The Big Switch

I finished reading Nicholas Carr‘s The Big Switch: Rewiring the World, from Edison to Google back in June, but am just now getting to blogging my thoughts on this book. Something about shipping software… :-)

Carr carefully recounts how the electrical industry evolved, and focused on the personalities of early central players such as Thomas Edison and Samuel Insull, as if to personify the computing industry into those “DC like” and those “AC like.”

As I read this book, my thoughts centered mostly around the nature of centralized and distributed systems, and around the human desire to control and to be emancipated.

A frequently recurring conversation I have with ECM customers, concerns centralized versus distributed systems. Often this discussion is about delivering the right façades to satisfy the needs of a particular role (e.g. supporting centralized auditing and compliance functions for systems administrators, empowering line of business users via delegated (edge) administration, including user/group management, etc.).

While I get that switch refers to electricity (i.e. a light switch), I wonder if a better metaphor would involve a dial or a set of dials. I think that software delivery is becoming more adaptive and less binary (i.e. on premise and hosted, not or).

I’m reminded of the following visual by my colleague Michael Hackney that I sometimes refer to as “dialed in” software:

There are, of course, more variables (rows) than those listed above, including whether software runs on premise or is hosted (i.e. delivered as a service). I talk with customers who require unstructured content to be managed on premise and delivered via software as a service. So, I guess I take some issue with the binary nature of a switch metaphor.

As a software architect building enterprise software for such adaptation, I think about the subtle ways our perceptions, ideas, and language change whenever we begin using a new tool (e.g. writing software for a so-called cloud environment). Consider the difference between the emphases of the printed page versus the web page:

  • Printed page – emphasizes logic, sequence, history, exposition, objectivity, detachment and discipline
  • Web page – emphasizes immediacy, simultaneity, contingency, subjectivity, disposability and speed

Here are a few more thoughts about this kind of transformation and disruption that cloud computing can yield:

  • Carr’s chapter, “The Great Unbundling,” talks to some of my thoughts in this blog [1] [2].
  • Does unbundling and re-bundling of content–a form of personalization and reuse–amplify isolation or promote knowledge sharing? Am I inclined to go beyond a particular filter even if my initial need is satisfied, or do shield (block) myself from more meaningful collaboration and thought instead (i.e. divide knowledge and magnify it according to differences, content balkanization, content polarization, etc.)?
  • What are the social consequences of pervasive content (i.e. content always at my fingertips)? (I’m already a bit frustrated with my Blackberry-slinging friends who feel compelled to thumb through incoming email despite being in face-to-face meetings with others.)
  • Advances in IT that weaken central control, inevitably are followed by a reassertion of control. That is, control… disruption… reassertion… (repeat).
  • On the Web, anonymity is just a false façade (illusion). So, does that mean the Internet can become an “integrity catalyst”?
  • Consider the value of distributed content services (i.e. bringing services to content and also content to services) in light of Eric Schmidt’s statement, “When the network becomes as fast as the processor, the computer hollows out and spreads across the network.”
  • Trust takes on special meaning where reliability, scalability, responsiveness (both software and customer service)
  • Capacity planning becomes even more important (i.e. knowing how a function/feature contributes load, establishing the right model and metadata to track demand, load and diversity factors).
  • Parallel processing and virtualization become powerful ways to deal with narrow application sets (e.g. Google) and wide application sets (e.g. Amazon), respectively.
  • Let’s say that collective intelligence is something to generally foster in a consumer or corporate environment and that something fostering collective intelligence is “green.” Then how green is your content and your content services? For example, instead of just performing calculations or executing queries, does your content (services) contribute to collective “sense making” (i.e. providing answers without knowing questions), to increased knowledge sharing (explicitly and implicitly), etc.?

Lastly, as you might imagine, there have been a number of other reviews of this book. Here are a few I found interesting:

  • Charles Fitzgerald (now with EMC via its Pi Corp. acquisition, after being a GM of platform strategy at Microsoft) – who reminds us of the power of reinvention (e.g. Edison and General Electric)
  • Ross Mayfield – who raises issues of perception and subjective value given his work with customers at Socialtext
  • Tim Bray – who does a nice job of outlining Carr’s work and validating its engaging nature (by taking the time to blog despite personal bias)
  • Chuck Hollis – who emphasizes the thought-provoking nature of Carr’s work and who leveraged his interview with Carr to discuss “how,” not “if”

Update 12/1/2008: For more of my book reviews and to see what else is in my book library (i.e. just the business-related or software-related non-fiction therein), please visit my Books page.