GIS Google Earth Technology Thoughts virtual earth

A picture is worth a thousand words ?

Not when it comes to Geographic Information I would argue…

Adena very well I think identified the massive interest in imagery demonstrated by the vendors at this years ESRI UC exhibition in her latest directions magazine editorial. Imagery is great as context to other types of spatial information, but on its own I believe it’s value is limited.

There are a lot of innovative ideas in this part of the geodata business driven both by the massive demand of the new generation of geographic exploration services from GYM and now ESRI, and from the fact that for much of North America there is no large scale topographic information otherwise available.

southampton in 3D

While technologies like Pictometry are interesting especially when combined with tools like SocketSet and OpenFlight to produce 3-D city models (thanks to John Allan and Rick Mort of BAE for the Southampton example above) , they can only provide contextual information which need expert human interpretation to generate true information let alone ‘Geospatial intelligence”

Using the example above, without access to other geographic information, can you tell.. where in the city are we ?, what is the name of the street in the foreground,? what is the address of the red building?, who “owns” this property ?

To answer these type of questions and indeed to really carry out any type of spatial analysis you need detailed feature based information and I would argue for a lot of analysis up to date information as-well.

So until more feature based information can be produced (it’s expensive !!) new tools like ArcGIS Server at 9.2 and the increasingly popular OGC WFS standard will be constrained..

Have we have invented the equivalent of the CD player, but are still producing 78-rpm mono gramophone records.

And a semantic Geoweb based on imagery.. forget it !!

Written and submitted from home, using my home 802.11 network.

3 replies on “A picture is worth a thousand words ?”


Interesting post – perhaps you should look at the site that has been built for the Australian NSW Department of Lands by AeM Group and ER Mapper: (follow link to Spatial Information eXchange).

The site shows how geographic information and image data is integrated so that each adds value to the other. Users can leverage the speed of ECWP to display imagery and supplement this with Geographic Information delivered from ESRI ArcIMS using ArcXML.

By itself temporal imagery can inform on land use change etc (assuming that you are familiar with the area on screen) but supplemented with the Geographic Information we start having the best of both worlds where users can use ArcXML/SOAP based searches to find the area of interest and have the location confirmed by the imagery on screen.

The site leverages NSW Lands services orientated architecture and will soon include channel personalisation to address the needs of other Govt agencies such as Government property, NSW Police, and the Emergency Services.

“they can only provide contextual information which need expert human interpretation to generate true information”

True, the above image couldn’t be used to answer any of your example questions, but it does provide us with an insight into the psychological and phenomenological aspects of a place. Insofar as it does this we’re able to start to answer more nebulous questions about how one experiences a place. It’s geography at the level of man. There are parallels here with these two items, which aren’t ‘traditional’ geography, but quite often enable us to make sense of a place better than a regular map:


I agree completely we have, as a community, ignored ‘sense of place” type information so far, and imagery and multimedia does help to portray this with greater effect.

We need both types of information together is my point.

Add a comment?

This site uses Akismet to reduce spam. Learn how your comment data is processed.