Trust in ability to express valuable trust statements

Many papers argue that one user should be able to express two kinds of trust in other users: Trust in characteristics and Trust in ability to express valuable trust statements.

Trust in ability to express valuable trust statements refers to how much a certain user values not the characteristics of the other user, but her ability to provide useful trust statements.

Actually it can be argued that it is not a wise choice to differentiate these two kind of trust and that it would be better to just rely on Trust in characteristics. The reasons are that: (1) explaining to a normal user these two kinds of trust and why she should enter it is very hard and (2) it is hard to convince users to enter two trust values when it is already very hard to get one trust value. Moreover it can be argued that in general these two values are very correlated so this difference would be useful just on a tiny minority of cases. One such example is captured by the Gandhi paradox.
 * I think the term trusting judgement is a good example when talking about people. I think it also significantly simplifies explaining this type of trust to the user. For example: John, how much do you trust Bob? John, how much do you trust Bob's judgement of other people? -- CallumTalk 17:38 (GMT+1), 24 Sep 2007

See also Trust in characteristics.