So, how did I get sidetracked? Let’s see… was it the information about Moore’s Law (processing power, i.e. number of transistors, doubles around every 18 months) coming to an end? Current processors use 45nm (45 nanometers, or billionths of a meter, is the “size” of a single transistor) technology (http://download.intel.com/products/processor/xeon/dc54kprodbrief.pdf), which allows Intel to fit 820 million transistors on a single chip. IBM is currently working on 32nm technology. The limit to shrinking transistor size is 16nm, at which point the distance between transistors can be measured by number of atoms. Based on our current knowledge, anything smaller will not work. We will reach 16nm by 2011. As transistors shrink, the chip uses more power and runs hotter.
Down a different, but similar alleyway, disk drive technology keeps shrinking – smaller, cheaper, and more capacity. Seagate just released their 1TB (terabyte), single platter disk drive – the Barracuda ES.2 SATA (http://www.seagate.com). See the accompanying table (information from “Roy Williams’ Data Powers of Ten”; University of California-Berkeley; link below) to get an idea of how much information that is. How do you manage that much data?
Which brings up yet another distraction – how do you understand the information contained within these vast amounts of bits and bytes?! There are a number of processing, visualization, and human interface technologies under exploration to solve this problem. Some examples:
- Microsoft Surface, Apple iPhone (touch screen)
- Veoh, Blinkx (video search)
- Inxight, NetOwl (entity extraction)
- TheBrain (visualization)
As you can see, it’s easy to get sidetracked! However, these explorations lead to some coherent thought on future directions of technology. Upon examining the list below, I think you’ll be able to discern how several of the items were formulated. Also, I have brought us back full circle to the original subjects at hand – SOA and Semantic Web. It’s important to remember that the technologies encompassed within the broad categories presented are continually evolving, either by intent or by happenstance (more likely in today’s environments of open source platforms and collaborative networks). In addition, there is a convergence between a number of these categories, which multiplies their importance.
- Standard, web-based architectures (Service Oriented Architectures)
- Collaboration – social networks, IM, text messaging, email, VoIP
- Geospatial Integration – geospatially enabled data holdings provide a common framework for understanding content
- Information assurance – secure information sharing, cross domain solutions, continuity of operations, information integrity
- Identity Management – single sign-on, biometrics, standard policies, portable authorization technology, tokens, Smart Cards, Java Cards
- Storage – orders of magnitude more, with smaller footprint and energy needs
- Web services/mash-ups – agile integration of functions and data to satisfy requirements and mission needs
- Knowledge discovery – information management, visualization, analytic tools, Semantic Web technologies
- Virtualization – of servers, storage, networks, and desktops
- Human Machine Interface (HMI) - widgets/gadgets, touch screen, visualization
As always, please feel free to post comments on my blog, and suggest future subjects. If you visit the site, you’ll notice some changes as I experiment with my methods of communication.
Ever curious... George
Resources: http://www.seagate.com http://www.microsoft.com/surface/
http://www.blinkx.com/ http://www.veoh.com/
http://www.apple.com/iphone http://www.thebrain.com
http://www.inxight.com http://www.netowl.com
http://download.intel.com/products/processor/xeon/dc54kprodbrief.pdf
http://www2.sims.berkeley.edu/research/projects/how-much-info/datapowers.html
