I have been reflecting for a couple of weeks on the significance of some of the young companies that presented at this year’s Sun Valley Conference always hosted so professionally and graciously by Allen & Co. Thingd presented its vision of the internet of things with crowd-sourced taxonomies of the attributes of these items. Square demonstrated how to extend mobile credit card payments to smaller transactions. Pandora showed its user-programmed internet radio service and the database of musical attributes that drives it.
More than ever these presentations got me thinking that all the pieces of internet infrastructure are falling into place to enable some truly fantastic services in the years to come. It is often said in tech circles that we overestimate what technology can do in the next year or two and grossly underestimate what can be achieved in the next 10 years. This results from the uniquely iterative nature of technology to build upon the foundations laid by the last innovation. Each generation stands on the shoulders of those who preceded it or more accurately each new services-oriented architecture subscribes to and builds upon a set of services introduced by the prior generation. Thus the pace of science progresses not linearly but exponentially as Ray Kurzweil has observed.
So what does this really mean in practice? For example the mobile e-payment system built by Square can rely on the location-specific service available for the iPhone or Android platforms to identify where a given transaction is taking place without having to spend the time or money to write that code itself. Similarly if you want to add mapping capability to your new service you would be well-advised not to go out and try to re-map the world from scratch; instead you would likely incorporate Google Maps or another service available to you through a programmatic interface. Almost none of these new companies build stand-alone datacenters populated with separate servers for each application; they simply subscribe to the Amazon Google or other cloud for commodity computing services. Cloud providers in turn rely on many of the technologies that preceded them virtualization software in particular.
This phenomenon of building upon the efforts of our forbearers is not unique to computing. Biochemists now routinely use gene sequencing and monoclonal antibody technologies pioneered by others and artists regularly "sample" and incorporate the works of their masters. However like the early and powerful computer language LISP in which a given line of code could be both instruction and operand digital technology permits much faster and richer iteration. So the next time you think that the digital generation has ruined our lives by speeding everything up just think of the amazing advances that you are likely to live to see. We are blessed to stand on the shoulders of giants.
You are exactly right. The cloud is becoming an innovation platform. I work at an Infrastructure as a Service company. Through the company one of our co-founders has helped give initial infrastructure and scaling support to Ruby on Rails WordPress Twitter and Gilt Groupe. And that is in addition to dozens of top Facebook apps and games. None of these companies or open-source efforts needed VC funds for the actual physical servers. They didn’t have to ask permission they just launched. In a previous life I was a derivaties systems manager. Any time we had a new system to support new ideas and new products it was a months long battle with IT to roll out the servers. So when do you think the first Secure Infrastructure as a Service Cloud for Finance is going to happen?
Hi Tom I came upon your blog when researching what working at Thomson Reuters is like and found it interesting that you made the point of thinking that it was internal members of the TR team that would read your blog. I’m willing to wager that more than half of your traffic comes from potential job seekers looking to see what their potential CEO (they are an optimistic bunch) has to say in order to get a true feel for what the company dynamic is like. From an outsider who is just starting to research Thomson Reuters I find your blog engaging welcoming and most importantly real. Thanks! Erika
Sir I totally agree with you on “we overestimate what technology can do in the next year or two and grossly underestimate what can be achieved in the next 10 years.” One day lag of providing end of day equity pricing was acceptable 15 to 20 years back now latency of real time prices are decreasing by nano seconds; I would guess entire quarter results of company would be provided in minutes in 2 to 3 years – can not think what would happen 10 years from now (may be information providers would have a copy of the report before the company and would publish it nano seconds after the company officially announces the results who knows…) Kind Regards Sudhi (Sudheendra Karanam)