On Monday I travelled to Cambridge University to attend the ‘Google Earth and other geobrowsing tools in the environmental sciences workshop‘, organised by NIEeS and ReSC. The workshop consisted of one day of presentations and one day of practical sessions. I only attended the first day, which entailed an early morning easyJet flight to Stansted, and then a late evening flight back to Edinburgh.
The most notable presentations came from Jason Chuck and Matthew Trewhella, two of the top Google Earth/Maps techies (who didn’t seem to understand what map projections were when asked about them), Maurizio de Felice (showcasing NASA World Wind), David Maguire, an ESRI evangelist promoting ArcExplorer, and Martin Daly (giving an overview of OGC activity).
What was most obvious to me about ArcExplorer (but which was not remarked upon by anyone) was how much it is so blatantly ripped off from Google Earth. A recurring theme throughout nearly all the presentations was how extremely useful Google Earth is for using as a platform for building custom map-based applications, and how much the academic community loves it, especially as it is free. ArcExplorer is free to download and use as well, although ESRI is trying to maintain an advantage by offering useful map data to use in ArcExplorer for subscribed customers.
Maguire did his best to beat off the Google Earth/Maps challenge by asking the assembled audience (of mostly environmental scientists) during the panel discussion at the end of the day if they had seen anything that could be called real GIS analysis rather than just fancy animated globes, but no-one seemed to side with him. In some ways he was missing the point, in that academic researchers see Google Earth/Maps as a lightweight way to do three things with data that have a geographic element:
- Visualisation (a ‘quick look’ at data, sometimes even real-time data from remote sensors).
- Initial error-checking of the data (data points that are in the wrong geographic area are easily spotted).
- Dissemination and outreach (easily exposing what you’ve done to the whole world by putting it on a website).
The fact that Google Earth/Maps is not a serious GIS tool was not seen as a problem by anyone (except perhaps by ESRI), as serious number-crunching of data is already done by other well-established methods – what Google Earth/Maps offers is a low-barrier way to implement the three things mentioned above, which can go a long way to making an academic researcher’s job easier.
Other notable things that the day brought up were:
- Google KML and OGC GML are soon to converge.
- Google Maps and Google Earth will in the future be more compatible, with both using KML as the underlying data syntax.
- Environmental scientists are very interested in the 4 dimensions of data (x,y,z,t) and they are driving applications like Google Earth to support the visualisation of this.
- REST-style programming APIs (e.g. the Google Maps API) have been much more widely taken up (at least in academia) than formal Web Service (SOAP/WSDL) APIs.
- Due to the success of Google Maps, AJAX-style internet communication is belatedly being seen by organisations like the OGC as something that they should address.
- Global internet-based geographic standardisation efforts (from the likes of ISO and OGC) have some way to go – despite WMS being widely used (and GeoRSS and GML less so perhaps), research data applications are still being developed with local needs (i.e. the interested community of environmental scientists such as meteorologists or oceanographers) in mind, rather than cross-discipline national or international requirements being considered. SDIs as a reality are not here yet.
- One NASA mission (the Space Shuttle SRTM mission) has contributed enormously to the development of low-cost academic research mapping applications. This is an example of a good scientific return on money that is spent by NASA (on behalf of US taxpayers of course) – as opposed to the ISS (a topic for another blog posting perhaps).