The architecture of the more advanced design of the BoA style than the BoE style lends itself to a much higher learning curve. In BoE, you have one compact file all done with nodes. In BoA, you have lots of little files all with their own scripts. Although the BoA model is more desirable from a software designer standpoint, I feel it is less so for a general scenario design tool.
An appropriate compromise would have been something that would read in the relevant files as needed. Users could script with traditional BoE style nodes, but a more advanced user could customize the script. In the end, Blades of Avernum forces the designer to keep track of lots of little compartments and such. Blades of Exile's biggest strength, I feel, was its compactness and ability to write a scenario with very little effort.
I think something such as a Blades Editor could work in the future if attached to a much larger game. One way to implement this is to hard code in the game a maximum town number or zone the player could enter without registering. Additionally, only the default game would be accessible so user scenarios would not be playable.
To summarize, I see two problems with the BoA model:
1) Too spread out of an interface, needs to be compacted for ease of use. Theoretically, this could be done for the BoA Editor with much effort.
2) A suboptimal marketing strategy of one small playable scenario. Big games tend to make the bucks.