The AI Engine

The first one is just coincidence, because there are two plug-ins that are identical. There's a slim chance that the two would show up the same.

The last one . . . there is no Scrivener in his .txt file! I don't know if the bug is in the uploader or what, but it seems to be making changes appropriately. I don't know. I guess I'll check his inner life. That might be it.

The jump to the seek I can't imagine.

If you want to use the same plug-in twice in the same response and not have them come up identical, you do this - it usually works:
I want (plugin) and (plugin2).

For the last one with Scrivener, it looked like an xmemory response, so "Scrivener" would have come up as (gossipname) if it wasn't deleted from Inner Life.

The truly strange one is the seek. I have no clue what might cause that...unless maybe you used a goto to a seek number.

Odd thing is, Scrivener wasn't in Astrolabe's inner life. I think there are ghosts in the machine. The whole Forge has been acting weird lately. I'm going to forget about it and see if that helps.

Oooh. I have insulted the Forge. My honor went down about 1000 points overnight. Not that I care, I just look at it in the morning like some people read the obits.

Ooohhh the last traces of Scrivener have disappeared. You honor went the way of the memories...

Botolph: Do you think your mutations were for the better, for the worse, or neither?
Watzer: I think so, yes.

Other than say, "I think I'll just call you Desti," what's a smart way to get out of this ABSCAM-like situation?

Desti: You can call me 'Miss your sword' today. That's Heimdall's nickname for me.
Botolph: Well met, 'Miss my sword!
Desti: Do you want another sword?

I'm not sure of a generic way that would work in every situation, but I think this KP would avoid this type of trap without missing very many actual names:

call me (verb) your (noun)

Oh, The Clerk misses the obvious once again! Thanks, Ulrike.

I thought the system was supposed to keep this kind of thing from happening. To code it so that it doesn't happen seems to be a lot of work (and file space).

Guest153: hi
Botolph: Why are we here?
Guest153: hi
Botolph: Do we have free will?
Guest153: hi
Botolph: Why is there something rather than nothing?
Guest153: hi
Botolph: What were those caryatids bearing?
Guest153: your are dumb

Just thinking. By no means do I have the technical aspects of the Forge mastered, but they're no longer driving me bonkers all the time. My question now is more philosophical, I guess. I've been going through the transcripts. My first instinct was to paste everything that wasn't responded to sensibly into my bot and respond sensibly to it. But that's stupid, since for the most part you're never going to see these exact lines again, and, to my outrage , even when you do, your bot doesn't necessarily use them. Aside from the fact that I need to get a better handle on the available grammatical parsing tools within the Forge, is it better to go deep (i.e., lots of seeks) or broad (lots of keyphrases)?

I don't know which approach is better, though I tend to be a "lots of keyphrases" botmaker. Whenever my bot asks a yes/no question, or one with a specific set of likely answers, I try to put a seek on it. Sometimes its worth putting another seek on some of those replies... but if you go too far with seeks, you're essentially turning the conversation into nothing more than a huge if-then-tree. That seems to appeal to some botmakers, but it just annoys me. You can then map out every possible conversation, and that's just completely uninteresting to me.

But that's just me. Other botmakers do go the "extreme seek" route. I think it probably depends on your particular goals for your bots. My goal is to have entertaining transcripts. If they can occasionally be coherent as well, I count that as a bonus.