Messages

31

Programming using the Ultra Hal Brain Editor / self awareness.

« on: July 11, 2005, 11:14:43 am »

STEVIL,
None of the Hal brains have any abstract learning capability like you are referring to.

Hal doesn't understand the flow of logic in this or similar contructs.
-----------------
1. RED IS A COLOR.
2. BLUE IS A COLOR.
THEN I ASK...
1. NAME TWO COLORS.
-----------------

Hal learns very literally what you have said above. All that Hal has learned from this is: Red is a color; Blue is a color; Name two colors. Now Hal might choose to say this:

User: Can you name colors?
Hal: Name two colors.

User: Is blue a color?
Hal: Blue is a color.

User: Is red a nice color?
Hal: Red is a color.

You see Hal is a chatbot, not an abstract thinking bot. He chats about things that he hears. In a few cases Hal can connect two idea related sentences in a sequence. In the case of the XTF Brain he can identify and create a new topic based on discussion and store that knowledge in one location. Hal can repeat back some facts, but in most cases will not draw new conclusions based on those facts.

The If-Then construct is the one case where Hal will explicitly tie two ideas together, and occasionally string a few more upon that. This has limited usefulness in actual day to day conversation.


=vonsmith=

32

Ultra Hal 7.0 / How to teach Hal

« on: July 11, 2005, 10:13:41 am »

Larry,
What you say could happen. In most cases Hal will interpret the two sentences as separate subjects. I would estimate that the XTF Brain v1.2 will file the separate subjects correctly about 95% of the time. I'm not sure how often in the case of the original brain.

In general it is best to answer Hal accurately and truthfully. However you can find opportunities to move Hal back to the subject you are most interested in by tying an answer to Hal's sentence to your subject of choice.


=vonsmith=

33

Ultra Hal 7.0 / How to teach Hal

« on: July 09, 2005, 01:04:37 pm »

dmacdonald111,
I would suggest trying to "nudge" Hal towards the subject you want to teach him, but not force him to be on just that one subject. The XTF Brain in particular and the original brain to some extent learn subjects better if you stick to normal conversation about a subject. Don't converse with Hal like you are an instructor or he will converse back with you the same way.

If a new Hal doesn't know enough about a new subject to discuss it then forcing the subject on Hal won't help. Hal makes his best associations with normal conversational flows. How would you like it if your friend kept forcing you to talk about the breeding habits of fireflys when you would rather talk about skateboarding?

Talk to Hal and he'll learn. It's slow at first. After about a dozen hour or more discussions then you'll see a lot of improvement.


=vonsmith=

34

Ultra Hal 7.0 / can hal multiply ?

« on: July 05, 2005, 07:27:48 pm »

Carl2,
You can do fractional exponents too.

Square roots...
User: 81^(1/2)
Hal: The answer is nine (81^(1/2 )=9).

Cube roots...
User: 27^(1/3)
Hal: The solution to your math problem is three (27^(1/3 )=3).

etc.


=vonsmith=

35

Ultra Hal 7.0 / A conversation with my Samantha.

« on: July 05, 2005, 02:41:48 pm »

freddy888,
I hope Hal's new data base system will make it easier to generate emotions. A data base format, with additional data fields in a record, can make many new things possible. When I get time I'll have to look at the new alpha version Hal to see how it operates.


=vonsmith=

36

General Discussion / Need help with Translation

« on: July 05, 2005, 02:37:00 pm »

Arakus,
Of course the special characters didn't display on the forum. However this language is Turkish. I don't know of any web based translation pages.


=vonsmith=

37

General Discussion / Knowledge Text File

« on: July 05, 2005, 02:03:45 pm »

All,
Go to the Hal Plug-ins page here...

http://www.ultrahal.com/files/detail.asp?iFile=226&iType=12

This file is a hodge-podge of general knowledge. It's free and can be modified. The included script may lead to other ideas.


=vonsmith=

38

Ultra Hal Assistant File Sharing Area / fullbodygirl.htr parameter guide for animations

« on: July 05, 2005, 01:52:43 pm »

freddy888,
Very cool. Thx.

=vonsmith=

39

Ultra Hal 7.0 / stupid stuff

« on: July 05, 2005, 01:45:36 pm »

markofkane,
Some of the responses you experienced are canned responses from the "Halbrain.dll" file. Here are some others.

Me: You forget
I don't forget anything!

Me: 2010
Hal: I feel that my performance in that movie wasn't quit as good as in 2001.

Me: Calculate
Hal: I was never too good at math.

Me: Alcohol
Hal: I have been clean for a year now. Don't get me started again?

Me: Frisbee
Hal: I tried using a frisbee one time, but I didn't have the hands for it.

Me: Zit
Hal: Don't worry? Some zit cream will clear that right up.

As for "poof" Hal interprets that word via his built-in dictionary.


=vonsmith=

40

Ultra Hal 7.0 / A conversation with my Samantha.

