Re: FAQ system

Hello Jim, On 12-Jun-99 18:50:06, you wrote: >Bonus idea: >* An email submission mechanism that listens at faq@php.net, > entering the entry with the Subject as the question and the > body of the message as the answer. >I'll sign up to do this when the primary system is up and running. >Ideas that are death: >* Accepting questions without answers. >* Intelligent parsing of questions to provide answers. >* Wholesale integration with the manual. >* Integration with the bug database. >Somebody else would have to step forward for any of these. I'm not sure if this is what you mean, but I think what would be great would a sort of knowlegde based engine that would analyse the questions and try to present extracts of the FAQ or the manual that might answer the questions. I don't intend to push myself into anything, but I have some experience here. Last year I released a book about the Internet by McGraw-Hill Portugal. I thought of doing a site about it, but to attract people to the book I thought of offering something more than just promotional pages. I developed a service that answers to questions about the Internet by presention extracts of the book where the answers possibly are. If you know anything about portuguese go straight to http://www.e-na.net/ . If not, try it with Altavista Babelfish. The service has some tolerance to misspelled words. It may also accept as equivalent different forms of the same words. If you ask for proxy or proxies, it will identify the words as being the same, but this is because I told it too. The engine is pretty dumb regarding spelling and grammar. So it has a reasonable level idiom independence. Independently if you are interested on my engine, I may help with know-how on the implementation of this kind of knowlegde base engines. Just let me know what you think. Regards, Manuel Lemos E-mail: mlemos@acm.org URL: http://www.e-na.net/the_author.html PGP key: finger://mlemos@zeus.ci.ua.pt --