The Theory

In Chapter 1 I discussed the current theories used for informed consent in the medical and research fields, comprising theories based on autonomy such as Faden and Beauchamp's autonomous authorisation and effective consent models, or the general autonomy-based theory of Appelbaum et al. I also looked at the alternative of Manson and O'Neill's informed consent as a waiver of normative expectations. I identified that the approach used for informed consent in information technology is more closely aligned to the Faden and Beauchamp effective consent model than an autonomous authorisation approach or that of Appelbaum's. In Chapter 2 I outlined the problems found in information technology that have crept in with the use of the effective consent model, and in Chapter 3 I argued that these problems, along with the more general concerns identified by Manson and O'Neill about autonomy-based theories for informed consent, makes the informed consent as waiver approach much more attractive for information technology, since it has a much better scope for dealing with the specific problems encountered, specifically the content and display of the agreement, given the greater problem of ``numbness'' encountered in information technology situations.

As previously established (in section 1.2.2), the effective consent model involves sets of policy rules that aim to regulate the consent-seeker in order to protect the consenter. It does this by utilising disclosure and competence tests that conform to a legal or generally accepted standard framework suitable for the context. It often uses appeals to autonomy as its justification, but requires little evaluation of any sort of autonomy on the part of the consenter [Faden & Beauchamp, 1986]. This is particularly obvious in the case of End User License Agreements which, as I discussed in Chapter 2, default to a de facto standard that was developed by the same set of people who also commonly act as the consent-seekers, or are closely aligned with other consent-seekers, and thus are more interested in providing the bare minimum for legally sufficient informed consent so that they are protected from liability. In doing this, they present a somewhat bare-boned version of effective consent, centring on disclosure yet providing almost no competence testing.

This is, as shown in Chapter 3, an unacceptable approach for informed consent in End User License Agreements, since it offers almost no attempt to inform the user outside of simple disclosure, which as discussed, is presented in such a way that it is almost always ignored. This definitely benefits the consent-requester over the consenter, and shows the need for an overhauled informed consent system for End User License Agreements. On a theoretical level, it shows the need for re-thinking the approach to informed consent in information technology, since the situations found in IT are very different from those in the medical and research fields. Therefore, I adopted Manson and O'Neill's objections to the basis of informed consent on justifications of autonomy, and their revised theory that relies on the establishment of informed consent as a waiver of normative expectations.

Informed consent as waiver, as developed in Chapter 3, shows how the identification of normative expectations, coupled with a good communicative transaction framework, can improve informed consent procedures in computing situations. Choosing normative expectations needing to be waived restricts the scope of the informed consent procedure to the values considered important by the users of the software rather than the more abstract and idealised values of the autonomy-based theories. A good communicative transaction framework allows for more direct disclosure of information that is relevant and accurate, rather than a general full disclosure that could easily flood the user with too much information, resulting in a confused or ignorant decision rather than an informed one.

The practical implementations and improvements discussed in this chapter aim to not only provide an example for the application of the above theory, but also to demonstrate ways of overcoming developer and vendor apathy toward change. It does this by recommending a third party mediation group that handles the development of new End User License Agreements through an easy-to-use interactive interface. This group would provide a modular license builder for simple standard license agreements waiving common normative expectations. These expectations would be standardised into language that can be easily understood, yet legally enforceable. More detail about the recommendations is in section 4.3.