• Open Data

    The White House announced on Monday, August 11th 2014, that it is formally launching a new U.S. Digital Service and that it has hired Mikey Dickerson to lead it. Mickey Dickerson is an engineer widely credited with playing a central role in salvaging HealthCare.gov. The idea behind the USDS is institutionalizing the approach that saved the US health care website.

  • Open Data

    Let’s Talk About Open Government First

    President Barack Obama, on his first day in office in 2009, signed an executive order stating that all government information that did not have to be kept secret for security or privacy reasons should be made public. The administration also launched the Open Data Initiative to publish government data and the data.gov website to distribute the data.

    My Administration is committed to creating an unprecedented level of openness in Government. We will work together to ensure the public trust and establish a system of transparency, public participation, and collaboration. Openness will strengthen our democracy and promote efficiency and effectiveness in Government.

    - President Barack Obama, January 21, 2009

  • CoreOS

    I have been playing around with CoreOS to get a sense of how everything works. The vision of this project is incredible.

    CoreOS describes itself as “a new Linux distribution that has been re-architected to provide features needed to run modern infrastructure stacks. The strategies and architectures that influence CoreOS allow companies like Google, Facebook and Twitter to run their services at scale with high resilience.”

    CoreOS displaces hypervisors and machine virtualization in favor of Docker and Linux containers. CoreOS uses Linux containers to manage your services at a high level of abstraction. A single service’s code and all dependencies are packaged within a container that can be run on one or many CoreOS machines.

    Clustering works across platforms, meaning there is no cloud vendor lock-in. For example, CoreOS runs on Amazon EC2, Rackspace, QEMU/KVM, VMware and OpenStack and your own hardware. Running a single CoreOS cluster on multiple different clouds or cloud + bare metal is supported and encouraged. This lack of lock-in is the reason why I have supported OpenStack and CoreOS takes this even further.

  • Docker

    Three driving forces of infrastructure efficiency in a large, distributed environment were:

    • Data center consolidation (KPI: How many DCs do you have?)
    • Host virtualization (KPI: what percentage of your servers are virtual?)
    • “Cloud” in its various forms (KPI: How many servers are you still running?)

    Assuming you have been aggresive you may not have any data centers left. You may be buying services rather than servers, and/or all of your remaining machines might be virtualized and hosted in someone else’s data center. Now what?

  • Google Enterprise Search

    One of the challenges in a large distributed enterprise is “finding stuff”. Sure we call it collaboration, knowledge management, or any number of other sophisticated terms but often it boils down to just finding stuff.

    To compensate the organization reacts by saving multiple copies and versions of a file in many different locations: email, shared file systems, intranets, etc. Of course this compounds the issue and drives the desire to save “my copy” to “somewhere where I can find it”.