My own perception is that those in CS/AI/AGI have made a subtle error that rightfully draws such derision. I will (try to) explain...
I have watched how over the last half century, many people have claimed to have demonstrated programs capable of (eventually with enough work) "understanding" NL. Similarly, I have seen things like OpenCog that, to an untrained eye, seem to parallel Ben's own early outlandish statements of how close AGI is to becoming a reality, only to reach a certain point and stall. After years of battling NL I **FINALLY** figured out what was blunting everyone's efforts and found a way past it, which of course does NOT mean there are no further barriers. Still there remain people here on this forum who are proceeding with NL projects without squarely addressing the issue I identified. >From postings of the last year, it appears that the light is starting to dawn in Ben's mind as to the true nature of the barriers in AGI that are now standing in his way. >From an engineering viewpoint, the next obvious step would be for Ben (and others who have invested at least a year of full-time-equivalent effort in AGI) to carefully document the barrier(s) that have been blunting their efforts. Perhaps/hopefully in this process they might see ways past their present barriers. The problem is that CS derides AGI because of the perceived nature of people working in the field, INSTEAD of their lack of sound engineering practice. It takes a certain kind of person to go through life while avoiding sound engineering practice, and the article seems to describe such people pretty well. Nonetheless, I suspect efforts to understand the barriers to fail due to POV issues, e.g. such computations may be (and I think they probably are) equilibrium in nature, i.e. computing the equilibrium point of many different "forces". This is easy to do with what I have labeled as "bidirectional computation" and possible to do with conventional computation, but the application of "conventional" methods is EXTREMELY counter-intuitive. Perhaps errant POV is the REAL barrier to AGI? Thoughts? Steve On Fri, Oct 24, 2014 at 4:43 AM, Jim Bromer via AGI <[email protected]> wrote: > I am happy to be on the fringes of science (and/or science > philosophy). I wish I made a little money at it and got to talk to > more people who were able to express constructive views about systems > directly related to AGI. I don't have any greater respect for people > who work within 'established' groups than I do for those of us who are > more on the fringe. Unfortunately I have again become one of those > people who think they can see the issues more clearly than almost > anyone else so I just am skeptical of any group who claims to be > working on AGI or strong AI. My only redeeming personal quality is > that I clearly see myself on the fringes of understanding AGI so I am > not one of those who goes around claiming that he has got it figured > out. But when it comes down to it we have to do something that can > stand clearly as an advancement in order for us to really get over the > pretentious status jockeying that goes on. Even if our advancement is > incremental and no one else really appreciates a true advancement is > something that doesn't just appear in vaporware. It either has to be > an actual program or it has to lead directly to a program that makes > some actual advancement. > Jim Bromer > > > On Tue, Oct 21, 2014 at 4:49 PM, Mike Archbold via AGI <[email protected]> > wrote: > > This is one of those summary of AI progress articles. The only thing > > I object to about it is this: > > > > "AI still attracts oddballs, lone wolves working in their basements 10 > > hours a week hoping to solve the AI problem once and for all." > > > > The hell with 10 hours a week!!! 50 hours and you might get > > someplace. More progress has to be made in the perception here. > > There isn't any working strong AI, AGI, so how do you single out an > > oddball? It seems to me an oddball stands exactly the same chance as > > established.... > > > > oh, I said it, the "established" word. I always stop talking or > > reading when somebody says that. God forbid, if AI ever becomes > > established. I take nobody as an authority in strong AI. Except > > maybe Aristotle. > > > > Mike A > > > > > > ------------------------------------------- > > AGI > > Archives: https://www.listbox.com/member/archive/303/=now > > RSS Feed: > https://www.listbox.com/member/archive/rss/303/24379807-653794b5 > > Modify Your Subscription: https://www.listbox.com/member/?&; > > Powered by Listbox: http://www.listbox.com > > > ------------------------------------------- > AGI > Archives: https://www.listbox.com/member/archive/303/=now > RSS Feed: https://www.listbox.com/member/archive/rss/303/10443978-6f4c28ac > Modify Your Subscription: > https://www.listbox.com/member/?&; > Powered by Listbox: http://www.listbox.com > -- Full employment can be had with the stoke of a pen. Simply institute a six hour workday. That will easily create enough new jobs to bring back full employment. ------------------------------------------- AGI Archives: https://www.listbox.com/member/archive/303/=now RSS Feed: https://www.listbox.com/member/archive/rss/303/21088071-f452e424 Modify Your Subscription: https://www.listbox.com/member/?member_id=21088071&id_secret=21088071-58d57657 Powered by Listbox: http://www.listbox.com
