Small pieces loosely joined: A useful way of looking at the web

The new century brought a lot of change to the young Internet, both good and bad. The unbridled enthusiasm of the 1990s that brought us the sock puppet, along with a number of other stupid ideas ultimately crashed and burned, and led to the bursting of dot-com bubble also brought in a more active way for users to interact the web.

Tech uber-publisher Tim O’Reilly captured this new potential in a 2005 article[1] where he declared the web had evolved to Web 2.0 and described a new level of user experience. Rather than passively sit back and absorb a series of static pages, users would now interact with dynamic elements that can be constantly updated and actively provide the tailored experience we have come to expect more than a decade later.

O’Reilly’s Web 2.0 pronouncement sits firmly on the work of a number of thinkers, including a researcher at Harvard’s Berkman Center for Internet and Society named David Weinberger. Trained as a philosopher, Weinberger has had a number of careers, including a stint as a comedy writer for Woody Allen and Internet adviser to John Dean’s 2004 bid for the presidency- praised for its innovative use of the web, but failing in a loud hoot. My how times have changed.

In 2002, Weinberger wrote a short but influential book entitled Small Pieces Loosely Joined: a unified theory of the web [2] which profoundly changed the way I understood the World Wide Web. To be fair, the book is somewhat dated at this point. I’ve tried assigning as student reading with poor results and perhaps the best takeaway is the title itself, which provides a profound insight into thinking about the web.

Weinberger’s big idea is this: Rather than think of a website as a monolithic entity, existing on some server, somewhere, and centrally controlled, a website is a collection of resources, a loose collection of resources — “many small pieces loosely joined.”

No physicality

There is no physical space to the web, so these resources can exist anywhere in the world, on any server, and it takes no more time to access them from Kansas than it does from the server in the rack next door. In fact, linking to external resources can actually improve access time, because your server does not need to provide 100% of the bandwidth. We routinely use services such as Flickr, YouTube, and Vimeo to serve images and movies that allows us to serve massive numbers of people with modest bandwidth resources.

Dynamic content

By relying on the networked nature of the web, you can easily include resources that change, such as news, weather, and a number of very particular data sources that are freely provided as web-resources. A web-resource is kind of communication directly between web pages.

This cooperation occurs when the user-facing web page sends a message, called a web service request, to another web page for some information, such as the current weather. The “invisible” page responds by sending that information back to the user-facing page, which then displays it.

When the user clicks on a page that contains a call to a web service, the page will begin to load from the page’s server. When the server reaches the part of the page where it wants to find the current weather, the server will ask a second server that contains a web service. This automated dance occurs because the makers of both pages agreed on a common format to facilitate this interaction. The page was “receptive when asked.”

SHIVA visualization tool

As a proof of concept to this strategy, at UVa we built a visualization tool, the SHANTI Interactive Visualization App (SHIVA) [3] that successfully leveraged a number of these external resources to create a tool that took data stored on Google Docs spreadsheets, images from Flickr, videos from YouTube and Vimeo, online maps, charts, graphs, word clouds, and timelines that makes it easy to create interactive graphical elements. These web-based elements can be in turn linked to other websites. SHIVA is freely available for anyone to use.

The true nature of the web needs to be embraced when designing modern web applications and to take advantage of this highly networked environment as an active player in that network, not just a deliverer of static content.

About Bill Ferster

Bill Ferster is a research professor at the University of Virginia and a technology consultant for organizations using web-applications for ed-tech, data visualization, and digital media. He is the author of Sage on the Screen (2016, Johns Hopkins), Teaching Machines (2014, Johns Hopkins), and Interactive Visualization (2012, MIT Press), and has founded a number of high-technology startups in past lives. For more information, see

[1] What Is Web 2.0? Design Patterns and Business Models for the Next Generation of Software:

[2] Small Pieces Loosely Joined website:

[3] SHIVA website: