Web 2.0, Web 3.0 and Beyond: Villainous Version Numbers

September 27, 2006

Just as I had come to accept the fact that Web 2.0 is an ambiguous term I am already hearing mention of Web 3.0.  In the context I have seen it so far, Web 3.0 is being used to refer to “Web 2.0 in the enterprise”.  As a side note, it would seem that an enterprise implementation of Web 2.0 concepts would be more aptly named “W2EE” (i.e. Web 2.0 Enterprise Edition), instead of increasing the version number.  This approach seemed to work for the Java guys for a while with J2EE.  But I digress.

First off, let’s take a look at things that we used version numbers for (before the practice of starting to add version numbers to concepts):
  1. Products:  Examples would include Windows 3.1or Oracle 10g

  1. Platforms:  Similar to the above, but examples might include .Net 2.0

  1. Standards:  Such as SQL92 or DOM1

  1. Documents:  More of an internal thing, but you could have “Version 3 of our standard sales agreement”.

You will find a common pattern in almost all prior uses of a version number.  Version numbers are generally a “short-hand” to describe something that is reasonably well delineated and distinct.  Oracle 8i represents some combination of things which is different from Oracle 10g.  Makes a fair amount of sense.  By labeling something with a version number we indicate that we have frozen it in time (to some degree) and decided to call it something.  As a community, we have found this useful.  It helps to talk about things that are changing in terms of discrete instances with a label.  No big surprises here.

Now, when we start looking at Web 2.0 (and now Web 3.0, Web 3.1 and related terms), things get a lot trickier.  Since Web 2.0 (where the madness started) is really just a set of concepts and broad technologies, things start to get difficult.  Now, one can argue that a label is a label is a label and there’s no reason to get upset because this particular label (Web 2.0) just happens to look like a version number.  But, I think some amount of semantic discipline is beneficial. 

Web 2.0 seems “villainous” to me for a number of reasons:
  1. Ambiguous:  If I told you a company or product was “Web 2.0” this may or may not mean anything to you (and likely means something different to different people).

  1. Misleading Granularity:  The fact that it’s two point zero (i.e. 2.0) seems to indicate a degree of specificity that is just not there.  It’s not just “Web2”, it’s Web two-point-oh.  This invites “incremental upgrades” like Web 2.1, Web 2.5, etc.

  1. Cascading Effect:  Now that we have come to accept attaching meaningless version numbers to amorphous concepts is ok, we are starting to use it all over the place.  In many cases, this is tongue-in-cheek (like my use of “Hindsight 2.0” in a prior article), but in some cases, people are actually expecting to convey something.  I’m also guilty of using (if not coining) the term “Small Business 2.0, our sister internet marketing blog”)

I think this whole “versioning of concepts” thing is going to be something we look back on with amusement in the years to come.  Similar to how we believed attaching “.com” to a company name somehow made it worth more back in the 1990s.  

Note:  This article is compatible with Web 3.0 and backward compatible with all versions back to and including Web 2.1, though attempting to read it with software on Web 2.0 may cause unanticipated behavior.  You are entitled to a free upgrade of the article for all 3.x “point” releases, but I reserve the right to charge an upgrade fee for Web 4.0 and beyond.

Topics: web20 web30

Written by Dharmesh Shah


Let's Connect

And, you can find me on Google+, Twitter, and Linkedin.