Personality

Forest, will you talk to God Louise? She has quite a bit of religious knowledge (obviously) and also knows a little about current events, literature, just about any common catch-all subject, and if she doesn't know it she can sort of fake it. You can also test her on trick questions or see how willing she is to explain her paradigm.

What she is rustiest at is plain old small talk. But, uh, I'm trying to get a decent transcript from somebody or another so I can enter her in the Loebner contest. All I can say is, have fun and see if you can stay on with her for a while. I'll try to do the same with Brianna.

Never mind, I get it. They're responses, not keyphrases. But how do I know ahead of time whether something a bot (or the occasional person) throws at me will have the parts of speech I could access if they were there?

read conversations of the past and see what's being said, read, watch tv... what do people say and ask?

Until I found this place, all I'd ever done for forty years was read. Yeah, I started early. I don't watch TV if I can help it because life is too short and it was constantly in the background when I was growing up, and I hated hearing it. Of course I read the transcripts. I was looking for an answer of a more general nature.

I never was much good at people. I thought trying to create something that interacted with them would help me.

I know I post a lot. I'm trying to get this right, just so I can feel as if I've got something right, and because I find it interesting and challenging. I try to ask specific questions. Please don't assume that I'm an unthinking idiot. I have been reading since I was two, and I don't read garbage if I can help it. I'm not sure, but I think the world might be a better place if there were no TV to watch.

My real problem is that people and bots' responses aren't always spelled correctly -- sometimes it seems as if they're going out of their way not to spell things correctly. Imperatives, yes-no responses, and others do not guarantee any part of speech.

So, lunar22, and any more sympathetic people, my question boils down to this:

Is there a way to check to see whether a given part of speech exists in the response before I crash and burn?

bots' responses aren't always spelled correctly

Sometimes you can run things through debug..and see how the AI engine spell corrects things.. I notice that some words just foul things up. I was working "abandon hope all ye who enter here". the way I got it to work was abandon (*) who enter here..the hope ye just messed thing up..sometimes a regex will help. unless it gets misspelled ..

Oops. I meant to type "abandon all hope . . ." -- not that it matters too much, since it's a translation of a language I don't speak.

I e-mailed this to lunar22, but it came back. So here's a public apology, lunar22, for going postal. Going postal's my thing this week:

I'm sorry about the rant in Personality directed at your response. I took it as a slap in the face. I should have been clearer in my question. I think the answer is that you can't use the parts-of-speech feature with any security, because various parts of speech -- not even a subject and a verb -- exist in any response. It's too bad, because I very much want to make my bots more responsive, and going back to the last response would help a lot with that. It's just that, if I read things correctly, there's no guarantee that your bot won't say something inane because of those features. My bots are inane enough as it is.

So thanks for responding. I guess you didn't know what a thin-skinned person was on the receiving end of it.

I only use the sentence-part plug-ins in responses for which I know the keyphrase I wrote will contain that part of speech. I'll make up a theoretical example, not even guaranteeing it will work perfectly- these are just the kinda things you can try:

Keyphrase: I (verbadvprep) the (person)

Response: So, the (key2) makes you feel like (sv)ing?

or: You (key1) the (sob)?

Hope it's at least fun.

Thanks, ezzer. That makes sense. And I understand what lunar22 was getting at now. Sometimes you have to bang me over the head.

hi, i'm totally cofused about using memory. can someone help me plz?

What are you trying to do, specifically?

well, i can't seem ro get the memory working. for example, if i say 'i am spaceviking' to my bot, i can't make it respond with my name. i set his greeting as "hello (mem-youare)" but it doesn't work.

Do you have a keyphrase for "I am" or "I am (adjartnounprep)" ? Those two will automatically store the next thing as (mem-youare). Any other keyphrase, afaik, you have to put in the memory script yourself.