When talking about the “web” what are we referring to? For most people it’s what can be experienced through their web browser including HTML, audio and video streaming, Flash-based animation, or rich Internet Application (RIA) interfaces. The key to this perspective is the web browser, which is viewed as essential for experiencing any type of content available via a hyperlink on the web. In reality, however, that definition doesn’t seem to encompass all of our uses of the web or, to be more precise, it doesn’t describe the full set of technologies that are accessible through the web.

If you download an application from the web and install it on your desktop, is it still a part of the web? Some people would say no; that once you take the application out the browser it’s no longer a part of the web. This raises an interesting set of questions: At what point is content “outside” the web browser? For example, if the presentation layer of an operating system includes the capabilities of a web browser, is the presentation layer a web browser or the desktop or both? If a web browser plug-in allows you to view web content outside the confines of the browser, is the content still a part of the Web? The truth is that as the desktop evolves, the border where the web browser ends and the desktop begins (or vice versa) is blurring. It’s becoming more difficult to determine when you are using a browser or using something else to peruse the web.

RESTafarians often apply a pretty narrow definition to the Web, requiring that anything that is part of the web be accessible as a URL and that those resource contain URLs to connect to other resources on the web. By this definition, email, although accessible via the web, may not be a part of the web depending on whether you use a web-based or a desktop-based e-mail client. For example, GMail is web-based so that individual e-mails can be referenced via URLs and can contain URLs. E-mail accessed through a desktop-based e-mail client, such as Outlook or Apple’s Mail application, do not expose individual e-mail messages through URLs, so are these desktop applications part of the web? My wife, a true computer neophyte who uses Yahoo! Mail, would tell you that all e-mail is a part of the web. My pragmatic but sometimes fundamentalist RESTafarian friends would tell you that Outlook and Apple Mail are not a part of the web.

Just as e-mail may be considered a part of the web or not, so to can rich Internet applications. A RIA application, while accessible via the web, may not make any of its windows, pages, dialogs, or information accessible via URLs. In this case some people will say it's still a part of the web (my wife) and others will say it is not (my RESTafarian friends). Can they both be right?

The discrepancy between the definition of the web as applied by RESTafarians and people like my wife (i.e., the general public) is pretty common. For example, most people think of butterflies and spiders as bugs, but entomologists consider them to be entirely different creatures whose only commonality is that they are both relatively small. A butterfly is a specific type of insect. A spider is not an insect; it's an arachnid. Well, in terms of modern science the entomologists are right. In terms of populist opinions – they’re all just bugs. Is there a real truth? Not really, both constituencies (i.e., the general public and entomologists) are simply applying their own rules of categorization on a commonly used term. In other words, the way in which we categorize things is always artificially based on a handful of characteristics that exclude other characteristics that don’t fit our mental model. The definition of a “bug,” just like the definition of the “web,” depends on who is making the definition.

The point is that the web, just like everything else, is whatever we define it to be and is different things to different people. This is an important perspective to maintain in all endeavors but is particularly useful when it comes to technologies. Our definition of the web 15 years ago is very different from our definition today – at least for the general public. Today there seems to be two prevailing views of the web.

For RESTafarians the web is a REST-based architecture defined by the combined use of a set of network standards including HTML, HTTP, URI, and MIME. To the rest of the world, the web is the Internet. It’s anything you access from your computer (and increasingly your mobile phone and game consoles) that resides on some other computer. The world at large uses the web and Internet interchangeably, which is wrong to all but about 1% of the population that thinks they know the difference. So which is the truth? There is no truth; there is only the interpretation of terms. The web is whatever you think it is and nothing more or less.

A great deal of energy has been wasted attempting to define terms like “Web,” “Web Services,” “Web 2.0,” “Internet,” and “RIA”. This is understandable as we want people to understand what we are talking about when we are referencing a specific piece of technology. Adobe Flex is a RIA; AJAX is not. Wikipedia is Web 2.0; eBay is not. SecondLife is a Massive Multiplayer Game; the wii is not. Agree. Disagree. It doesn’t really matter. All that matters is which technologies people actually use. There is no absolute truth in defining the web. Remember this the next time you try to mentally categorize a technology that claims some type of webiness – the web is in the eye of the beholder.

This column appears exclusively at SYS-CON.com. Copyright © 2008 Richard Monson-Haefel.
(This copyright notice supersedes the one auto-generated at the foot of this page.)