444A national intelligence network using K-Logs
- May 3, 2004Dear K-Loggers,
As some of you may know, I have been working on national security issues for a while (with a particular emphasis on networks and social technology). I wrote an article today on building a national intelligence network that may be of interest to you since it involves using K-Logs -- many of the same problems we face building a nation intelligence network are very similar to those faced by large companies and organizations.
Here it is:
Non-state terrorist networks enjoy many advantages over the traditional hierarchies of nation-states. They learn, share, innovate, and survive disruption better than hierarchies. This strongly implies that best way to defeat these terrorist networks is to adopt a network architecture ourselves. The best approach to begin this process of transformation is to establish a nation-wide intelligence network. The advantages of a national intelligence network are real. For example: almost all of the information needed to identify and stop 9/11, the DC sniper, and the first World Trade Center attack was present in the nation's intelligence/information system prior to the occurance of the events. It was the current network's flaws: stove piping, one-way flows, gaps, over centralization, security restrictions, interfaces, and multiple technologies that prevented the timely use of this information. To prevent future attacks, the Markle Foundation has prepared a 2003 report, "Creating a Trusted Network for Homeland Security" that recommends building a dynamic and decentralized national intelligence network.
See: http://www.markletaskforce.org/Report2_Full_Report.pdf (PDF)
Their proposed national intelligence network (dubbed SHARE -- Systemwide Homeland Analysis and Resource Exchange) includes the following:
A decentralized architecture.
Real-time information flow.
Ad hoc connections/information flow.
Scalable systems (to hundreds of thousands of participants -- 14 federal intelligence agencies, state and local law enforcement, hospitals, fire departments, and private industry).
Provides multiple nodes of analysis.
Many-node information sharing (within limits of security access level).
Easy directory development.
Open architecture (XML).
Able to handle multiple data types (docs, spreadsheets, multimedia, etc.).
What's needed: A national Weblog/Wiki network
These requirements beg the question: what system would enable all of this functionality? The simple and powerful answer is (it is the only system I know of that can accomplish this): a weblog/wiki network with fast search services (along the lines of what we have been talking about on K-logs for the last three years). A network of this type will:
A decentralized architecture for publishing and consuming information (via Really Simple Syndication -- RSS).
Real-time information flow through the use of a search engine (ie. Google appliances) and prompted indexing (through weblog pings to the search engine when new posts are published).
Ad hoc connections and information flow through RSS subscriptions (an RSS aggregator ala Bloglines, NewsGator, Radio, or Technorati).
Scalable systems -- already demonstrated via the millions of weblogs on the Web today.
Easily filterable through Intranet search and through applications built on an expanded Google API (see: http://www.google.com/apis/ ).
Provides multiple nodes of analysis (filtering and independent analysis) through analyst weblogs. The operation of the weblog world demonstrates this everyday.
Provides many node information sharing through easy to use Web interface standards (which implies zero training for users).
Easy directory development through a Wiki (see the Wikipedia http://en.wikipedia.org/wiki/Wiki for an example) and OPML (Outline Processor Markup Language -- based on XML -- see: http://www.opml.org/ )
Open architecture using XML. Weblogs use XML. Both RSS and OPML are XML based.
Able to handle multiple data types. Weblogs allow the publishing and the reciept (automated if necessary through RSS enclosures) of multiple data types. Also, search engines like Google allow these data types to be easily searched once published.
Let's get moving!
A national intelligence system of this type can scale to a global level, linking up open societies everywhere there is an Internet connection. The only requirement necessary to start this is the decision to start -- the expense would be negligible (tens of millions), training would be almost unnecessary (Web standards are used -- almost everyone knows how to type an e-mail or search using Google), and time needed to impliment short (less than a year from the word go to launch of the basic system). If you would like to discuss this, please contact me directly.
[Non-text portions of this message have been removed]
- Next post in topic >>