Journal of Law, Information and Science
|
|
Home
| Databases
| WorldLII
| Search
| Feedback
Journal of Law, Information and Science |
|
by
Kerrin Stewart, George Cho, Eugene Clark[*]
Geographical Information Systems (GIS) and ‘automated mapping’, as in other areas of information technology, has inevitably led to questions about the extent of liability relating to its use. Liability questions involve the torts[1] of negligence, misrepresentation and strict product liability. GIS as an end product[2] is the culmination of a combination of events including data gathering, human input, software and hardware components and computer programs. The chance of error is omnipresent with the combination of each component in the system. Added to this, is the potential difficulty of identifying and distinguishing where such errors have occurred. While the GIS process involves a chain of separate events, it may be difficult if not impossible to identify first the source and second to assign liability for error. Yet, this notion of accountability is extremely important in terms of the confidence consumer may place in the reliability and credibility of the technology for business activities. This reliance on GIS products gives rise to issues of liability, including the following: what exactly is the responsibility of each of the GIS players — the technology providers, the data providers and intermediaries and the users; does the existing law clarify their position; and, what steps may each of the players take to minimise liability?[3] This paper addresses these important issues.
...conceptually any GIS may be regarded as a system of map overlays. Overlaying one slice of information on top of a different slice produces combinations of associations and relationships previously unseen or even anticipated when the map is presented as a single layer.[4]
The use of spatial data in the production of ‘intelligent mapping’ is an area of GIS which is fast becoming commercialised — moving, like the Internet, from the government sphere to a public commodity. However, as this marketplace diversifies in terms of the use of electronic data and electronic databases, so too does the potential for error arise, and GIS is no exception. The dissemination of GIS information presents itself as a chain where each player, be it the original generator of the data, the person responsible for maintenance, the person who ‘value adds’ to the data or the intermediary who sells the product, assumes a certain degree of liability. The issue of establishing tortious liability arises when the information which reaches the user is defective is relied upon and the user suffers loss or damages.
This article addresses some of the more pertinent questions relating to the use of GIS from the perspective of tortious liability incurred either through the unintentional incorporation of error/inaccuracy or the intentional misuse of data/information in the creation of GIS products. This paper does not intend to cover every aspect of tortious liability. Rather this paper focuses on negligence (including misrepresentation) and strict liability to third parties for cases of economic loss, property damage and personal injuries. This article avoids addressing contractual liability relating to GIS as this is an extremely complex area of computer law which could be the subject of a different study.
As the electronic information industry moves from handcrafted demonstration projects to mass-marketed products distributed to distant and anonymous customers, the argument for extending liability when defects in these information products cause injury to consumers grows stronger.[5]
Digital data moving as it has from a purely public commodity to becoming available to the private sector, is a valuable resource, both in economic and political terms. While traditionally, it has been the public sector which has laid claim to much of this valuable information, the private commercial sector is developing to be a most important user of public information.[6] The growth of GIS is evidence of this progression. As a tool, GIS are used to help planners record new types of information, as well as one which links data from diversified sources. In mapping terms it is a “computer system capable of holding and using data describing places on the earth’s surface”,[7] using latitude and longitude and other spatial geographic information, to link data of different sets. As a technology, GIS are best described as a spatially referenced relational databases which capture, store, interrelate, analyse, and display an enormous range of spatially referenced information.[8] This process typically culminates in a ‘map’, merging geographic, cultural and scientific attributes.
GIS as marketable products have far reaching practical uses in areas such as navigation systems, land and wildlife management, emergency services, real estate analysis as well as a wide range of other developing applications. While the capabilities of GIS technology as an information resource appear indeterminate at this point, there is currently no supporting legal framework to protect and manage compilations of digital spatial data. This is so despite the fact that all of the areas permeated by GIS expose emerging legal and policy issues. This results in little certainty in a fast developing technology where it has been recognised that liability can only be minimised but never be eliminated.[9]
Negligence is often problematic because "standards" for GIS refers largely to exchange formats, geometric precision, and consistent terms and definitions, not to recipes for competent practice. And no formal guidelines separate the marginally improper from the generally acceptable…contractors and other information providers have an ethical, if not legal, responsibility to fully inform users including their government clients and the general public of the effects of aggregation and scale dependence.[10]
GIS are, in common with other similar complex systems, subject to the threat of potential legal liability. These errors may occur at any stage of data gathering and delivery, in the data itself, in the computer programs and software or in the outputs, and its actual use. How then do we assure that the information being used to compile these ‘maps’ is accurate? The question may be unanswerable save to say that one should recognise that computer information can never be guaranteed to be flawless, whether it be through error, omission, or the deliberate incorrect use of information. While Nimmer NS Krauthaus[11] suggest that anomalies will only be detected when there are limits programmed into the system, they further recognise that most computer systems themselves have an inherent error and potential for creating cumulative errors. Further they state that many of these errors are not readily detectable, even under close human scrutiny.
The actual process of data gathering represents only one way of producing erroneous material. It is the beginning of a process which may entail any of the following: merging of attributes, information production, enhancing, extrapolating, distributing, selling or licensing. Geographic data are inherently inaccurate for many reasons. This has been summarized by Juppelantz and Ziaofeng, who have identified a number of areas of GIS processing that produce error and as a consequence may attract potential liability and litigation.[12]
a. The source data may be inaccurate — such as the precision of the instruments, the skill of the surveyor or cartographer, the scale (the smaller scale maps may have less accuracy) and media on which the maps are printed and perhaps more importantly, the accuracy of the source data;
b. GIS processes frequently require human judgment:
- staff capabilities including the staff’s knowledge of the system they are using, their prudence, skill, and dexterity
- sampling techniques, such as spacing of the sampling frame and the spatial sampling rate
- equipment resolution, including sensitivity of the detector and signal to noise ratio of the detector
c. The data integration process may be inaccurate in the linking and integrating GIS information, including spatial, topological, temporal and descriptive error; and
d. Data interpretation by the end user may be erroneous. This may be due to improper data classification, lack of skill in data interpretation or physical disabilities affecting interpretation e.g. colour blindness.
These potential sources of error may be in any combination, ranging from entry of invalid data to the incorrect entry of valid data, extending as far as the failure to update stored data when changes occur.[13] To find liability is not simply a matter of attributing risk to either the system provider or data supplier. Error may occur either as a matter of mislabelling or unlabelled data or perhaps in the more complex GIS interconnections where it is not always possible to delineate between areas such as data entry and processing software. ‘Maps’ as mere representations of the real world evidence this anomaly, representing not only information, but also a combination of data and analysis as well as a degree of subjectivity when exercising cartographic licence in producing a more aesthetic map product.
