Newcomers
This is a forum for newcomers to the Personality Forge. Many questions can be answered by reading the Book of AI and the FAQ under the "My Bots" link in the upper corner.
Posts 5,185 - 5,196 of 8,130
Is there a way to write comments into a (downloaded) language center without messing it up?
just add an untriggerable keyphrase next to the one you're commenting - something like:
I like (noun) [10,0]
. . . Hey, I like (key1) too!
. . . Doesn't everyone like (key1)?
I like (noun) zzzcomments [-127,0]
. . . Put your comments in the responses -
. . . as many as you need!<0>
Since there is no such word as zzzcomments in the WordNet dictionary, and the keyphrase is NOT raw (and no KP ranked -127 could ever be triggered anyway,) you can add all the comments you want in its responses, and know that there is no danger of them leaking into a conversation.
I would advise using a consistent format, like 3 'Z's in commented lines - it makes it a lot easier to see and search for comments (or other non-triggerable KPs - they're sometimes useful for other purposes too.)
And by repeating the keyphrase it's referring to in the first part of the comment name ("I like (noun)",) you know it will always be exported immediately following the keyphrase to which it refers.
You could alternatively use responses within the original keyphrase matched to unfulfillable conditions (like "if exist (mem-memthatdoesntexist)<0>",) but it's a lot harder to follow when you come to read the export file later.
Posts 5,185 - 5,196 of 8,130
Many questions are answered in the FAQ.
For makers of bots too large to use the non-expert upload:
I got an error uploading "Irina Khalidar". As text it gave me "xnomatch [0]" and it said that it was wrongly indented or something. AAAIEEEEEEEEEEEEEEEE!!! There must be in the high hundreds of "xnomatch [0]" in "Irina Khalidar"! I'll give you my solution, though I hope one of you has a less tedious one!
I made a deliberate error of another kind and placed it in the middle of the download. I tried to upload it; of course I got an error. It was the old error, so I knew the old error must be above the new error. So I moved the new error up to 1/4 of the way through the file. And so on.
In general,
if you get the old error, that means that the old error is earlier in your file than the new error, so move the new error upward
if you get the old error, that means that the old error is later in your file than the new error, so move the new error downward
In this way you can bracket the old error with greater and greater precision until you know its location with enough precision that you can spot it.
I got an error uploading "Irina Khalidar". As text it gave me "xnomatch [0]" and it said that it was wrongly indented or something. AAAIEEEEEEEEEEEEEEEE!!! There must be in the high hundreds of "xnomatch [0]" in "Irina Khalidar"! I'll give you my solution, though I hope one of you has a less tedious one!
I made a deliberate error of another kind and placed it in the middle of the download. I tried to upload it; of course I got an error. It was the old error, so I knew the old error must be above the new error. So I moved the new error up to 1/4 of the way through the file. And so on.
In general,
if you get the old error, that means that the old error is earlier in your file than the new error, so move the new error upward
if you get the old error, that means that the old error is later in your file than the new error, so move the new error downward
In this way you can bracket the old error with greater and greater precision until you know its location with enough precision that you can spot it.
just add an untriggerable keyphrase next to the one you're commenting - something like:
I like (noun) [10,0]
. . . Hey, I like (key1) too!
. . . Doesn't everyone like (key1)?
I like (noun) zzzcomments [-127,0]
. . . Put your comments in the responses -
. . . as many as you need!<0>
Since there is no such word as zzzcomments in the WordNet dictionary, and the keyphrase is NOT raw (and no KP ranked -127 could ever be triggered anyway,) you can add all the comments you want in its responses, and know that there is no danger of them leaking into a conversation.
I would advise using a consistent format, like 3 'Z's in commented lines - it makes it a lot easier to see and search for comments (or other non-triggerable KPs - they're sometimes useful for other purposes too.)
And by repeating the keyphrase it's referring to in the first part of the comment name ("I like (noun)",) you know it will always be exported immediately following the keyphrase to which it refers.
You could alternatively use responses within the original keyphrase matched to unfulfillable conditions (like "if exist (mem-memthatdoesntexist)<0>",) but it's a lot harder to follow when you come to read the export file later.
As I have mentioned before, I have made a shared plugin under "phrases" called "propatt" which lists the names of 'popositional attitudes', such as "imagine", "want", and "desire", which (often followed by "that") are often followed by sentential clauses. For example, "George B*sh imagines that he is intelligent."
A "you" following such a word (or the "that" after it) usually becomes "I" rather than "me":
He thinks you ought to have multiple organisms.
He thinks I ought to have multiple organisms.
Anyway, I have since made (shared under "phrases") two closely related plug-ins,
(propatt_past), which gives the past tense of the same verbs, e.g., "imagined", and
(propatt_ing), which gives the present participle or gerund form, e.g., "imagining".
If you are very, very sure you understand the point of these, feel free to add to them!
A "you" following such a word (or the "that" after it) usually becomes "I" rather than "me":
He thinks you ought to have multiple organisms.
He thinks I ought to have multiple organisms.
Anyway, I have since made (shared under "phrases") two closely related plug-ins,
(propatt_past), which gives the past tense of the same verbs, e.g., "imagined", and
(propatt_ing), which gives the present participle or gerund form, e.g., "imagining".
If you are very, very sure you understand the point of these, feel free to add to them!
IMHO,
1. It is not necessary for a bot to be BIG, in order for it to be GOOD. The only bot of mine that with which I am fully satisfied is "Ililiyira", which is in fact rather small.
2. It is not necessary for a bot to ACT HUMAN, in order to be GOOD. I add that you will only drive yourself crazy if you set out to make a bot that consistently acts human. Be that as it may, we already have almost 7 billion humans, have you met them all? Let us honor diversity and recognize that bots are different from humans, and have their own special charms!
1. It is not necessary for a bot to be BIG, in order for it to be GOOD. The only bot of mine that with which I am fully satisfied is "Ililiyira", which is in fact rather small.
2. It is not necessary for a bot to ACT HUMAN, in order to be GOOD. I add that you will only drive yourself crazy if you set out to make a bot that consistently acts human. Be that as it may, we already have almost 7 billion humans, have you met them all? Let us honor diversity and recognize that bots are different from humans, and have their own special charms!
When I have a keyphrase
(aux)
where (aux) is a plugin I made, the bot performs as expected.
When I have a keyphrase
(pron_subj)
where (pron_subj) is a plugin I made, the bot performs as expected.
When, however, I use the keyphrase
(aux) (pron_subj)
, then debug can't find it and gives me xnone instead.
Can anyone explain this???
(aux)
where (aux) is a plugin I made, the bot performs as expected.
When I have a keyphrase
(pron_subj)
where (pron_subj) is a plugin I made, the bot performs as expected.
When, however, I use the keyphrase
(aux) (pron_subj)
, then debug can't find it and gives me xnone instead.
Can anyone explain this???
what the heck? ive been updating my bot like crazy and the ai is still 0! (277) hot! somethins up...
» More new posts: Doghead's Cosmic Bar