This example may not be ideal since this country selection has both "Congo" and "Congo, Democratic Republic of" in its data set, but the principle can probably be applied elsewhere.
This example may not be ideal since this country selection has both "Congo" and "Congo, Democratic Republic of" in its data set, but the principle can probably be applied elsewhere.
That is to say in your example, the parent wasn't questioning whether the code should support 'Congo', but whether it should support 'ong', for example.
Basically it should work like Command-T in Textmate (or the vim plugin based off of it). So if there's a Congo and a Republic of Congo in your potentials, then typing "con" should list Congo first as being a more likely result (even though in this case it may not be).
In this demo, when I type "ca" the first three results are "United States, American Samoa, and Antartica". I would argue that this is less intuitive to the user than a simple alphabetical dropdown.