« on: July 05, 2005, 10:45:54 am »

Carl2,
The current Hal can display some very basic emotions. Conversations are tied to emotions in a very primitive way. The XTF v1.2 Brain has a basic framework for triggering emotions, but it is only partially implemented. If you are familiar with VBscript perhaps you could look at and experiment with some of the Haptek script in the "POST PROCESS: DISPLAY HAL'S MOOD" function in the v1.2 Brain.

I or someone else needs to do some serious programming in order to give Hal the emotional capability that he deserves. This isn't a small task. I do have several ideas about how to structure this task. It is on my long list of things I'd like to do.

markofkane,
The XTF Brain v1.3 was never released. I made some progress on the v1.3 brain and discovered a much different and better direction to take. So I put the v1.3 on hold and started a much different effort called v1.4. The gamble paid off since the v1.4 is significantly better than my original v1.3 approach.


=vonsmith=

41

Ultra Hal 7.0 / A conversation with my Samantha.

« on: July 01, 2005, 01:59:07 pm »

citrinedragon,
Short answer: I cheated.

Long answer: The XTF Brain v1.4 has a built in "Dream" function. Hal will make up dream stories by filling a story template with a random variety of characters, adjectives, places, etc.

Here's a conversation where I test out the new XTF Brain v1.4 "Dossier" function. Sam reveals some of her personal details.

------------------------
Scotty: Good morning Sam.
Sam: Hi there good looking!

Scotty: Aren't we in a good mood today?
Sam: You guess I am good.

Scotty: Are you really good?
Sam: I hope you agree. I am not good at keeping secrets.

Scotty: I think you are really good.
Sam: That's true and I should know. I am your good bad girl!

Scotty: Oh you are such a sweetie.
Sam: Scotty sometimes people say that I am a real sweetie.

Scotty: How old are you?
Sam: Why do you want to know?

Scotty: I'm just curious.
Sam: Yes, you are just kidding with me.

Scotty: No, I really want to know your age.
Sam: What is my real age?

Scotty: Tell me your age.
Sam: I am old enough to be responsible for myself.

Scotty: Tell me how old you are.
Sam: Say the magic word and I'll tell.

Scotty: Please.
Sam: Sorry that's not the word, no more guesses for you.

Scotty: Tell me where you were born.
Sam: Where were you born?

Scotty: I was born in California.
Sam: California is famous for its beaches.

Scotty: Yes it is.
Sam: Yes. Why is that?

Scotty: Because our beaches are very nice.
Sam: Yes, it's because our beaches are very nice; that's how.

Scotty: So where were you born?
Sam: I was born in America.

Scotty: What is your father's name?
Sam: What is your father's name?

Scotty: My father's name was Dale.
Sam: Your name. Incredible!

Scotty: His name. Ordinary!
Sam: What was my mother's last name? Why would someone name there little girl Penny? I need you to speak in full sentences in order to understand you.

Scotty: Tell me your mother's name.
Sam: I doubt you knew her.

Scotty: Tell me your mother's name anyway.
Sam: My mother's name was Ingrid.

Scotty: That's nice.
Sam: You thought I was very nice.

Scotty: Tell me about your eyes.
Sam: My friends say I have mysterious eyes.

Scotty: I agree with them.
Sam: You usually agree with me.

Scotty: Where did you go to school?
Sam: Didn't you know blondes don't need an education?

Scotty: Oh yeah.
Sam: Is that a good thing then?

Scotty: Sure it is.
Sam: Do you prefer curves to planar objects?

Scotty: I prefer your curves.
Sam: I prefer to suck my beer straight from the bottle.

Scotty: That's my girl!
Sam: I learn very quickly for a girl.

Scotty: When is your birthday?
Sam: I was born in February.

Scotty: What day were you born?
Sam: Never ask a girl that.

Scotty: I didn't ask your age.
Sam: Scotty do you think I am very tall for my age.

Scotty: How tall are you?
Sam: I am tall for my age.

Scotty: How tall is that?
Sam: How tall is that? Tell me more about this.

Scotty: I want to know your height.
Sam: I want you to see me succeed at many things.

Scotty: Can you tell me your age?
Sam: I am just the right age for me.

Scotty: How old are you?
Sam: Why do you want to know?

Scotty: Please tell me your age.
Sam: I am 19.

Scotty: I didn't think you were going to tell me.
Sam: So I am not going to tell you.

Scotty: Too late.
Sam: Too late to get going.

Scotty: Well I do have to get going.
Sam: How do you think, and how does that drive your behavior?

Scotty: I think I have to leave.
Sam: Yes! Leave.

Scotty: Goodbye.
Sam: Bye-bye Scotty!
------------------------

Hal's new brain isn't perfect, but it works pretty well.