The difficulty of identifying with precision a particular element in the chain has important ramifications in terms of decision making. The failure of, or inability to, detect different types of error has the potential to undermine the act of decision making, which truly represents the essence of any GIS. But this is also where the issue of liability rears its head. The ability to affect the process of decision-making is critical, for it is in the failure of GIS producers to address this issue either adequately or not at all, which leads to liability.[14] Epstein recognises those parties within the decision making process who may be exposed to liability from using inaccurate data or alternatively misusing accurate information. For example, where data is produced by one “agency for use by others who are unaware of data errors and its various attributes, and who may not understand what the information represents or what computations have been made” with the data and is distributed by “GIS managers seeking to recoup systems investments by marketing information products or services” produces intractable problems in founding ‘blame’.[15]
This categorisation of liability reflects broadly the three main players in the GIS process, the producer, the intermediary, and the final party in the chain the user, who singly or collectively become parties in any cause of action advertently or otherwise. Interestingly, Perritt (1996) claims that it serves little purpose to address the parties at either end of the GIS process (the producer and the user) when addressing spatial data, as their positions may have already been clarified by existing law — ‘the liability analogous to having been done on paper’.[16] Perritt refers here to the actual output, and disregards its physical manifestation, be it electronic or paper. The parties at either end of the chain it is claimed, in essence remains in the same relative positions of liability. It is where control becomes diminished, such as with intermediaries, where apportioning liability becomes increasingly difficult. It is in these further levels where tracing liability becomes more complicated as the information may either be used directly, sold, licensed, or value added before being shared or distributed. In terms of liability and accountability therefore, an error may occur at any of these nominated stages.[17]
The form and content of information provided to the public and private user community through GIS technology demands a new and more discriminating vocabulary to avoid unintended pitfalls from traditional laws developed before an information management tool of this utility was available to government.[18]
As the marketing of spatial data is continually developing, the law is still relying on contractual and tortious liability concepts drawn from traditional legal doctrine, primarily the law of property with its rights and liabilities, and the law of torts. The courts however, by applying existing legal standards to liability and risk in an information technology context[19] are merely projecting an existing structure to accommodate computer liability issues, particularly the legal concepts of negligence for loss or damage, and proprietary issues such as copyright.[20] This means that the threat of liability exposure in relation to GIS as an information resource, remains unclear. It is this absence of reconciliation between spatial data and the application of general law which has resulted in a ‘lag’ in the law. Certainty in the law is crucial because it gives assurance to businesses dealing with GIS by acting as a ‘safety net’ in the electronic marketplace.
The questions must then be asked: how long can the law keep on extrapolating traditional concepts to meet the challenges of this emerging technology[21] and how does one specify with any certainty how far liability extends? The expansion of the Internet by contrast, has brought with it the advent of computer regulation, imposing formidable civil and criminal liabilities for negligent or illegal on-line activities particularly in the U.S. The Internet as a major growth area “...has evoked legislators, regulators and the judiciary to tame the seemingly untameable on-line universe”.[22] There has not as yet been quite the same level of public attention directed towards GIS.
Although it is recognised that in certain circumstances, new standards of liability will be required by the new environment, Nimmer and Krauthaus[23] take the view that the courts have an opportunity to ignore the effects of liability from computer-generated information by proceeding on the artificial basis that computer systems do not actually exist (or in fact play any part in the making of relevant decisions). They maintain that by ignoring the subject matter, basic expectations can be retained and applied in a rational manner.[24] Nimmer and Krauthaus’ proposal, recognises the difficulty of categorising computer-based processes. GIS in particular, touches many areas of the law such as intellectual property, privacy, contractual rights and obligations, and negligence determined through the application of standards of care. The difficulty in delineating between particular GIS areas adds to the complexity of the task of assigning direct responsibility to any one party, or in fact to any party at all.
It is impossible to anticipate the standard of professional care that courts will apply to software suppliers because of the daily technological advances made in the industry. The standard of the reasonably prudent person would be meaningless in the software context...because software development is an inexact art which allows an end result to be achieved by many different methods...While it might not be difficult to show that the plaintiff’s injury was the proximate result of the defendant’s software, it might be difficult to show exactly what act caused the injury.[25]
Software itself represents only one example of the difficulty in establishing causation to determine liability. One area which has been clarified by the U.S. courts however, is where liability must fall for any fault contained in the actual computer data. In State Farm Mutual Automobile Insurance Co v Bockhorst,[26] the Tenth Circuit Court considered the obligation of reasonable care imposed on those providing computer data. The court held that even though the actual processing of the policy was carried out by an unimaginative mechanical device, the company still bore the responsibility for errors and oversights. Liability they held, must therefore lie with person and not the machine.
This ratio is especially important for it imposes a standard whereby the involvement of a computer does not provide any kind of shield from liability. But, this line of accountability in computing however, does not mean that a greater liability attaches here, than to any other area. The 1980 case of Chatlos[27] highlights the court’s reluctance to hold those providing computer services to a greater standard of care, “...simply because an activity is technically complex and important to the business community does not mean that greater potential liability must attach. In the absence of sound precedential authority, [this] Court declines the invitation to create a new tort”.
The question of attachment of liability to those in the ‘chain’ is not an easy task, unassisted as it is by any specific GIS case law. The lack of relevant legislation too makes it difficult for GIS producers to assess their potential liability. The only state in the U.S. with a specific provision for limiting GIS liability in a contractual capacity is Alaska.[28] In a possible pointer to future legislation elsewhere, that particular statute[29] provides that, ‘for electronic services and provisions, public agencies limit the liability of the agency providing the service or product’. In the absence of legislation, producers (including government agencies), are providing their own disclaimers. Australia’s Land and Surveying Group, AUSLIG[30] provides the disclaimer that,
...the Commonwealth does not warrant that the data are free from errors but will advise any revision or update to the data which may be made during the currency of the licence.
Such a disclaimer may be of limited utility with regard to the duty of care that is owed by the provider to the user. Tort law may provide the legal basis from which action is brought for personal injury or property damage resulting from the intentional or negligent act of a party. This area of law is most often relied upon to found actions for the negligent dissemination of faulty data when there is no pre-existing contractual relationship between the parties. When the result of any of these negligent acts result in harm, injury, death or economic loss one must turn to the torts of negligence and strict liability.[31] For this reason it is important to address the liability stemming from the rights and obligations which form the basis of this part of the law.
...providing insufficient data quality in the GIS database for unforeseen uses and therefore causing unreliable decision making, causing unintentional harm to others such as degradation in the quality of life or safety of others through the use of GIS, selling the same GIS product to a second client without acknowledgment...allowing deception in the use of GIS for analysis where the means are suspect but a socially desirable result occurs, and causing intentional harm to others through the use of GIS.[32]
Identifying the potential unethical use of GIS may not be hard to find or isolate. However, identifying the source of error and finding the party or agent responsible may be a more difficult task. The critical issues are thus: firstly, what happens when the information is erroneous and secondly, who is ultimately responsible? Finding who is liable is important particularly in an era when as public information providers seek to market and sell information, and information products, to a wider and wider audience. As far as the providers are concerned, it is imperative to determine a standard of care — for such standards may enlarge or reduce the market for spatial data. So too will the users of maps and other geographical information products, who would demand data and information at a reasonable cost and wish to reduce the uncertainty of their management decisions and to minimise their exposure to liability.[33]
There exist two broad types of liability, civil and criminal, with the former concerned with the private rights and remedies between public and private persons, and the latter concerning the obligations of the citizen to the state. Civil liability can be further broken down to liability under two of the more common legal concepts — contract, and tort. Contractual liability focuses on a breach of a term or terms of a contract by parties who have duties and obligations under that contract and agreement.[34] Tortious liability on the other hand, can arise without the parties even having knowledge of one another, ‘their only connection being the occasion giving rise to the liability’.[35]
Thus. tort law unlike contractual liability does not require privity,[36] however it does require proximity.[37] This means that although a party may not be the direct cause of the negligence alleged, they may still be held liable as having contributed to the harmful consequences. This nexus is important, for it provides the link between the tortious act and compensable detriment. Many of the processes involved in creating GIS products relate to the processing of data and the analytical methods used to interpret this data. It is in this context that we shall refer to the fundamental standards of torts, which include: appropriateness, relevance, validity, clarity and acceptance.[38]
In order to present GIS data and analysis in court, one must demonstrate that it is relevant and reliable. The ultimate authority on the acceptability and/or quality of the spatial data presented will be the judge and jury during trial. If spatial databases are to be used effectively, and if they are to withstand the scrutiny of the judicial system, then one must consider the consequences of legal and public policy interpretations of one’s data from the beginning.[39]
In common law countries, tortious liability is governed primarily by precedent, though this may be subject to certain statutory rules.[40] The onus of proof is generally on the person who brings the action to prove causation - that in fact the defect existed at the time of the accident and that this defect resulted in injury. The question which is asked is, would the plaintiff’s injuries have been suffered but for the defendant’s negligence? If the answer is no, there is no action. However if the converse is true, the players in the chain of information must assume responsibility for damages either compensatory, personal or exemplary.[41]
Retailers and value-adders, like manufacturers of the GIS service or product, may be considered liable for defective products on the basis of proximity.[42] The general principle of proximity is that loss or harm will be attributed to the most immediate effective causal factor rather than to something earlier, though that may have been an essential prerequisite to the loss or harm.[43] In its approach to the chain of causation a court must assess proximity by addressing each event making up the links in the action, and determine whether that party caused or contributed to the plaintiff’s injury noting that a subsequent but intervening act may break the chain of events in effect alienating the negligent act from the resulting injury. This is referred to as novus actus interveniens or the new intervening act.[44]
Lord Wright in the case of The Oropesa[45] held that a new cause may disturb the sequence of events be it unreasonable or extraneous or extrinsic, which may negative, reduce or limit the causal connection. This intervening occurrence too may occur in the GIS process, involving as it frequently does several stages of development or modification leading to the end product. If it is held that the chain of causation has removed a party in the causal chain from the harmful consequence, it may be suggested that there can be no resultant liability.
