About OpenCog, you say > The gap between present technology and the target technology is so vast - > essentially it is being claimed that an entire independent machine *system* > can be built, when there is no such thing as an independent machine ystem - > and there is no explanation of how this can/will be achieved, starting from > highly dependent hyperspecialist programs. > > This illustrates that you can apparently have the odd specific target, and > yet still be wildly unrealistic. > > More particularly, it shows that you have to present not just specific > targets, but a realistic idea of how you are going to achieve them.
There is a detailed explanation, in a 1000-page technical book manuscript. However, whether the design described there, and the teaching methodology described there, will actually work -- is admittedly not proven. There is no ironclad mathematical argument that OpenCog will work as hoped. The preliminary results, as of now, are very primitive (though subsystems of OpenCog have done rather good narrow-AI stuff on their own) ... which is expectable since only a fraction of the system is implemented So, the workability of OpenCog as a powerful AGI is a bit of an intuitive leap at the moment, admittedly That's how progress happens. People place their bets on various intuitive leaps, spend their time accordingly, and some are right and some are wrong... -- Ben G ------------------------------------------- AGI Archives: https://www.listbox.com/member/archive/303/=now RSS Feed: https://www.listbox.com/member/archive/rss/303/21088071-c97d2393 Modify Your Subscription: https://www.listbox.com/member/?member_id=21088071&id_secret=21088071-2484a968 Powered by Listbox: http://www.listbox.com
