Configuring a chatbot with remote language processing

1. A computer implemented method of generating a chatbot application, comprising:

generating a designer user interface with a visual authoring canvas that shows visual elements identifying a logical flow of processing in a chatbot application being designed and that shows linguistic elements output by the chatbot application during operation and linguistic elements expected by the chatbot application, from a bot end user, during operation;

detecting an input, on the designer user interface, identifying a language processing system that is to be accessed by the chatbot application during operation;

detecting an input that changes a state of the chatbot application from a previous state to state N;

storing a representation of the chatbot application, in state N, on an undo stack; and

configuring the chatbot application to access the language processing system during operation.

2. The computer implemented method of claim 1 wherein detecting the author input, on the designer user interface, comprises:

detecting the author input identifying a natural language understanding system the chatbot application is to use during operation of the chatbot application.

3. The computer implemented method of claim 2 wherein the chatbot application is generated to perform actions to run a dialog and wherein detecting the author input identifying a natural language understanding system comprises:

selecting a visual element on the visual authoring canvas identifying an action for which natural language understanding is performed;

displaying a natural language understanding system selection mechanism on a property configuration pane corresponding to the selected visual element; and

detecting author actuation of the natural language understanding system selection mechanism indicating a selected natural language understanding system.

4. The computer implemented method of claim 1 wherein detecting the author input, on the designer user interface, comprises:

detecting the author input identifying a regular expression recognition system the chatbot application is to use during operation of the chatbot application.

5. The computer implemented method of claim 1 wherein generating the designer user interface comprises:

displaying an undo actuator and a redo actuator on the designer user interface.

6. The computer implemented method of claim 1 and further comprising:

detecting an authoring input that changes the state of the chatbot application from state N to state N+1;

detecting author actuation of the undo actuator; and

storing a representation of the chatbot application in state N+1 on a redo stack.

7. The computer implemented method of claim 6 and further comprising:

retrieving the representation of the chatbot application in state N from the undo stack; and

populating the visual authoring canvas on the designer user interface with state N of the chatbot application.

8. The computer implemented method of claim 7 and further comprising:

detecting author actuation of the redo actuator; and

storing the representation of the chatbot application in state N on the undo stack.

9. The computer implemented method of claim 8 and further comprising:

retrieving the representation of the chatbot application in state N+1 from the redo stack; and

populating the visual authoring canvas on the designer user interface with state N+1 of the chatbot application.

10. The computer implemented method of claim 9 wherein generating the designer user interface includes displaying a property pane and displaying a serialization pane, the property pane having author actuatable input mechanisms that are actuatable to configure an action corresponding to a visual element selected on the visual authoring canvas, the serialization pane displaying a serialized representation of the visual element selected on the visual authoring canvas, and further comprising:

populating the property pane and the serialization pane on the designer user interface with state N+1 of the chatbot application.

11. The computer implemented method of claim 7 wherein generating the designer user interface includes displaying a property pane and displaying a serialization pane, the property pane having author actuatable input mechanisms that are actuatable to configure an action corresponding to a visual element selected on the visual authoring canvas, the serialization pane displaying a serialized representation of the visual element selected on the visual authoring canvas, and further comprising:

populating the property pane and the serialization pane on the designer user interface with state N of the chatbot application.

12. A computer system, comprising:

one or more processors; and

memory including an undo stack and a redo stack and storing instructions which, when executed by the one or more processors, cause the one or more processors to perform steps of:

generating a designer user interface with a visual authoring canvas, that shows visual elements identifying a logical flow of processing in a chatbot application being designed, and a property pane having author actuatable input mechanisms that are actuatable to configure an action corresponding to a visual element selected on the visual authoring canvas, the designer user interface including an undo actuator and a redo actuator;

detecting an authoring input on the designer user interface that changes a state of the chatbot application from a previous state to state N;

storing a representation of the chatbot application, in state N, on the undo stack;

detecting an authoring input on the designer user interface that changes the state of the chatbot application from state N to state N+1;

detecting author actuation of the undo actuator;

storing a representation of the chatbot application in state N+1 on the redo stack;

retrieving the representation of the chatbot application in state N from the undo stack; and

populating the visual authoring canvas and the property pane on the designer user interface with state N of the chatbot application.

13. The computer system of claim 12 and further comprising:

detecting author actuation of the redo actuator;

storing the representation of the chatbot application in state N on the undo stack;

retrieving the representation of the chatbot application in state N+1 from the redo stack; and

populating the visual authoring canvas and the property pane on the designer user interface with state N+1 of the chatbot application.

14. The computer system of claim 13 wherein generating the designer user interface includes displaying a serialization pane that displays a serialized representation of the visual element selected on the visual authoring canvas, and further comprising:

populating the serialization pane on the designer user interface with state N+1 of the chatbot application.

15. The computer system of claim 12 wherein generating the designer user interface includes displaying a serialization pane that displays a serialized representation of the visual element selected on the visual authoring canvas, and further comprising:

populating the serialization pane on the designer user interface with state N of the chatbot application.

16. The computer system of claim 12 wherein the designer user interface shows linguistic elements output by the chatbot application during operation or expected by the chatbot application, from a bot end user, during operation, and further comprising:

detecting an author input, on the designer user interface, identifying a language processing system that is external to the chatbot application and that is to be accessed by the chatbot application during operation; and

configuring the chatbot application to access the language processing system during operation.

17. A computer system, comprising:

one or more processors; and

memory storing instructions which, when executed by the one or more processors, cause the one or more processors to perform steps of:

generating a designer user interface with a visual authoring canvas that shows visual elements identifying a logical flow of processing in a chatbot application being designed and that shows linguistic elements output by the chatbot application during operation and linguistic elements expected by the chatbot application, from a bot end user, during operation, wherein the chatbot application is generated to perform actions to run a dialog;

based on detecting an author input on the designer user interface, selecting a visual element on the visual authoring canvas identifying an action for which natural language processing is performed;

displaying a natural language processing system selection mechanism on a property configuration pane corresponding to the selected visual element;

based on detecting actuation of the natural language processing system selection mechanism, identifying a natural language processing system that is to be accessed by the chatbot application during operation; and

configuring the chatbot application to access the natural language processing system when the chatbot application is operating.

18. The computer system of claim 17 wherein detecting actuation of the natural language processing system selection mechanism, comprises:

detecting the author input identifying a regular expression recognition system the chatbot application is to use during operation of the chatbot application.