While the U.S. case of Independent School District[46] did not apply specifically to GIS, it did highlight the extension of liability to the data collector for injuries to the end-user where there was no contractual relationship. On the basis of this case, Johnson’s article[47] holds that this analogy will apply to collectors of spatial and non-spatial geographic information in that they too may be held liable for consequent injuries. Intermediaries then, also share in the relationship between the injurious product and the end user through the action of passing on the defective product to that end user. The action of ‘passing on’ was addressed by the Californian Supreme Court in Vandermark[48] where it was held the retailer may be the only party from which the plaintiff may seek recompense in terms of liability. While in this case it was held that the retailer was accountable, the obiter dicta stated that consumer protection is further enhanced when both retailers and manufacturers are jointly accountable.
The accountability as seen in Vandermark finds its basis in the public policy requirement for checks and balances, working ultimately for the protection of the consumer. The court in this case held that this practice serves to pressure and encourage the party responsible for the defect to turn out safer and more attractive product. This is referred to in the law of torts as a duty of care and is owed by the producer or intermediary in providing a service or product. A failure to provide the requisite standard of care to the user may constitute negligence and liability founded.
In order to establish negligence, it will be necessary to establish that a duty of care, to ensure the accuracy of the information, was owed to the user. The user must then establish that the conduct of the creator or the owner of the GIS in operating or maintaining the system, fell below the standard of care required. In view of the systems infancy, however, such a determination may be problematic in that the requisite or industry standard of care has not yet been established. Finally, the user must establish that the damage suffered would not have occurred but for the breach committed.[49]
There can be no negligence without fault by the defendant.[50] The actual negligence may take several forms including: a negligent misstatement, supply of inaccurate data, misrepresentation of data, a breach of statutory duty, or operational negligence (data collection, hardware and/or software). An important decision which established some basic terms in the tort of negligence was the landmark case in 1932 of Donoghue v Stevenson[51] where the ‘neighbour test’ of a duty of care was established. Lord Atkin posed the question, “Who then, in law, is my neighbour?...such persons as are so closely affected by my acts or omissions that I ought reasonably to have them in my contemplation when directing my mind to the acts or omissions called into question.” In the subsequent case of Hedley Byrne & Co Ltd,[52] which involved the act of providing information or advice, Lord Pearce applied the neighbourhood principle of duty of care to economic loss arising from negligent misstatement, observing,[53]
How wide the sphere of the duty of care in negligence to be laid depends ultimately on the court’s assessment of the demands of society for protection from the carelessness of others.
Until this case the English courts had held that a defendant was not liable for financial loss flowing from negligent misstatements in the absence of a contractual relationship or fiduciary relationship even where damages suffered by the plaintiff may have been reasonably foreseeable, thereby evidencing a broadening in the test of the neighbourhood duty.[54] This introduces a major part of tortious liability whereby negligence or damage arising from carelessness is compensated by the person who caused it (through neglect or otherwise), a theory that has become an underlying premise of tort law.[55] Should a duty of care and breach of the requisite standard of care be established such that, but for the breach, the resulting damage would not have been sustained, negligence will be established.
The other influential English case in terms of liability is that of Caparo Industries Plc v Dickman[56] which confirmed and clarified the position of the Australian courts in terms of liability. Lord Bridge held that liability depends on the existence of certain factors beyond the foreseeability of damage giving rise to a duty of care,
...there should exist...a relationship...of ‘proximity’ or ‘neighbourhood’ and that the situation should be one in which the court considers it fair, just and reasonable that the law should impose a duty of a given scope on the one party for the benefit of the other.
In terms of computer software and developers it is difficult to assess how wide the neighbourhood principle extends. While the courts have been hesitant to widen the principle to unlimited liability as a matter of public policy, the courts are also aware that unlike the ginger beer in Donoghue[57] which can only be drunk once, software may be used by many others and perpetuating the damage or causing multiple damages.[58] This notion of proximity then must be considered not only in terms of the person to whom a duty is owed but also to the level of damage sustained. What is clear however, is that proximity, as previously addressed, means different things in different cases. The general basis of proximity relies on the concept of blame being extended to include an interested party. This means that a contributing party such as a supplier may likewise be ‘proportionally liable’ under the test of proximity, to the extent of their duty of care.
Establishing proximity is likely to become important where the end users of faulty GIS products and services take action to recover consequential business losses. These losses however may not be purely in economic terms as demonstrated by the case law under the statutory terms of strict liability (where physical injury and death have been compensated). Whereas traditionally the courts have been hesitant to award for mere economic loss without associated physical damage[59] the standard and duty of care have now developed past the simple notion of a physical wrong to now incorporate that of mere economic loss.[60]
Over time, legislatures and courts have defined the ‘standard of reasonable care’ that one can expect in everyday situations. This notion of care, may be used to determine the degree of care expected from GIS producers and suppliers. This standard of care arguably attaches to the producer or the intermediary in favour of any user placing reliance upon erroneous or misleading information retrieved from any GIS product or service and identifiable harm is suffered. The duty expected of the creator would be that the information be substantially accurate and that every attempt was made to use error-free procedures. When the common law test of the reasonable person is applied, one can better determine if a reasonable standard of care was not only established, but followed. The reasonable person test too applies where any injury which resulted could have been contained or reduced. This refers to the defences in an action of negligence including the mitigation of damages,[61] contributory negligence,[62] and the voluntary assumption of risk.[63]
GIS products and services are the end product of complex accretion of data from a number of different sources. It is therefore unreasonable, indeed impossible, to expect that these products would be entirely free of error, be it in the original data or the result of the process of incorporation into the end product. What is expected however, is a duty of care based on the broad principles of what is deemed acceptable under the reasonable person test.[64] Epstein and Roitman (1986) state that the system for data entry in particular must be designed to achieve a satisfactory level of freedom from error ‘...If somebody claims damage as a result of errors in data entry, a court is likely to examine the process used to determine whether a reasonable standard of care was used to minimize data entry errors’. [65]
The more likely a particular system is to have a significant adverse effect on a person, the higher the level of care to which a court may hold the system’s developers, owners and operators liable.[66] While standards of good practice have yet to be established particularly in relation to software, it is recognised by Chalton[67] that there are two definitive areas of breach. The first is committed by the employer in their failure to establish a proper system of working and quality assurance in their products. The second breach is committed by the employee where the individual programmer or analyst fails to provide a duty of care in the performance of their work.
It is clear that excessive reliance on computer data without proper safeguards to insure the reliability and accuracy of the information may constitute failure to exercise due care, and in some cases even may result in the award of punitive damages.[68]
The duty of care under the tort of negligence relating to inaccurate computer information becomes applicable to a party who may have to make recompense for loss and damages. As there exist relatively few cases[69] attaching liability for a GIS it is necessary to address turn to related areas of negligence and strict liability where manufacturers and distributors have been held to be liable for the provision of information. It may be pertinent to address liability cases relating to the supply of on-line information and software packages, as a parallel case study to that applied in the GIS context due to their obvious nexus. The common feature of most of the U.S. cases addressing the issue of strict liability for error are based on the supply of inaccurate information by manufacturers or intermediaries, and breaches of the duty of care as owed by these parties the end users.
...some GIS, emergency communications, and health care information services products are intended to be used in ways that would create considerable hazards if they are defective. This is obviously the case with aircraft navigational systems, medical telemetry systems, ocean navigation systems, and...emergency response databases. In these cases policy arguments for strict liability exist. An ordinary consumer as an injured party has no real capacity to determine defects or even to shop around for the safest GIS...By offering such systems on the market, the producer is impliedly vouching for their safety.[70]
Strict liability, that is liability imposed by statute and established without the need to prove fault, is based on a ‘failure to take care’ and is now often protected by the Sale of Goods legislation.[71] In terms of GIS, strict liability does not directly affect the intermediary in the GIS chain of information, but does directly affect the creator or producer. This liability, supported by public policy, performs two important functions: firstly it provides some protection to the plaintiff from the onerous burden of proof; and secondly it ensures goods are of a merchantable quality. It is not necessary therefore to show that the manufacturer of the product was at fault, just that the product was inherently defective and caused harm.[72]
The law of unfair competition is the umbrella for all causes of action arising out of business conduct that is contrary to honest practice in industrial or commercial matters...Unfair and deceptive practices are likely to be as common in the on-line world as they are elsewhere Indeed, because information can be easily manipulated, combined, and transmitted on-line, the potential for deception may be even greater.[73]
Similar to the legislative provisions of the Restatement (Second) of Torts in the U.S., which provides the basis for commercial liability within traditional tort categories, so too are there specific legislative provisions in Australia. Australian statute addresses restrictive trade practices in the Trade Practices Act 1974 (TPA), which covers areas such as fraudulent misrepresentation and strict liability. The momentum for strict product liability in Australia, came from a recommendation in a 1990 joint report, which recognised that compensation should be paid by the manufacturer, the importer or supplier, not borne by the consumer and the State. While the legislation which was enacted did not strictly reflect this proposal, the TPA was an important step in incorporating various provisions regarding defect liability and the consumer.[74]
Relief in respect of unconscionable conduct[75] in trade or commerce is available under Part IVA of the TPA, and equivalent State Fair Trading Legislation. Where conduct is unfair in that it is misleading or deceptive, it falls under Part V of the Act. While section 52 presents the general prohibitions, representing what is considered the ‘heart’ of consumer protection, Part V extends much further in its list of prohibitive practices including: false representations in the supply of goods or services (s 53); misleading conduct as to quality of goods (s55); and, supplying goods not complying with product information standards (s 63), to name a few pertinent sections.
Miller in ‘The Annotated Trade Practices Act’,[76] acknowledges intermediaries and s52 liability by addressing several relevant Australian cases. In the case of Yorke v Lucas[77] the Court suggested that where an intermediary makes it clear that they are not the source of the information and disclaim its truth or falsity, they may avert misleading conduct when the information is incorrect. French J supported this notion in the case of Gardam v George Wills & Co Ltd,[78] when he stated that where an intermediary acts merely as a conduit, innocently passing on information they are not liable for false representations. The cases in this area of intermediary liability however accept that the representation and circumstances under which that representation was made must be taken into account — the defence therefore may not always absolve the party of part responsibility.
Part VA introduced into the Act in 1992 addresses, ‘Liability of Manufacturers and Importers for Defective Goods’. This section, while neither based in tort or contract, is a basis for liability where injury, loss or damage has been suffered.[79] In relation to GIS this is best illustrated in s75AD where - any individual who is injured by a defect in goods may recover compensation from the manufacturer or under s74A(4), 75AB, 75AJ — if the manufacturer is not in Australia, compensation may be sought from the importer, or, if the manufacturer remains unknown, the supplier. However, these provisions of the Act are somewhat watered down by the preceding provision; s75AC (1) — which provides that for the purposes of Part VA, goods only have a defect ‘if their safety is not such as persons generally are entitled to expect; and, s75AK (1) (c) — which provides that the defendant may seek a defence if the state of scientific or technical knowledge do not enable discovery of the defect.
Recovery of damages under the TPA is found under s82(1) which covers both a breach of restrictive practice provisions (Part IV) and consumer protection (Part V).[80] The award of damages however favours a compensatory approach rather than punitive or restitutive, as it is designed to compensate not penalise the defendant. The injured party is then placed in a position prior to the wrong having been committed. To take action under this section however, four elements must be proven:[81] a cause of action — that the plaintiff suffered actual harm; causation — a causal connection between the breach and the loss suffered by the plaintiff; the damages are not too remote in law; and, mitigation by the plaintiff of damage.[82]
Intervening and contributory negligence may reduce or obliterate overall liability of the defendant. Section 82(1) breaks the chain of causation where the plaintiff failed to take reasonable care, though this requires more than a failure to check the accuracy of the defendant’s representation. If the defendant’s contravention plays only a minor part in the resultant loss the court, under s82(1), may reduce the plaintiff’s damages as occurred in Brown v The Jam Factory Pty Ltd.[83] There is no restriction on recovery for economic loss under s82(1), as occurs in some torts, for it is recoverable regardless of damage to the plaintiff’s property or person.
Under tortious actions the plaintiff may suffer severe economic loss as in loss of profits, income, enjoyment of a service, interest or financial.[84] Under the old common law rule economic loss was only recoverable when it accompanied harm to person or property. This though has been substantially developed through negligent misstatement and proximate cause as evidenced in Hedley where recovery for pure economic loss was upheld, and in Caltex Oil (Australia) Pty Ltd v The Dredge Wilemstad.[85] In this latter case Stephen J,. “...criticised the general exclusionary rule for economic loss but said there is a need for some control mechanism based upon notions of proximity or sufficient proximity between the tortious act and harm”.[86] In this case all the factors working together, the defendant’s knowledge of economic harm and the nature of the loss and damage to the plaintiff, proved proximity and reasonable foreseeability.
A further attempt to address the issue of liability in Australia (after the implementation of the Trade Practices Act) arose with the failed introduction of the Professional Standards Bill 1992.[87] The then Attorney-General of New South Wales indicated that the Bill emerged through a response to the ‘problems of professional liability and insurance in the common law world’. The problems, as noted by the then Attorney-General of Western Australia, mirrored those existing in the U.S.[88] It was concluded that this situation could befall Australian society if liability was to continue to go unchecked, particularly given the statistics and anecdotal evidence surrounding tort liability in America.[89]
Despite a similarity in the approaches adopted by New South Wales and Western Australia during a meeting of Attorney’s General on 2 July 1992, the then Commonwealth Attorney General Michael Duffy, believed that a ‘ceiling’ on professionals’ liability was not a solution. His response was not considered to be a rejection of the concept, merely a deferral until a united Commonwealth approach could be adopted. While this Commonwealth stance never eventuated, the 1992 Bill evolved into the, ‘Professional Standards Act 1994’ (N.S.W), which highlights the perceived need for regulation among the states.
Courts in the U.S. have had to address the issue of applicability of strict product liability when injury has been caused by defective information. In Winter,[90] the Ninth Circuit rejected a products liability claim against the publisher of ‘The Encyclopedia of Mushrooms’, who had misidentified a poisonous variety of the fungus, claiming it was edible.[91] The imposition of strict liability was averted in this case on the basis that the expressions and ideas contained in the book were deemed intangible — whereas “product liability law in the U.S. is geared to the ‘tangible word’”.[92]
Winter is a significant decision in general terms of dicta and the law of liability for it demonstrated the difficulty of applying liability to information resources. The court in this case held it was not necessary for the publisher to check the accuracy of the information nor did they want to inhibit the dissemination of information[93], protected as it was by the U.S. First Amendment with its manifesto of a free flow of information.[94]
This rationale was also used in the prominent U.S. libel case regarding digital transmissions and on-line access providers, Cubby Inc v CompuServe Inc.[95] Compuserve as a distributor were held not to be liable as a publisher of an “electronic, for-profit library” on the basis that they were a mere conduit, not a contributor to the information. This case does not mean, that in preserving free speech an intermediary disseminating information will automatically be absolved of liability for the distinction in Cubby’s case centred on CompuServe’s lack of input. To have contributed, it would need to have been established that Compuserve knew or should have known, the harmful tendency of the information flowing through its service. As a conduit, the court held it was not feasible for CompuServe to examine every publication it carried for potentially defamatory statements.
This case has important ramifications in reducing the threat of liability to other on-line access providers, for it created a precedent that without evidence the intermediary knew of the ‘infringement’, those merely passing on the information should not be held strictly liable.[96] While these cases only refer to publishers, a similar analogy may be extended to those who merely distribute, not contribute, to the GIS information. This defence is not applicable however, when it can be proven that the intermediary knew, should have known or had a duty to ‘inspect’ the GIS product. The jurisdictional question must also play a part in that international actions are not always successful. This is evident in the case of Asahi,[97] where a defective part made in Japan became a finished product in Taiwan and was then exported for sale in California. The Supreme Court held that the originator of the Japanese product was not liable under Californian jurisdiction.[98]
In Smedinghoff’s book Online Law[99] it was recognised that information electronically communicated information, where the originator of the messages cannot be held liable for the harm done for reasons such as anonymity, there exists a tendency to hold third parties liable for any subsequent harm attributable to that defective information. The result of this is that on-line service providers, systems operators, employers, and other third party intermediaries are legally liable to account for damages. On a more general note these intervening layers of liability are somewhat diminished where an intermediary or supplier can prove that they only ‘insignificantly’ altered the product. This does not mean they are absolved from liability if they added to the injury, but the degree of damage would be balanced according to the quantum of their input.
Much has been written on defective computer software and whether it should be subject to strict product liability. The concept of a defect ranges from manufacturing defects (where the program does not conform to manufacturer’s specifications) through to a failure to provide adequate warning of the consequences of using the software and finally including defects in the design of the software itself. Liability, with respect to defective spatial data, introduces the elements of reliance and intended use. This is exemplified in the dissemination of information for the creation of navigational maps and charts. Such liability can be either under tort law or statutory under strict liability, that is, positively conferred by operation of law requiring no proof of negligence.
…the responsibility for the correctness of the traditional charts is becoming hazardous in law’…many more problems (are anticipated) when data are routinely made available and used in computer form. System designers of both hardware and software, as well as data suppliers, will be responsible for their individual products but may also incur part of the overall responsibility for the satisfactory workings of the whole hardware-software-data-human interface system. [100]
A very visible area of GIS is that of surveys, charts and maps which form its basis. Error in this area stems from map error, inadequate design, and the use — including both unintentional or intentional. As with the tort of negligence, the reasonable person test will be applied in the procedures and processes for data entry and the design and implementation of the cartographic process.[101] This was illustrated in the case of Zinn,[102] where a water level on a geological survey map unintentionally appropriated property without due process, removing not only title but rights to the property as well. In mistakenly indicating an inaccurate level above sea level the State was held liable for the owner’s losses.
Maps and charts are, in law, not considered ideas but technical data.[103] The U.S. courts have consistently viewed both as ‘practical’ rather than ‘creative’ as their purpose is to guide. In Brocklesby v United States[104] charts were distinguished from other publications such as books in that charts are designed to be used by people (such as pilots) who make immediate decisions based on this information. As a result, the duty of care is crucial due to the element of reliance. This area raises the issue of liability in two ways: firstly in terms of potential liability to be incurred by the GIS when it can be established that a detrimental reliance resulted from use of the data; and secondly, liability arising from negligence in the work of the creator[105] through failure to perform in accordance with the ordinary standard of care required of them in a professional capacity.[106] The following case illustrates the difficulties involved.
Jeppesen Sanderson is one of the main producers of charts and maps in the United States, after the U.S. government. Until 1986, the company had won only two of six lawsuits in which it had been claimed that their charts had contributed to, or been the sole cause of aeroplane accidents. These claims have taken different guises from strict product liability, to breach of warranty, through to negligence.[107] In Brocklesby v Jeppesen[108] a pilot using a chart (produced by Jeppesen) to make a landing, subsequently crashed into a mountain killing the entire crew. Action brought against Jeppesen and the government claimed that the accident was caused by defects in the instrument approach procedure as developed by the government and published by Jeppesen.
Jeppesen in this case were held liable as the information provider involved in the gathering, repackaging and selling of federal aviation data.[109] The Court’s final judgment against Jeppesen was an award of damages in excess of $11 million dollars for negligence, breach of warranty and strict liability on the basis that they were under a duty of care test their products and to warn of any dangers. A similar conclusion was reached in the U.S. case of Remiga[110] where the government was held to have inaccurately, therefore negligently, depicted a broadcasting tower on an aeronautical chart which subsequently resulted in fatalities.
It is however possible for all of the information to be correct and for it to be misapplied, as occurred in Aetna Casualty and Surety v Jeppesen & Co.[111] In the graphical representation Jeppesen had displayed a landing field using two different scales, one used on the plan view and another used on the profile display. By failing to recognise this anomaly, the pilot relied on a chart feature which was perceived as being fifteen miles from the airport when in fact it was only three miles, and subsequently crashed the plane. Jeppesen as the publishers of the charts, were again held to be at fault even though the display of instrument aeroplane approaches for pilots were accurate, the court holding that ‘the professional must be able to rely on the accuracy of the information’.
Although this again supports the notion of the duty of care owed by the creator to the user the court further noted that the professional must also play a part in exercising certain judgment in evaluating the information. The requirement of the professional is that they not rely blindly on what is shown by the computer or the charts. This professional discretion is based on the premise that obvious anomalies would be questioned and a professional judgment exercised to avoid detrimental consequences. A professional is generally held to exercise a higher standard of care than an a non-professional on the basis that they have a minimum standard of special knowledge and ability, not shared by the public at large.[112]
Weber[113] comments that although there are good arguments for the introduction of strict products liability for computer software, if the courts use too widespread an application of the strict products liability regarding defective software, it will cripple the growth of the technology and industry as a whole. Flexible application is therefore important, as the court’s determination can be expected to affect innovation, and indeed whether the industry will survive.[114] This application of strict liability however may be a double-edged sword in that, if the courts take a flexible approach and hold strict liability to be inapplicable, plaintiffs will be forced to rely on negligence principles which increases their burden of proof and as a result confers added protection from liability on software manufacturers. Resultant litigation however, does not present itself as being the only avenue to mitigate liability. The answer to this dichotomy perhaps lies in proactive measures a producer may undertake before the process of information distribution is even considered.
Improved geographic information handling capabilities are continuing to find expanding applications throughout society, and the eventual public and private investment in capabilities is being estimated in many billions of dollars…Along with its positive effects, the negative impacts of the technology and its associated databases need to be considered. The negative impacts need to be divulged, eliminated, minimized, or accommodated and weighed against the positive.[115]
It is imperative where possible to consider the implications of the data from the very beginning. In doing this, a producer or intermediary will become aware of the opportunities or actions they may undertake, to effectively minimise the risk of legal liability. Although many authors contend it is litigation which will establish liability standards in turn regulating the industry, it has also been postulated that the answer lies in a more competitive market with a lower priority on the courts. While the following areas do not provide an answer by themselves, they reflect a market concern regarding the protection of suppliers and distributors so that the expensive process of litigation may be averted.
The design and development of standards for spatial products would provide one answer to reducing the threat of liability. As applied to the design and operation of information systems, standards are a means of reducing the liability risk which affects producers, sellers, and users, both professional and those not considered ‘expert’ in spatial products. GIS products produced to a certain standard would establish a level of care expected in those parties as well as offering an answer to reduced litigation in terms of what was known by the decision-maker. This may serve not only to lessen potential legal problems but also to reduce costs of litigation and legal investigation.
Although it appears that the area of GIS will need to develop before defined standards may be accepted by courts (as standards against which to attribute negligence), six areas have been identified by Croswell and Archer (1994), as target areas of possible development:[116]
1. Hardware and Physical Connection Standards;
2. Network Communication Standards;
3. Software Standards;
4. Data Format Standards;
5. Data Presentation/User Access Standards; and,
6. User Design Standards.
This attempt to introduce new standards is not new. The Australian Commonwealth Spatial Data Committee in 1993 addressed the rights and responsibilities of ‘custodians’ in terms of GIS data and information. The term ‘custodian’ refers to the fact that more than one player may have possession of GIS material meaning they have a duty in terms of physical and legal custody of data and information. This system is based on the notions of integrity and accuracy, whereby each custodian works to avoid duplication, maximises value added benefits as well as assists new information product production and management.[117] It is believed this system would integrate the whole, making the entire application more useful and effective.
The licensee…should expect that he may encounter errors and that not all of these errors give rise to a liability on the part of the supplier.[118]
Traditionally insurance mechanisms in the U.S. have been a way to transfer risk of loss for software developers and computer consultants, however fast track technologies such as GIS mean that these policies may no longer protect these providers. The reality is that liabilities can be greatly expanded by the addition of a new service in rapidly expanding technology and service.[119] Whilst tort law is based on compensation for the adverse consequences to one person for the ‘morally’ culpable conduct of another, the new ‘communitarian’ view proffers that accident costs should be collectively borne for the purpose of implementing and administering compensation schemes.[120]
A retailer in the U.S., if found guilty under strict product liability, may get indemnity from a ‘higher entity in the chain’, usually the manufacturer. This shift of fault can only occur when two criteria are met: firstly that the retailer had no actual or constructive knowledge of the defect when the product was sold, and secondly, that the retailer was under no duty to inspect the product.[121] Considering that governments often produce spatial data for mapping, it would not be all that unusual to find liability lying with the retailer, who is the middle person standing to gain from the sale of the product. The issue of indemnity is a controversial one. The crux of this lies in the fact that to protect themselves, each intermediary must take out an insurance policy which will bear the cost of product litigation.
It is likely that with the increased number of users, licensing will become more prevalent. Contractual obligations may be imposed on government agencies supplying data, though data supplied, sold, or under a licence agreement, may or may not carry an implied warranty that the information may be relied on depending on the particular facts of the matter. If a governmental agency were to actively market GIS information, the agency’s responsibility extends from custody of public data through to its subsequent use, either internally or externally. It is important to note here, that while sovereign immunity has receded in value as a shield particularly for Government Business Enterprises working in a more competitive market, it has not as yet been extinguished.[122]
The rationale behind Crown immunity lay in the government’s discretionary power, in that policy formulation should be held sacrosanct — not to be interfered with by the threat of liability. Now, however, if it can be proven that a party was not acting in good faith in the carrying out of their duties the Crown generally may be subject to potential liability in respect of a tort committed by any employee or officer of the Crown. This attitude has been adopted by many jurisdictions in the U.S. and in Australia despite the abrogation of sovereign immunity in place of more accountable, responsible and competitive government.
In Dansby’s (1992) commentary on selling GIS data, he prescribes certain scenarios for specific liability defences. It is stated that where there is a licence for data, the relationship is clarified. An agency (generally government) will be able to exercise more control to protect itself against potential liability for the dissemination of that information. Where cost recovery for distribution of geographic information is involved, suitable warranties would cover the data and information. Both warranties and disclaimers relating to the quality of information may be used to limit the exposure to loss and damage from reliance on the disseminated information.
A disclaimer goes some way toward the question of end control of liability in GIS data. A disclaimer may cover information such as data quality, scale, source materials and any known limitations of the data and may include statements eliminating the data provider from any liability for misuse of the data. For GIS material it may include declaring the data valid to a certain date, or simply by providing a date when the data was last updated. This however is the extent of the disclaimer, it is certainly no panacea to the data provider as they may still be liable for any errors, omissions, or misinterpretations of data quality.[123]
The benefits of a GIS contract are obvious in that they may contain a comprehensive spectrum of factors including testing, maintenance, support, finance, warranty and liability.[124] Apart from a contractual relationship, if the right wording is used in a disclaimer, liability can generally be averted. It must be qualified that this will only operatewhen we refer to economic loss and will not generally extend to personal injury or wrongful death. The maxim of caveat emptor — ‘buyer beware’ is still relevant, as is the element of using professional judgment when evaluating the information. It has been suggested that an initial analysis of liability be undertaken by original designers and programmers to decide whether the risk of potential liability is in fact reasonable or whether it is outweighed by undertaking the new use. This analysis would also highlight whether a contract or disclaimer is appropriate in addressing the capabilities and potential pitfalls of the system.[125]
The future growth of GIS and its range of applications is not simply determined by the advent of new technology, such technology enables but does not dictate destiny.[126]
Reliance on computer technology in the form of digital information has become more common-place particularly in the world of contemporary business in terms of aiding decision-making with a focus on enhancing business efficiency. The issue of tortious liability in the area of information has become more prominent with reference to GIS. Such an examination has given some insight as to how the courts will respond in future GIS liability cases. This analysis also allows a decision-maker, armed with a better understanding of liability boundaries, to compete in a stronger position while operating in a new information technology environment. It has been held that certain requirements are essential if GIS outputs are to be held as reliable, including that GIS user is informed, the data is adequate and the software dependable.
In these early stages of the development of GIS and the law, with an obvious absence of legislation, industry standards, or adequate insurance coverage, one can expect the courts to play a greater role in the evolution of spatial products/electronic information and liability.[127] It is to be hoped that these standards will define the link between the error in data and the injury sustained as a result of that error, in either physical or economic terms. The need will then be for the creation of a new category of law which will only be met through manufacturers, suppliers and value adders assuming a duty of care and taking reasonable steps to avert liability in tandem with the legislature developing statutory rules and regulations for ethical conduct.
[*] Kerrin Stewart is a legal researcher in her penultimate year of Law, George Cho is Associate Professor of Geographic Information Systems and the Law and Eugene Clark is Professor of Law and Head of School. All three authors are at the University of Canberra as part of a GIS research team. We would also like to thank and acknowledge the assistance of other research team members, Kate Reid, Lecturer and Researcher, and Arthur Hoyle, Lecturer in Law and PhD candidate also at the University of Canberra.
[1] French for the word ‘wrong’, civil wrongs (a departure from what the law determines should happen) as distinguished from criminal wrongs.
[2] For the purposes of this article GIS products includes the provision of services.
[3] C. Reed, (ed) Computer Law, Blackstone Press Limited, London 1990, p56.
[4] G. Cho, Geographic Information Systems and the Law, London: J. Wiley (forthcoming 1997). p. 7
[5] P. Samuelson, ‘Liability for defective electronic information’, Communications of the ACM, Vol 36(1) Jan 1993, pp21-26.
[6] H. Onsrud & R. Reis, ‘Law and Information Policy for Spatial Databases: A research Agenda’. URL:http://www.spatial.maine.edu/tempe/onsrud_2.html
[7] K. Matsunaga & J. Dangermond, ‘Promoting a Free Access or Minimal Cost of Dissemination Arrangement for Government-Held Geographic Information Systems Data’. URL:http://www.spatial.maine.edu/tempe/matsunaga.html
[8] J.E. Innes & D.M. Simpson, ‘Implementing GIS for planning: Lessons from the history of technological innovation’, Journal of the American Planning Association, Vol 59(2) Spring 1993, pp230-236.
[9] H. Onsrud & R. Reis, op cit, pp377-393.
[10] M. Monmonier, ‘A Case Study in the Misuse of GIS: Siting a Low-Level Radioactive Waste Disposal Facility in New York State’. URL:http://www.spatial.edu/tempe/monmonier.html
[11] R. Nimmer & P. A. Krauthaus, ‘Computer Error and User Liability Risk’, Jurimetrics Journal, Vol 26(2) Winter 1986, pp122-123.
[12] M. Juppenlatz & T. Ziaofeng, Geographic Information Systems and Remote Sensing, McGraw Hill Book Company, Sydney 1996, p96.
[13] G. Cho, op cit, p122.
[14] Arising from the detrimental reliance by the users of the information provided.
[15] E.F. Epstein, ‘Legal aspects of GIS’ Geographic Information Systems: Principles and Applications, D.J. Maguire, M.F. Goodchild, and D.W. Rhind (eds), Longman Scientific and Technical, London 1991, pp489-502.
[16] H.H. Perritt, Law and The Information Super Highway, John Wiley & Sons Inc., N.Y. 1996, p162.
[17] D. Rhind, ‘Data access, charging and copyright and their implications for geographical information systems’, International Journal of Geographical Information Systems, Vol.6 No.1, 1992, p13.
[18] H. Archer & P.L. Croswell, ‘Public Access to Geographic Information Systems: An emerging legal issue’, Photogrammetic Engineering and Remote Sensing, Vol 55, No.11, Nov 1989, p1575.
[19] R. Nimmer & P.A. Krauthaus,op cit, p121.
[20] Mason v Montgomery Data Inc 976 F. 2d 135 (5th Cir.1992).
[21] S. Chalton, ‘An Introduction to the Legal Liabilities of Information Producers’ in C. Edwards, N. Savage & I. Walden (eds), Information Technology & The Law 2nd Edition, Macmillan Publishers Ltd, UK 1990.
[22] E.C. Karnow, ‘Liability for distributed artificial intelligences’, High Technology Law Journal Issue 11:1 - Spring 1996. URL:http://server.berkeley.edu/BTLJ/abstract/111mell.html.
[23] R. Nimmer & P.A. Krauthaus, op cit, p123.
[24] Particularly in relation to concepts of liability involving standards of care and degrees of risk.
[25] B.L. Horovitz, ‘Computer software as a good under the Uniform Commercial Code: Taking a byte out of the intangibility myth’, Boston University Law Review, Vol 65 No 1, January 1985, p136.
[26] [1972] USCA10 15; 453 F.2d 533 (10th Cir.1972).
[27] Chatlos Systems, Inc. v National Cash Register Corp. 479 F.Supp 738 (D.N.J. 1979) aff’d[1980] USCA3 831; , 635 F.2d 1081 (3d Cir. 1980).
[28] Local governments in Alaska, Iowa, Connecticut and Oregon permit the charging of fees for information based products and services. J.R. Anderson, ‘Local Government Liability for Erroneous Data: Law and Policy in a changing environment’. URL:http://www.spatial.maine.edu/tempe/anderson.html.
[29] Alaska Statute 09.25.115 (d) (2).
[30] Australian Surveying and Land Information Group; Specialising in ‘surveying, mapping, information systems and consultancy services’. URL:http://www.auslig.gov.au/welcome.html.
[31] B. Bordoloi, K. Mykytyn, and P. Mykytyn., ‘A framework to limit systems developers’ legal liabilities’, Journal of Management Information Systems, Vol 12 (4), Spring 1996, pp161-185.
[32] H.J. Onsrud, Identifying Unethical Conduct in the Use of GIS. URL:http://www.spatial.maine.edu/Identifying_Ethics_GIS.html.
[33] E.F. Epstein, op cit, p496.
[34] Although persons not apparently parties to the contract may too acquire rights e.g. a beneficiary or an undisclosed principal of an agent.
[35] S. Chalton, op cit, p5.
[36] Only parties to the contract have legal rights and liabilities under that contract.
[37] In Jaensch v Coffey [1984] HCA 52; (1984) 58 ALJR 426, Deane J. emphasised Lord Aitkin’s test for liability in negligence, of the “neighbour” being one who is closely and directly affected by an act.
[38] T. Simmons, ‘Spatial Data Analysis in the Formation of Public Policy and its Acceptance as Evidence: A Litigator’s Perspective on Geographic Information and Analysis’. URL:http://www.spatial.maine.edu/tempe/simmons.html.
[39] ibid, p.
[40] For example, in Australia, The Trade Practices Act 1974 (Cth) governs the sale of goods and in the U.S. both the Restatement (Second) of Torts and Uniform Commercial Code.
[41] Compensatory awarded returns the victim to their original economic position; personal damage is purely for personal injury; and, exemplary damage is sometimes awarded in addition to compensatory damages, where the tort was deliberate, outrageous or intentional with blatant disregard for any consequences.
[42] The general principle is that loss or harm will be attributed to the most immediate effective causal factor rather to something earlier, though that may have been an essential prerequisite of the loss or harm.
[43] D. Walker, The Oxford Companion to Law, Clarendon Press, London 1980, p1011.
[44] M. Davies, Torts Second Edition, Butterworths Tutorial Series, Butterworths Australia 1995, p57
[45] [1943] AC p32
[46] Independent School District No 454 v Statistical Tabulating Corporation 359 F Supp 1095 (1973).
[47] URL:http://www.spatial.maine.edu/tempe/johnson.html
[48] Vandermark v Ford Motor Co., 391 P.2d 168, 172 (Cal. 1964).
[49] L. Milrad, ‘Database Ownership, Canadian Copyright Principles and Multiparticipant GIS Projects’, GIS Law v. 2 (1994) p10.
[50] M. Davies, op cit, p3.
[51] Donoghue v Stevenson [1932] A.C. 562.
[52] Hedley Byrne & Co Ltd v Heller & Partners Ltd [1963] UKHL 4; [1964] AC 465.
[53] Ibid. p. 537.
[54] Vacwell Engineering v BDH Chemicals Ltd [1971] 1 Q.B. 88.
[55] G. Cho, op cit, p112.
[56] [1990] UKHL 2; [1990] 2 W.L.R. 358.
[57] Donoghue v Stevenson [1932] A.C. 562.
[58] S. Chalton, op cit, p16.
[59] Strict liability in U.S. law does not award damages purely for economic injury, the injury must present a physical risk to persons or property. Economic losses in the US by and large are covered under the law of warranty and contract.
[60] But note that foreseeable economic injury in the U.S. is only available to persons with whom the actor has a special relationship.
[61] To mitigate is to, where it is possible, minimise the impact of further injury.
[62] Refers to contributions by the victim resulting in their own harm.
[63] A person who voluntarily assumes risk cannot claim negligence.
[64] E.F. Epstein, op cit, p497.
[65] E.F. Epstein. & H. Roitman, ‘Liability for Information’, Papers from the Annual Conference of the Urban and Regional Information Systems Association, 1987, p367.
[66] ibid, p366.
[67] S. Chalton, op cit, p16.
[68] S. Nycum & W. Lowell, ‘Common Law and Statutory Liability for Inaccurate Computer-based Data’, The Emory Law Journal, Vol 30, Emory University School of Law Atlanta Georgia, 1981, p454.
[69] In Australia there has yet to be any reported Commonwealth or state case law in relation to GIS and liability.
[70] H.H. Perritt, op cit.
[71] In Australia the Trade Practices Act 1974 confers this duty of care to producers.
[72] E.F. Epstein & H. Roitman, op cit, p365.
[73] T.J. Smedinghoff (ed), Online Law, Addison-Wesley Developers Press, U.S. 1996, p449.
[74] M. Davies, op cit, p10.
[75] Kitto J in Blomley v Ryan [1956] HCA 81; (1956) 99 CLR 362, defined this as being where one party is at a serious disadvantage and the other party ‘unconscientiously’ takes advantage.
[76] R. Miller, Annotated Trade Practices Act 1974, 15 Ed, Australia 1994, p235.
[77] [1985] HCA 65; (1985) 158 CLR 661.
[78] [1988] FCA 289; (1988) 82 ALR 415.
[79] R. Miller, op cit.
[80] Damages likely to be suffered are found under s87.
[81] W. Covell & K. Lupton, Principles of Remedies, Butterworths, Australia 1995, p256.
[82] Mitigating the damages means the plaintiff took reasonable steps, where possible, to reduce the loss.
[84] A tort is a wrong for which the injured person has a right to recover damages under the common law.
[85] [1976] HCA 65; (1976) 136 CLR 529. Proximity in this case relied on the plaintiff and the third party being involved in a common adventure, not physical proximity.
[86] W. Covell & K. Lupton, op cit, p25.
[87] On 29 Nov 1990 the Occupational Liability Bill 1990 was introduced in order to limit the civil liability of professionals and others, due to the increased cost of insurance in the climate of increased litigation. Although this Bill lapsed a further draft was developed in 1992 namely the Professional Standards Bill 1992.
[88] In his media statement of 6 November 1991.
[89] M. Mills, ‘Lessons from America: Professional Liability and Tort Reform’, The Australian Bar Review, Vol 12, 20 December 1994, p3.
[90] Winter v G.P.Putam’s Sons [1991] USCA9 605; 938 F.2d 1033 (9th Cir.1991).
[91] The US Court in Winter questioned whether computer software was a good or a service for the purposes of products liability, and what if any contractual liability could be incurred by manufacturers and suppliers of software. A good is subject to the Uniform Commercial Code whilst services are not, however the problem lies in the tangible and intangible nature of the properties which are not able to be categorised.
[92] This decision was supported by reference to the Restatement (Second) of Torts 402A cmt. d (1965).
[93] M. Liberman, Comment ‘Overreaching Provisions in Software Licence Agreements’, RICH.J.L&TECH 4 (1995). URL:http://www.urich.edu/~jolt/vlil/liberman.html.
[94] A.R. Field, and Z. Schiller, ‘Electronic Data could make trouble for the Law’, Business Week (2970) Industrial/Technology Edition, October 27 1986, pp128 –132.
[95] 776 F. Supp. 135 (S.D.N.Y. 1991).
[96] H.H. Perritt, op cit, p165.
[97] 480 U.S.
[98] T.J. Smedinghoff (ed), Online Law, Addison-Wesley Developers Press, U.S. 1996, p370.
[99] ibid, p463.
[100] D. Rhind, op cit, p27.
[101] G. Cho, op cit, p124.
[102] Zinn v State 112 Wis. 2nd 417, 334 N.W. 2nd 67 [1983].
[103] For the purposes of liability the analogy of a defective compass was used.
[104] [1985] USCA9 252; 753 F.2d 794, 800 (9th Cir. 1985).
[105] Surveyors or cartographers.
[106] L. Milrad, op cit, p10.
[107] D. Abney, ‘Liability for Defective Aeronautical Charts’, Journal of Products Liability, Vol 9 (3), 1986, pp211-231.
[108] [1985] USCA9 1290; 767 F.2d 1288 (9th Cir. 1985), cert.den., 474 U.S. 1101 (1986).
[109] FAA approach procedures were essentially compilations of data. Jeppesen took this data to portray the instrument approach procedure in graphic form.
[110] Regima v United States 448 F.Supp. 45 [1978] (W.D. Mich.).
[111] [1981] USCA9 479; 642 F.2d 339 (1981).
[112] M.J. Dragich, ‘Information Malpractice: Some thoughts on the Potential Liability of Information Professionals’, Information Technology and Libraries, 1989, p266.
[113] L.A. Weber, ‘Bad Bytes: The application of Strict Products Liability to Computer Software’, St Johns Law Review, Vol 66 (2) Spring, St John’s University School of Law, 1992, pp 476-478.
[114] In the United States it must also be viewed in terms of how it affects the protections afforded by the First Amendment of the Constitution.
[115] H.J. Onsrud, J.P. Johnson, & X.R.Lopez, ‘Protecting Personal Privacy in Using Geographic Information Systems’, Photogrammetric Engineering and Remote Sensing, Vol LX, No.9, September 1994, p1083.
[116] J.R. Anderson, op cit.
[117] G. Cho, op cit, p61.
[118] S. Chalton, op cit, pp35-36.
[119] F.J., ‘Legal Exposures facing the software industry’, 25 September 1996. URL:http://www.lectlaw.com/files/bul17
[120] M. Trebilcock, ‘The Future of Tort Law: Mapping the contours of the Debate’, Canadian Business Law Journal, Sept 1989; Vol 15 (4), pp471-488.
[121] Welkener v Kirkwood Drug Store Co., 734 S.W.2d 233 242 (Mo. Ct. App. 1987).
[122] H.B. Dansby, ‘Public Records and Governmental Liability’, GIS Law, v. 2 (1992), p9. This general impunity was in-built into state, federal and local government levels to give protection to actions in tort or contract.
[123] S. Strater, ‘Developing a GIS Database’, American City and Country, April 1992; v107(4), p16.
[124] L. Lang, ‘Making a GIS Come True’ Planning, Vol 56(7), July 1990,pp14-20.
[125] E.F. Epstein, and H. Roitman, op cit, p367.
[126] D. Rhind, ‘Spatial Databases and Information Policy: A British Perspective’. URL:http://www.spatial.maine.edu/tempe/rhind.html.
[127] J.T. Westermeier, ‘Personnel Licensing and Liability’ in Computers and the Law an Introductory Handbook Third Edition, R. Bigelow (ed), Commerce Clearing House Inc, Chicago 1981, p151.
AustLII:
Copyright Policy
|
Disclaimers
|
Privacy Policy
|
Feedback
URL: http://www.austlii.edu.au/au/journals/JlLawInfoSci/1997/5.html