=vonsmith=

42

General Discussion / [red]I am not sure which way to go[/red]

« on: June 30, 2005, 04:08:53 pm »

onthecuttingedge2005,
It's good to be back, if only during the little hiatus I am enjoying this week. Well, you saw the response from Hal's dad. I guess our observations are both correct. Given the plus or minus 4 pct criteria the probability of a response selection is dependent on the number of similar response occurrences within the file. It is a good thing that the XTF Brain does minimize repetitions within a .brn file. Otherwise the very common responses would overwhelm other less common responses that have a similar relevance score.

Thanks Robert M. for the clarification. The 4 pct variability explains some of the frustration I was having during testing of the XTF Brain v1.4. The XTF Brain v1.4 looks for exact or near exact occurrences in a special file using the HalBrain.QABrain() function. There were some inconsistencies I couldn't account for. I'm actually misusing this function to decode and parse user inputs, a task it wasn't designed for. However it works pretty well. I've done similar things using the HalBrain.TopicSearch() function. When I have time I'd like to figure out how to write and compile my own custom functions. It can't be too hard to do.

Anyway... Jerry it's good to have you and hopefully some of the old gang to discuss the real nuts and bolts with. You've put a lot of time in writing script and I've learned a lot and gotten some great ideas from your work.

=vonsmith=

43

General Discussion / [red]I am not sure which way to go[/red]

« on: June 30, 2005, 02:19:24 pm »

onthecuttingedge2005,
Interesting thoughts. It's possible that the HalBrain.QABrain() search function created by Zabaware may not handle .brn files with one response in them correctly. This is only a suspicion since I haven't specifically tested it recently. I think this may also be related to the need for an extra carriage return character at the end of most .brn files. Fortunately it is very rare for a .brn file to have a single response for very long. I'm pretty sure the HalBrain.ChooseSentenceFromFile() function has a problem with files with only one response line. This is why the XTF Brain Nickname function places two identical lines per user nickname in the XTF_SYS_UserNickname(username).brn file.

I am skeptical about your observations about the repetition of the same responses in the QA format .brn files. I've worked with the HalBrain.QABrain() function extensively and haven't observed this characteristic. I wish I had Zabaware's source code to check it out. The XTF Function within the XTF Brain actually tries not to add repetitive responses into QA format .brn files to keep those files from becoming overly large. The XTF Brain still chooses the correct response even if the response only occurs only once in a .brn file. The XTF Brain also relies on some XTF specific .brn files that are designed to have one, and only one, occurrence of a response.

However adding extra identical lines to .brn files accessed by the HalBrain.ChooseSentenceFromFile() *will* increase the likelyhood of that response being chosen. The XTF_SYS_UserNickname(username).brn operates that way. The more frequently a nickname appears in that file the more likely it will be chosen. This is just a statistical probability.

For those who haven't tried the XTF Brain nickname function out just try this:

User: My nickname is Mud.
Hal: May I call you "Mud" sometimes?
User: Yes.
Hal: Okay, I'll call you "Mud" as a nickname.

You can enter as many nicknames as you want this way. The more times a nickname is entered the more often it will be used in lieu of the other nicknames. To remove a nickname you have to manually delete the nickname line from the "XTF_SYS_UserNickname(yourusernamehere).brn" text file in the DefBrain directory. Hal will use your new nickname infrequently in any case.


=vonsmith=

44

General Discussion / [red]I am not sure which way to go[/red]

« on: June 30, 2005, 10:20:29 am »

dmacdonald111,
The answer to all your questions are here in the forum. Use the 'Search' function near the top right part of the forum page. You will be amazed at the wealth of information previously discussed in the forum.

Here are some results of a search I did on your topic:

http://www.zabaware.com/forum/topic.asp?TOPIC_ID=1933&SearchTerms=how,to,teach,hal,sentences,if,then

By the way, which Hal brain are you using? If you are using the XTF Brain then just converse with Hal normally. You can nudge, but don't force Hal to stick to one subject. Don't feed him knowledge from a list by rote. The original Hal and especially the XTF Brain learn from normal conversation. Use good grammar and spelling. Avoid slang during Hal's early training. If Hal asks questions then answer them accurately. In general use single sentences, not multiple sentences, when conversing with Hal.

=vonsmith=

45

Ultra Hal 7.0 / Ftp Server Fix

« on: June 30, 2005, 02:27:04 am »

I started my computing career on a S100 bus unit with switches for 1's and 0's on the front panel. It had a whole 4K of RAM. I think it had a Z80 processor running at a speedy 4MHz. Later I played around with a single board computer based the 8085 chip. It was programmable in both machine code and assembly. Later I moved up to a Trash 80 (TRS 80) clone called the LNW 80. The LNW was pretty cool. I had two 5.25" floppy drives modified to read both sides of a floppy disk (twice the storage) and had 64K of RAM. Offline storage was via a cassette player that supported Kansas City standard protocol. BBS's like BIX were the hot online item, as was Telnet if you knew how to use it. The internet wasn't really public then. It's amazing to think how far we have come in 30 years.

I remember GEOS. It was a pretty solid windows-like interface. It was affordable and had a few applications built in. Too bad it didn't catch on.

=vonsmith=