FTD: Optimizing Outcomes, Part III

By Enrique Castro-Leon

(Back to article)

When deciding how best to manage the technology development processes, there is a choice to be made between a totally centralized approach to a totally decentralized approach: A choice that is usually there, but implicit.

We assert that if this decision is not made explicitly with an objective analysis, the outcomes for the development program will be less than optimal. Yet, there is a sweet spot between the two extremes that yields optimal business results. It's called FTD, or federated technology development.

In this article, last of a three-part series (see Part I, Part II), we look at some examples where FTD has taken place. In many cases, this has been the product of visionaries who applied their intuition. The application of FTD makes this process more predictable and less dependent on a strong visionary to make it happen.

We dwell on two technologies, the Unix operating system and grid computing

FTD patterns happened with the Unix operating system, not once but twice: first during the operating system’s initial inception and, once again, during the open source software, or Linux, revolution.

Grid computing technology constitutes another example of FTD in mid-flight.

It is not driven by a single company, but by multiple players through standards-making bodies. Organizations with an understanding of the FTD dynamics behind the evolution of grid technology can translate this understanding into specific strategic plans and, hopefully, competitive advantage.


The Unix operating system was developed at Bell Labs by Ken Thompson and other researchers in 1969. Anti-trust considerations affecting Bell Labs parent company at the time, AT&T, prevented the company from building products out of this technology.

Partly because of this, the OS was made available to universities and research institutions in source form for a nominal fee.

The Unix code base, because of its relatively small size, was ported to literally dozens of different machines and became the subject of research for countless number of papers and Ph.D. theses. Each instance can be counted as an example of an FTD parallel jump.

Unix’s compact size was at the root of its competitive advantage. The OS succeeded because the action of licensing it to universities unwittingly triggered FTD dynamics.

Students in these projects eventually graduated and worked or founded companies that built new computers: Apollo, Sun Microsystems, Multiflow, Convex, Alliant, NeXT, Apple and many others.

Unix became the OS of choice and, at this point, its place in history was assured. The diversity of Unix “flavors,” often criticized, is a testament of the dynamism of concurrent technology development in myriad organizations.

A similar phenomenon, from an FTD perspective, occurred around 1991. This time initiated by a single person, Linus Torvalds, instead of a company. In case you missed it, Torvalds developed a kernel for a Unix-like system that eventually became Linux.

The FTD dynamics were triggered by the emergence of open source software. The continued development and adoption of Linux was assured by the participation of thousands of individual contributors, organizations and companies; with each instance of participation becoming an instance of an FTD touch point.


Grid computing is interesting from an FTD perspective in that it’s hard to productize in a traditional sense: You can’t go to the store and purchase a grid.

Every grid is unique and cannot be shoehorned into a line of products. This suggests that FTD dynamics can better capture the essence of the grid evolution and point more easily to useful business development strategies than a traditional product approach.

Yet, a product-centric view of the universe gives an incomplete picture of the dynamics behind grid evolution. A key factor that bets consideration—especially in an enterprise context—is that grid adoption will be driven by business processes and the IT processes supporting the business.

Once these processes are understood, it will be possible to determine which products and technology building blocks will be most appropriate. At this point it will be possible to assess desirable features and incorporate specific features into product planning.

Today, grid adoption is slow because the customer community that would potentially benefit from the technology has no awareness of how useful the technology can be. Because of this, even if “grid-ready” products were available, they would not sell because the processes that would provide the context for grid deployment are not there.

Hence a strategy to grow the grid market needs to encompass the business processes in the business segments targeted. The strategy must be holistic, integrating various sources of expertise. It must also be process based, linking various players through FTD touch points.

Grid computing will make strides to the extent the various parties are able to work independently, but cooperatively through mutually agreed standards. This captures the essence of FTD.

Enrique Castro-Leon is currently an enterprise architect and technology strategist for Intel Solution Services with 22 years at Intel. He has taught at the Oregon Graduate Institute, Portland State University, and has authored over 30 papers. Castro-Leon holds Ph.D. and M.S. degrees in Electrical Engineering and Computer Science from Purdue University.