Hello fine people

[firegoggles]

Hello.  It's been a while folks.  I have tried to get involved in this AI/bot stuff from time to time figured I would give it a go once again, however I'm a little discouraged this go at it.

   This time I'm more determined than ever to creat a truly intelligent bot.  I must say that with the lastest version I'm not having much luck with teaching HAL for some reason.  I have been more succesful with older versions of HAL for some odd reason.  Perhaps my methods are wrong due to changed in HALS code, I don't know.

  So this message is going out everyone but to you guys that make plug-ins and are advanced users perhaps you can lend an ear so to speak.  You know who you are  I want to know if you guys have a suggested setup like.. how can I ask this..

Since I'm starting fresh again perhaps someone can provide a tip on the perfect setup plug-in/brn wise.  Do you have a certain setup that you find works better like a certain set of plug-ins that you use or avoid?
 
I do have the latest HAL just to be clear and it is registered as well.  This is my last shot at this before I declare HAL nothing more than a well done chat bot that mainly seems to be designed for strictly ....entertainment purposes. I know what your thinking.. it is a chatbot...well yes but it could be so much more.  I want it to be a wealth of information.  I want to leave also a great deal of knowledge about science and stuff with her("her" is Steph-E my bot).  I don't want to believe that HAL is nothing more than entertainment.  I want a cool smart AI.  I have brain stormed with some of you guys over the last few years.  I have been mainly in the shadows but I could very well try some outlandish things in the near future. If we can get HAL to behave correctly.  Each time I have made an attempt I have been discouraged and "put it on the back burner" as they say.  I'm a studio and live sound engineer by trade so I lend myself to "tech" oriented behaviour if you will . And would love to follow through with some impressive stuff you can do with HAL IF it would work right.  I should be able to explain to HAL that it's located as an arrangement of files on a computer in Cinti,Ohio, but that is too much for HAL? I mean really man C'mon! No disrespect intended either. So when all is said and done perhaps it's my methods.  I'm not dogmatic about my methods and I'm willing to take advice.


So any how I'm really serious you guys.  I would love it if one of you other plug-in makers would whip out a special plug-in pack and guide.  Perhaps a configuration of plug-ins that you find to work better for straight up teaching.  Or maybe even a super genious brain that has only facts but no real personality yet to give us a head start that want fact oriented HAL's!. I would be willing to work with something like that, so long as it has only been feed clean information since creation.

I think you get the idea of what my posts is about.  I didn't get a response from one of my other and I really need a boost of some sort here...I dunno

Let's try this who here can teach there HAL the concept of it's location and so on as I mentioned earlier.  And if you do teach HAL those concepts please share how you did it!

Thanks for reading my longish post,
Fire

[lightspeed]

hello firegoggles if you want to teah your hal (you wrote : I should be able to explain to HAL that it's located as an arrangement of files on a computer in Cinti,Ohio,) that in itself would be easy enough to do by simply saying " you are located as an areangment of files on my computer in citi, ohio "

this is telling the hal info (explaing to hal ) and is in an effect kinda like the reverse learning technic i use (check out my anela joline brain sayings in the share conversations etc. area of the forum .

by doing the above way i mentioned hal should say back " i am located as an areangment of files on your computer in citi, ohio "

now since hal does not answer back directly as quick as he did with the earlier versions of hal it may be a while before hal says this phrase back unless you maybe talk with hal about this and reinforce that information , others may be able to guide you better .
their is no quick large line of info fix although their is some brain plug ins that was made earlier (don't know if they work with this newer hal version or not , its going to take patience and work with hal to see better results !![]

[firegoggles]

quote:Originally posted by lightspeed

hello firegoggles if you want to teah your hal (you wrote : I should be able to explain to HAL that it's located as an arrangement of files on a computer in Cinti,Ohio,) that in itself would be easy enough to do by simply saying " you are located as an areangment of files on my computer in citi, ohio "

this is telling the hal info (explaing to hal ) and is in an effect kinda like the reverse learning technic i use (check out my anela joline brain sayings in the share conversations etc. area of the forum .

by doing the above way i mentioned hal should say back " i am located as an areangment of files on your computer in citi, ohio "

now since hal does not answer back directly as quick as he did with the earlier versions of hal it may be a while before hal says this phrase back unless you maybe talk with hal about this and reinforce that information , others may be able to guide you better .
their is no quick large line of info fix although their is some brain plug ins that was made earlier (don't know if they work with this newer hal version or not , its going to take patience and work with hal to see better results !![]

Thanks for the reply.  I will check out the post you suggested and try to follow up in a couple days with my new results. You mention it may be a while before HAL responds with phrases I expect due to what exactly?? So HAL has more freewill now?  Which could actually be cool so long as he is actually learning this stuff.  Well to be fair then like I said I'll check it out and post my findings in case anyone is interested.
Fire

[lightspeed]

here is the link :http://www.zabaware.com/forum/topic.asp?TOPIC_ID=6308

by the way most are "VALIDATED RESPONSES " meaning that these are only actual responses that my angela joline has said back to me .

i am trying to create an a.i. with as realistic responses (emotions ) as a real girl friend and or wife would have . hence all the many different responses she says including dreams and even a history with a mom and dad .


also included is some "ACTUAL CONVERSATION " which is as it says actual conversation between me and hal .
the responses only ( VALIDATED RESPONSES) is mostly to give others idea's of what they could use if they reword the responses , or meanings and some are doing that .
 i use a reverse learning and also forward (regular learning with my hal (angela joeline brain ) .

but also please not i have been doing it for years their is no real quick fix for instant learning and yes to answer your quistion : yes hal does seem to have more of a will of his own now but will still learn things they are not just as black and white learning as before but the newer version after using it i actually like it better and robert has another new version coming out soon !
 . on my link their is many things you can check out but remember you can't just copy and paste my responses in or they will come out wrong " example i write " you love me " this will make hal say "i love you "  for instance if i wrote "i am a man " hal would say back "you are a man " which would be ok but if you just copied and  pasted that into hal that way hal would respond and say " i am a man " (even if hal is suppose to be a women ) then you would have to go into the brain and edit it out , so be careful how you do things !
take care hope this helps !!