09 May 2007 - Meltdown and the Maze - Toward a Real Time Geography

The defense community has long talked about "Situational Awareness" and the need for a common operating picture. While some might think this is only a matter for defense interests - there is increasing demand for a real time geography (e.g. see)  "picture" of the world that we can readily access and which reflects the current state of the world - from the condition of our roadways to the state of the climate. Such a real time geography is the objective of the GeoWeb.

To create such a real time geography requires that we bring together a great diversity of kinds of data - sourced from a similar diversity of creators - from the man in the street to governments and corporations. No one has all of the data - or even a sizable slice of it. Furthermore, a great deal of data is generated from other data - traffic sensors update real time traffic models - meteorological sensors can update parameters in road weather models - and these in turn can provide estimates of possible and safe driving speeds. The richness and variety of the data and the data sources demands data integration - but data integration that can happen when we need and for the purpose in hand. Data appropriate to hurricane response will be substantially different than that required to respond to a traffic jam or a terrorist incident - at the same time much of the data will be the same or similar.

A real time geography is thus ONLY possible if we embrace open standards for representing geographic information of all kinds. This demands that we have a common framework and the basic primitives to represent the richness of the world around us. This is the intent of Geography Markup Language (GML) and it is promise that is being delivered as GML adoption increases in more and more fields. A real time geography is also only possible if we can move data from one place to another - GML assists with the integration - but we also need the means to ask the questions to get the data. Here the OGC/ISO WFS specification plays the key role - enabling changes to be propagated from source to end user in seconds and minutes rather than the situation today when it can be hours to months.

It is also essential that we have a means to integrate ALL kinds of information and not make artificial boundaries between say sensor data and geography, or between sensor data and imagery. GML can be used in all cases as reader's of this blog are aware. Of course people will say that this imposes performance restrictions and of course this is the case today - yet this is a weak argument in the face of the fact that existing technology cannot hope to provide us a real time geography - it just was not created with that objective in mind. Performance issues will be dealt with - partly through Moores law - partly through technology improvements - but there is no question that this will be a non-issue in very short order.

Others worry about the supposed complexity of these standards - and pine for a simplicity that anyone can grasp and understand. There is no doubt that we should not embrace complexity for its own sake - however - it is also overwhelmingly clear that the variety and complexity of the issues for real time geography are NOT going to be handled by simplistic and ad hoc solutions. KISS is a guiding principle - but it must match the problems at hand.

Many people will think that real time geography can be handled merely with the arrival of GPS. Indeed GPS is a key element as it enables us to sense our position accurately - or the position of a vehicle - a vessel or an aircraft. This information is no good however without the capacity to connect this position to other objects. It does no good to know where you are, if you don't know what is there.

Of course GPS, as important as it is - is just another piece of sensor data - but one that provides, along with time, the ability to integrate the other sensor information. Sensor data is key to the GeoWeb revolution and to the vision of a real time geography. The next decade will see the emergence of hundreds of thousands (millions?) of sensors that will monitor every aspect of our planet - and which tied with more conventional notions of "geography" will make the GeoWeb an ever present reality for all of us.

To understand the GeoWeb - the real time geography - will require good "pictures" - we are visual creatures - and tools like Google Earth and its successors will be key to communicating the state of the world - and our perception of it from one to another. This will require that our GeoWeb enable the easy flow of data from sensors - individuals - professional surveyors and technicians of all kinds - satellite and aerial imagery - in situ sensors in the atmosphere and the oceans - to the tools for presentation. Again this emphasizes the need for broadly accepted standards such as KML and GML (see).

The emergence of "situational awareness" for the individual - the corporation and all levels of government is going to have profound consequences in all spheres of our lives from the daily experience to the large scale enterprise of managing our impact on the planet. The GeoWeb will be an important factor in creating that awareness.

Comments

No comments yet.