(Sorry - this is not a use case but I didn't know where else to put it.) I just came across an essay ( http://www.acm.org/pubs/cacm/AUG96/antimac.htm ) which touches on this issue and puts the point quite well, I think. In particular, we know that command line interfaces as they stand are far from being perfect, but we can imagine something that combines the fexibility and power of language based interaction with the ability to prompt and give options and assistance when they are required. The idea of 'negotiating a query' (see below) is quite intriguing.

[begin quote] Over the past million years, humans have evolved language as our major communication mode. Language lets us refer to things not immediately present, reason about potential actions, and use conditionals and other concepts not available with a see-and-point interface. Another important property of language missing in graphical interfaces is the ability to encapsulate complex groups of objects or actions and refer to them with a single name. An interface that can better exploit human language will be both more natural and more powerful. Finally, natural languages can cope with ambiguity and fuzzy categories. Adding the ability to deal with imprecise language to the computer interface will increase the computer's flexibility and its fit with people's normal expressive modes. As Susan Brennan has shown [4], natural language, in addition to being natural for people, has several advantages as a medium for human-computer interaction, including the role of negotiated understanding, use of shared context, and easy integration with pointing and other input/output channels.

We are not proposing, however, what AI researchers would call a "natural language interface." [...] Instead, we have in mind something more like the interfaces of text-based adventure games, with their understanding of synonyms, relatively simple syntax, and tolerance for error in the input----a pidgin language for computers. The critical research question is, "How can we capture many of the advantages of natural language input without having to solve the 'AI-Complete' problem of natural language understanding?" Command line interfaces have some of the advantages of language, such as the large number of commands always available to the user and the rich syntactic structures that can be used to form complex commands. But command line interfaces have two major problems. First, although the user can type anything, the computer can understand only a limited number of commands and there is no easy way for the user to discover which commands will be understood. Second, the command line interface is very rigid and cannot tolerate synonyms, misspellings, or imperfect grammar. We believe that both these deficiencies can be dealt with through a process of negotiation. Think of the way a new library user might interact with a reference librarian. A librarian who had a command line interface would understand only a limited number of grammatically perfect queries, and the novice user would have to consult an obscure reference manual to learn which queries to write out. A reference librarian with a WIMP interface would have a set of menus on his or her desktop; the user would search the menus and point to the appropriate query. Neither interface seems very helpful. Instead, real reference librarians talk with the user for a while to negotiate the actual query. Similarly, we envision a computer interface that utilizes a thesaurus, spelling correction, displays of what is possible, and knowledge of the user and the task to take part in a negotiation of the user's command. We could imagine, for example, a dialog in which the user makes a free-form request, the computer responds with a list of possible tasks that seem to match the request, and both engage in a dialog to focus on the request the user actually intended. [end quote]

-- SamerAbdallah - 02 Mar 2007

-- BenFields - 21 Mar 2007