There was a recent change (confirmed with support) from listing the most recently created intent at the top of the list when creating the trigger for a dialog node, to expecting user to start typing the name of the intent and then your "type ahead" feature would narrow down the list of intents for you to choose from.
This change creates a problem with longer intent names and also when there are hundreds of intents to choose from. In my situation it is a legal chatbot and the client has decided to put the intent category as the first part of the intent name so during reporting they can see how many intents of a certain category are seeing action. So with legal content it sometimes deals with new laws and acts which sometimes have long multi word names. So the intent name begins with the category and then continues to mention a longer legal act, it can push the intent name into having 50+ characters. So now when I'm adding a dialog node for these longer intents, because the most recently added intent is not at the top of the list, I have to first remember the exact name of these 50+ character intent names and then start typing them out "hopefully" not making any typos as I'm doing it so the "type ahead" feature eventually narrows the list down to the one I'm after.
I don't see how the "type ahead" feature excludes the possibility of still having the most recently added intent at the top of the list as there is still a list that shows up initially until you start typing.
If you don't want to have the most recently added intent at the top of the list, then please create an option to turn this off and on and just use the type ahead feature.
Why is it useful?
|Who would benefit from this IDEA?||The Watson Asssistant developer spends less time creating dialog nodes as they don't have to type 50+ characters to eventually get to the intent they need.|
How should it work?