this issue: