Craign argued that the iiNet case was one where the ICT infrastructure was separate from the content carried. He suggested that where there is more processing of the data by an application on the network, the IT professional who designs the network has more responsibility for the content. That is a view I agree with, but I disagree with Craig on the extent of the responsibility of the IT professional.
The IT professional cannot become an expert on all things and a dictator to all people. The extent of the IT professionals responsibility depends on the nature of the system and the extent of the relevance of their expertise. The IT professional needs to design the system well, so it works with an adequate level of reliability. But it is not possible to design perfect systems and the level of reliability depends on the application.
In the case of a content management system, records management system, or a user interface design, the IT professional can be expected to know a reasonable amount about how such a system should work. So they have a large part of the responsibility for that system working correctly. In the case of a medical system, they could be expected to have less subject matter expertise.
As an expert witness I am called on to advise courts and other bodies on if a failed system should have worked. Such cases typically happen years after the system was built and I need to put myself in the mindset at the time. It is not enough to say "now we know what to do", I need to look at what was known and what was reasonable then. Also it is necessary to consider what should be the extend of the IT professional's knowledge and role.
An IT professional cannot simply assume that providing accurate information, or more information is better. At the Australian Bureau of Statistics, part of my job was to make sure that some of the information provided was inaccurate by introducing random errors. There are so few individuals in some categories of data the ABS provides that individuals could be identified from the data. To protect privacy, this data has to be modified. In this case providing correct and accurate information is the wrong thing to do.
A similar example is the case of Synthetic vision systems for aircraft. These show an artificial view of the terrain, generated from mapping data. It might be assumed that providing a more detailed view, using more data, would be better for the pilot. But more information makes the image hard to interpret. In this case more accurate data is harmful.
In 2000 I gave evidence in the case of Bruce Lindsay Maguire v. Sydney Organising Committee for the Olympic Games in the Human Rights & Equal Opportunities Commission. My role was to assess if the website for Sydney 2000 Olympics was accessible to someone who is blind. My conclusion was that it was not, which the commission agreed with. However, it was decided that the Sydney Organizing Committee for the Olympic Games (SOCOG), a NSW statutory corporation, were responsible, not IBM who designed the web site.
Following the SOCOG case I was asked by the Australian National University to help teach web design. I decided that accessible web design should be a basic part of such training and incorporated that in the course. Due to such courses and advice by HREOC, it would be difficult today for an IT professional to argue it was not part of their responsibility to make web sites accessible. An IT professional who claims expertise in web site design can be assumed to know about accessible design and know it is part of their job to implement this.