Feature Article |
|
Dangermond@OZRIby Jon Fairall |
|
Jack Dangermond was in fine form at this year’s ESRI user conference on the Gold Coast, holding sway for a couple of hours during a late plenary session. And why not? Australian government GIS users love the man’s software, and he loves them back. And even to a disinterested observer in a later interview, the objective fact is that Dangermond – the man who invented commercial GIS, and is today still the leader of the pack – has a lot of very interesting things to say. Jack’s message: ‘you ain’t seen nothing yet’. The world is getting more fragile. Management is getting more important and more difficult. GIS is the pre-eminent tool for dealing with it, so doing GIS will become more significant as time passes. Is it difficult? Yes. Is it often done wrongly? Sure. But the consequences of not doing it are so dire that its importance can only grow. Possibly the greatest shift in geographic sciences over the last two decades is that we are no longer short of data. The challenge, however, is to turn that data into information, knowledge and wisdom. And that is no easier now that it has ever been. Dangermond says that if we are to manage our world, we need to measure, analyse and model the world appropriately so that we can plan, make decisions and finally, act. Communications is the key to this. None of these things can be done in isolation. The more people involved, the better the result. So a key feature of the direction of software in the future must be about the development of communications methodologies.Dangermond says this reduces to just two key technological issues. One is to design software so it can take advantage of the web. The other is designing software that can read datasets independent of the format in which they have been written. The former of these – using the internet – is largely an exercise in creativity for code cutters. The second is more about philosophy and corporate advantage than technology, and all the more controversial for that reason. Not that the internet is without its issues. Dangermond says its usage has now moved to a second generation. The first had to do with publishing. That is, the whole challenge of internet use was to get data to all the people that could possibly use it. There are many reasons why that still does not happen, but the technology is no longer an impediment. The big challenge for technologists now is to devise software that encourages collaboration. This is all the more difficult because it also implies a move from the simple assembly of datasets, through to modelling the processes described by those processes. Essentially, he says we need to start to think of IT systems as building blocks of large systems – of systems, of systems in which we can recreate models of the real world and its processes. This idea of connectiveness is at the centre of Dangermond’s thinking, so it is no surprise that he is also a passionate advocate of the need to eliminate the format issue that has plagued GIS since its inception. However, he is by no means convinced that the nostrums of the Open GIS Consortium, the International Standards Organisation or other bodies, are the right way to go. The OGC has proposed a series of standards, such as web feature services and web mapping services, with which compliant software must be able to interact. ESRI has supported these activities. Nevertheless, it seems that ESRI will concentrate on building its relationship with Safe Software in Vancouver. This way, ESRI users will have the ability to generate files in just about any imaginable format, and be able to read them as well. It is not that Dangermond has bad things to say about standards. He just seems to think they are not necessary, and given the level of interconnectivity heralded by the company in the forthcoming ArcGIS 9.2 suite, he may well be right. If Open Systems represents one threat to the dominant GIS paradigm, so does Easy GIS, or Google Earth as it is sometimes called. But Dangermond says that many pundits in the industry have overstated the case. Google, and other tools like it, are superb display engines, but it is just a mistake to think they are GIS. They have no analytic capability and their ability to link to a database is extremely limited. Having said that, ArcGIS 9.2 will have a very Google-like interface. Dangermond admits that Google has lifted the bar on what a mapping interface should look like. ‘I love, competition’, he said. Jon Fairall interviewed Jack Dangermond in Surfers Paradise on 3 November 2005. |
|
Top of PageTable of Contents
|
|