Editorial
Death in a Back Paddock
Editorial |
|
|
Death in a Back Paddock
Liablity for bad data rests with the user.by Jon Fairall |
|
A recent finding in the coroner's court in NSW has elicited little response so far from senior executives in the country's mapping agencies, but it should have, because it could have wide ranging implications for the way liability issues are handled in the mapping community. At issue was the death of George Fitzsimmons, a 52 year old fire-fighter, and three members of his team, Eric Furlan, 51, Claire Dean, 25 and Mark Cupit, 32. On the basis of evidence presented to her by survivors of the incident, Jan Stevenson, the senior deputy state coroner, told the court that Fitzsimmons was conducting a hazard reduction burn in Ku-ring-gai Chase National Park on 8 June 2000, on behalf of the National Parks and Wildlife Service. The fire trapped the team, and three of them died from smoke suffocation. Cupit survived but died later in hospital from complications associated with his burns. Although Stevenson identified a list of procedural problems in the management of such hazard reduction burns which contributed to the situation, she identified the primary cause of the deaths of the team as Fitzsimmons' decision to flee from the fire by moving uphill, contrary to all accepted fire fighting principles. However, she declined to lay any blame at Fizsimmons' feet. She said he had acted responsibly on the best information he had. That information was contained in a map issued to him by the NPWS, which showed a cleared hill top, which would have provided shelter. The map did not show a 30 metre cliff which stood between the team and the cleared area. She also noted that the map showed a path, which Fitzsimmons followed, running directly to a local motorway. In fact the path twists into impenetrable bush. She said the original botanical map had not been ground-truthed to include specific detail and had not had areas marked with such things as safe refuges where firefighters could retreat as required by the incident control system adopted by the NPWS. As a result of the coroner's decision, the NPWS is overhauling various procedures which might have made a difference on the day. However, no map-making authority was cited or blamed for the errors in the maps. Liability is, of course, one of the hot topics in current spatial information management. The question has been given urgency by typical Internet management strategies, which make it possible to distribute spatial data widely, in ways that cannot always be predicted. Data can be mixed and matched with other data, or included at inappropriate scales, all of which increase the possibility that it might be used inappropriately. |
Data managers have been afraid that in this environment, it would be all too easy for blame to be laid at their door. Frequently, they have used this as an excuse to restrict access to the data. Mercifully, coroner Stevenson seems to have provided a very clear principle for all common law jurisdictions - caveat emptor. The map-maker is not responsible for either errors or omissions; the map reader must use the map data with due caution. To a large extent, Stevenson is simply acknowledging life's realities. Maps will have errors so long as government's quibble about the cost of producing them. Even then, maps will be out of date long before they leave the shelf. Nevertheless, one cannot help but think that had Stevenson decided the other way - had she laid the blame more firmly at the door of the map-makers, it would have provided map-making authorities in all common law countries with a very large stick with which to belabour the holder of the purse strings. It doesn't require self interest to argue that more money for map-makers would be a good thing, because this situation will not get better by itself. In the future, spatial information will become imbedded in IT systems to an extent never seen before. It will be the basis of a terrain avoidance system for an airline pilot; it will underpin a system that will enable a shop-keeper to tout for business among passing mobile phone users; it will form the core of an ambulance dispatch system. Largely it will be invisible to the user, but the impact of spatial information in such systems will be profound. Unlike other IT problems, programmers will never be able to troubleshoot such errors. They will only become apparent when someone depends on them - when it's too late. As they become more ubiquitous, the probability of the failure of such systems will rise until it reaches certainty. The question is not 'if', but 'when'. It will be open to government to manage this risk by measuring the likely cost of legal action against the likely cost of fixing the problem. To a certain extent, this is inevitable, but it would be prudent to spend more rather than less on minimising the risk. Whether the public will wear such cynicism is an open question. |
Top of Page |
Table of Contents |
|
|
