Two additional thoughts on Definition of User Agent

Two additional notes from today's discussion of the "proposal for adding 
test conditions to the Definition of User Agent":

1. We should include examples in the document to help readers understand 
the three categories and their criteria. Simon, can you post the list of 
UA and non-UA you've run through the criteria and which categories they 
fit into?

2. I suggest that even if the section on compliance includes detailed 
criteria, it should also state that the criteria should not be 
considered exhaustive because new technologies may not fit easily into 
existing categories. Thus, when evaluating products to determine whether 
they should comply with UAAG, even if a component does not meet the 
stated criteria it should still comply if its failure to do so could 
hamper users' ability to carry out their tasks. (This is following up on 
our discussion of general vs. specific guidelines from the 3 September 
2009 conference call.)

    Thanks,
    Greg