In agricultural conservation a “Greenfield” is considered an undeveloped, uncontaminated piece of property; in technology solutions that require "game changing" technological and political investments to be successful are often called “Greenfield technologies”.
A flattering way of thinking of Greenfield technologies is to think of them as revolutionary technologies vs. evolutionary ones; I typically have a problem with this model in that it’s been quite some time since I have seen any technology that feels truly revolutionary.
In any event, what I wanted to talk about today was how the chains of our past successes in technology make it difficult for us to make these revolutionary changes in the future.
There are many examples of this, take IPv6 as an example we are so entrenched in IPv4 technology and the associated deployments that despite the fact that for the last decade IPv6 standards have been completed and support for them is in every major operating system we have made only minor progress to this new technology.
Then there is the migration to digital television broadcasts and High Definition, these concepts/technologies were first developed in 1969, yet products based on them did not start to become available until the late 1990’s and here we are in 2007 and we have still not transitioned off of the legacy analog counterparts.
There are lots of reasons these transitions take a long time, one of the most significant being that the transition technologies that enable the broad deployment of these technologies (for example Teredo in IPv6 or the Digital to Analog Converter box in the case of High Definition) often come very late in the deployment cycle; this can be attributed to the fact that its common for those who develop a core technology to stop there (this reminds me of the line in Real Genius where the young scientist says “Let the engineers figure out a use for it”). In the end this usually attributable to the inability to identify a business model that is immediately sellable to justify the investment in these long tail investments.
Interesting side effects result; for example consider TLS and PKI, this technology clearly has been extremely successful yet I suspect less than 10% (I can’t find any numbers right now but this has to be a safe bet!) of the internet accessible websites are protected with these technologies; the rational for this is long and complex but the implication is that the majority of the websites out there do not get the protections these technologies offer as a result alternate technologies evolve in an attempt to address the gap.
One such example would be DNSSec, this is a technology that is just another form of PKI; instead of having a separate entity perform the role of certification it makes the DNS server that entity; there are some measurable benefits to this approach, for example there are fewer DNS servers than there are web servers so although DNSSEC doesn’t offer client to server confidentiality and integrity it does offer name authenticity (arguably the most valuable service of TLS PKI) in a cost effective manner.
While at Blackhat this year I had a opportunity to see a presentation by Kenneth Geers on the “Ten most Orwellian computer networks”, one common trend that was apparent was is that these countries leverage their control of their “network” and its DNS to control and track the information the residents have access too; removing this control (right or not) has the possibility to be a "deal breaker" for the ubiquitous deployment of DNSSec.
In the end I think the largest issue DNSSec will have is its political ones, mostly rooted in Sovereignty and not the technical ones; this reminds me of a Voltaire quote: “It is difficult to free fools from the chains they revere.”
As you can see from the handful of examples I have provided the migration from technologies deployed successfully in the past to newer more modern equivalents (with their own strengths and improvements) clearly becomes more complicated once the prior technologies become entrenched, this entrenchment can literally take decades to overcome and in many cases cannot be.
This begs the question of if we will see more “game changing” technologies in the future, I would expect we will but the rate in which the rate in which they appear will slow after all once this pattern becomes painfully obvious a rational person would question their involvement in efforts that are likely to never see the light of day; I do hope however this is just me being a pessimist
.