Back to Trust Metrics

I think I've come to the conclusion that there isn't a good way to handle trust, only a least bad way. In this case, I can import a certain baseline trust level from journal articles--well, I assume I can at least, since I haven't found an automatic way to do so (at least, automatic and free.) But even assuming that I could use the number of articles and citations to create an 'expertise level' (or maybe a 'notoriety level') that still wouldn't speak to what area that level is valid in. I could have three dozen highly cited articles in string theory, but that doesn't mean I can claim expertise in civil engineering. It would be even worse to not even make that claim myself, but to have an automated system claim it on my behalf when I have no way to override.

Obviously, while a system based on the degrees somebody has obtained might be technically accurate, it would not always (or often?) represent a person's experiences after graduation. It seems that some level of self-classification will be necessary...but this self-classification will have to be on a per-post basis. Maybe radio buttons:

• I am an expert on the topic I discuss in this post
• I am familiar with this field
• I have no specialized knowledge of this field

Which I find a little worrisome, but it may not be that horrific given that everybody's credentials will be exposed. Maybe a meta-mod system would be helpful for this...