In this month's Wired, Clive Thompson has the best single piece of analysis on the netbook phenomenon. While he makes the standard argument that this anemic but affordable low-end disruptor took the big PC vendors by surprise and threatens to eliminate the last bastion of attractive profits in the commodity PC business, the feature-laden laptop, where the piece really shines is in the way that he traces the confluence of factors that led a Taiwanese company, Asustek, to become a powerhouse in launching this new category.

For Thomspon it was the combination of the design work started at OLPC, the design muscle that Asustek had built up over a decade of making, and eventually designing, PCs for the big American companies, and the shift in computer use to the web browser for most everyday computing tasks that allowed the EeePC to explode on to the scene. It is worth considering the power of these overseas ODMs (as opposed to an Original Equipment Manafucturer, an Original Design Manufacturer can play a big role in the design of a product top-to-bottom).

Thompson's money quote:

What Asustek proved is that the companies with real leverage are the ones that actually make desirable products. The Taiwanese laptop builders possess the atom-hacking smarts that once defined America but which have atrophied here along with our industrial base.

I can attest to the fact that every product we make these days has big contributions from the ODM even very early in the planning phases. For the most part this is a good thing, after all specialization and factor inputs tend to lead industries in this direction; however, as a non-hardware person, I've been surprised at how some design decisions seem to turn significantly on what the ODM can or can not do for a given project. More often than not American companies present specifications in the form of Word docs and not real engineering to companies like Asustek, Foxcom, and Quanta. It makes one wonder whether we have not opened ourselves up to a dependency that might be unhealthy for the types of R&D engineering innovations that made companies like HP in the first place.

Meanwhile the best analogy in the article is of the standard laptop to the SUV— a computer that is far too powerful and sold for the promise that one might actually do some simulated annealing work on it someday. And just like no momtank ever goes off road, most computer users (nongamers and nonprogrammers) barely venture off of the communications paved highway.

That said, it may just be that we are currently in a phase of personal computing where we are not delivering the software applications around experiences like immersive video or voice recognition that could really exercise local computing power in the client. Maybe, but as the power in a smartphone approaches that of a standard Atom netbook (and as these in turn start to adopt ARM cores for battery reasons), we might have just painted ourselves into a corner where those applications will either reside on server farms, or worse yet, not get invented at all.