Well, we have attempted to use sound software engineering principles to architect the OpenCog framework, with a view toward making it usable for prototyping speculative AI ideas and ultimately building scalable, robust, mature AGI systems as well
But, we are fairly confident of our overall architecture with this system because there have been a number of predecessor systems based on similar principles, which we implemented and learned a lot from ... If one has a new AGI idea and wants to start experimenting with it, SE is basically a secondary matter ... the point is to explore the algorithms and ideas by whatever means is less time-wasting and frustrating... OTOH, if one has an AGI idea that's already been fleshed out a fair bit and one is ready to try to use it as the basis for a scalable, extensible system, SE is more worth paying attention to... Premature attention to engineering when one should be focusing on science is a risk, but so is ignoring engineering when one wants to build a scalable, extensible system... ben g On Mon, Dec 22, 2008 at 9:03 AM, Richard Loosemore <[email protected]>wrote: > Valentina Poletti wrote: > >> I have a question for you AGIers.. from your experience as well as from >> your background, how relevant do you think software engineering is in >> developing AI software and, in particular AGI software? Just wondering.. >> does software verification as well as correctness proving serve any use in >> this field? Or is this something used just for Nasa and critical >> applications? >> > > 1) Software engineering (if we take that to mean the conventional > repertoire of techniques taught as "SE") is relevant to any project that > gets up above a certain size, but it is less important when the project is > much smaller, serves a more exploratory function, or where the design is > constantly changing. To this extent I agree with Pei's comments. > > 2) If you are looking beyond the idea of simply grabbing some SE techniques > off the shelf, and are instead asking if SE can have an impact on AGI, then > the answer is a dramatic "Yes!". Why? Because tools determine the way that > we *can* think about things. Tools shape our thoughts. They can sometimes > enable us to think in new ways that were simply not possible before the > tools were invented. > > I decided a long time ago that if cognitive scientists had easy-to-use use > tools that enabled them to construct realistic components of thinking > systems, their entire style of explanation would be revolutionized. Right > now, cog sci people cannot afford the time to be both cog sci experts *and* > sophisticated software developers, so they have to make do with programming > that is, by and large, trivially simple. This determines the kinds of > models and explanations they can come up with. (Ditto in spades for the > neuroscientists, by the way). > > So, the more global answer to your question is that nothing could be more > important for AGI than software engineering. > > The problem is, that the kind of software engineering we are talking about > is not a matter of grabbing SE components off the shelf, but asking what the > needs of cognitive scientists and AGIers might be, and then inventing new > techniques and tools that will give these people the ability to think about > intelligent systems in new ways. > > That is why I am working on Safaire. > > > > > > Richard Loosemore > > > > ------------------------------------------- > agi > Archives: https://www.listbox.com/member/archive/303/=now > RSS Feed: https://www.listbox.com/member/archive/rss/303/ > Modify Your Subscription: > https://www.listbox.com/member/?&; > Powered by Listbox: http://www.listbox.com > -- Ben Goertzel, PhD CEO, Novamente LLC and Biomind LLC Director of Research, SIAI [email protected] "I intend to live forever, or die trying." -- Groucho Marx ------------------------------------------- agi Archives: https://www.listbox.com/member/archive/303/=now RSS Feed: https://www.listbox.com/member/archive/rss/303/ Modify Your Subscription: https://www.listbox.com/member/?member_id=8660244&id_secret=123753653-47f84b Powered by Listbox: http://www.listbox.com
