Hi,
Cliff Stabbert wrote: > In other words, in the case of a word processing application the > package would come with the "basic concepts" of document commands > (creating, opening, editing saving), editing actions (typing, > inserting, deleting, moving), formatting actions, and document > contents (chapters, TOCs, paragraphs, words). (These concepts having > some sort of "referents", i.e. not being hollow phrases.) > > Where a current user would need to write a "macro", a user of this > package would instead communicate with the package about the word > processing program in terms of those abstractions. "Copy this list of > bullets and make each one a new chapter heading" doesn't require a > sophisticated user model, nor that sophisticated a model of the word > processing software... I agree. This would be a highly nontrivial software application to create, but it could be done by pushing the limit of standard narrow-Ai methods. > Where I see the proto-AGI elements coming in is in allowing the > language the user uses to instruct the computer to evolve, and in > interpreting that language, which requires some sophistication in > applying analogies. An example conversation: > > "What I just did is "expanding" a list of bullets. Now expand this > list." (gestures with mouse) > > "Now expand this paragraph." (I envision a CopyCat-like exploration of > possible ways to apply the "expand" concept to a paragraph arriving at > the ~90% strong conclusion that the sentences of a paragraph most > closely correspond to each bullet in a bulleted list.) > > " "Collapse" is the opposite of "expand". Now collapse chapters 3-7." > > The above to me seems reasonably doable; Hmmm.... Thanks for these examples! I understand what you're thinking of a lot better now... I think if you eliminate the NLP aspect, then you may indeed have something that could be done using a proto-AGI system in a reasonable time-frame. And it would be a damn useful thing -- I'd definitely use it. Not for e-mailing or browsing so much, but definitely for word processing, which I do a lot of. There are a lot of semi-repetitive actions in word processing, which are a little too loosely defined to put in macros without a lot of work, but could be explained to any human (except some really dumb secretaries I've had ;-p) very very simply. Spreadsheet operations are another excellent example. However, including the NLP aspects will make it a LOT harder, because the app would require pretty sophisticated NLP understanding, or it will be incredibly annoying to phrase things in ways the system understands. [And I think that NLP comes quite late along the AGI path, because an AGI has to learn natural language thru its experience, not through hacking parsing rules, linguistic DB's and so forth into it.] The trick is then to come up with a really user-friendly way for the user to communicate with the system in a flexible, evocative way without requiring sophisticated NLP. I guess this is what will make or break the product idea... I guess that if you came up with a user-friendly non-NLP communication scheme, and made a spiffy non-functional demo showing "what the product would do if it were completed", it could be possible to get funding to create such a thing. Definitely there is promise here... -- Ben G ------- To unsubscribe, change your address, or temporarily deactivate your subscription, please go to http://v2.listbox.com/member/
