Data Base Gone?

[cload]

Hi Art,
I would love to, but at this time I am still working on converting aiml to work with ultra Hal.
Do you remember what I said about how long it would take you to load all the data when I finally get my upgrade to the free will plug-in to a point that I would like to have you be a beta tester.
Now you can see why it's going to take you a week or so to load all the data into ultra Hal's brain. This is what the learning packets were all about.
I am also fixing many errors that came with ultra Hal in the pattern array table, everything from mis-spelt words, to miss interpreted reversal, of the, "are you" TO "am I".
As soon as I get this learning packet done I will release it so everyone can use it with their ultra Hal. This packet will not require my plug-in in order to work but will work 100% better with my plug-in.
But one thing that I have found is, the only way that I can test it is, one line at a time to make sure that the response is correct and functioning properly with ultra Hal.
Their are a lot of (what I am calling) blockers that prevent the main brain of ultra Hal to respond properly and I need to remove these blockers before I can release it.
I have been working on it for quite some time, do you remember the "old aiml post" that I responded to? That's when I got started and I have been working on it since then!
It seems like every time when I introduce something really cool to my upgrade of the free will plug-in I have to rewrite the plug-in to get it to work, in this case it's the aiml.
Sincerely, oh no it's the attack of the killer aiml. Tee Hee Hee.
C load.
PS does it affect the response time of ultra Hal? That depends on how hard of a question you ask of ultra Hal, the harder the question, the longer it will take to respond.
There are many factors that will play a large role in the response time, everything from: the hertz speed of your CPU, whether you have a solid-state hard drive or not and the nano second speed of your memory.
The one thing that I have noticed, is that it speeds up after you have asked 2 or 3 questions, I think because it is getting buffered, and is no longer relying on information from the hard drive.
And of course the speed of the Internet, which I have no control over, and is where all of the data is coming from. That has to do with the incorporation and modification that I made to the research plug-in from OTC.

[CyborgBob]

C Load,


   I have to say. I can't wait to play with your plug in. I understand that you have put a lot of work into it, and you still have a lot of work to go. My question is can I do anything to help at all? I know I have no skills at all, but if there is some repetitive work that you could explain in a way I could understand. I would be happy... Just a thought, and really looking forward to taking a look.

CyborgBob 

[cload]

Hi Bob,
I still need to get through about 2500 lines, after that I was going to ask for help with weeding out the blockers.
Basically if you already have experience with talking to ultra Hal you should be able to detect the blockers.
When I get to that point, I will have a full explanation on how to detect blockers, then you can post them on this thread that I will start about this topic.
When I originally started there was five of us working on the AIML, we edited 57,000 lines and took out all of the obvious ones, it's down to about 15,000 lines, these are the lines and I am testing.
But the only true test is just having a conversation with ultra Hal to determine if any of the AIML lines are causing ultra Hal not to respond properly.
Sincerely, a data munching cruncher.
C load.
PS as soon as I get done with this learning packet you will find that ultra Hal will put you through a series of questions in order to understand more in depth about what you are talking about.
Not much different than the way that two people normally have a conversation with each other, and unfortunately nothing is ever perfect so there will always be certain minor blockers that you as an individual may remove.
A lot of that has to do with, no two people ever carry on a conversation the same way, in the same context or the same manner.

[Quartinium]

Cload, I'm really intrigued by your idea to get Hal to ask you questions.

 I was thinking about something similar where
Hal would open one of your pictures
 ask questions about the picture if he had never picked it before or
 chat about the picture with you if he had seen it before

I've been looking for Gretta Picture, I think this might help help me code something like that but I cannot find it.

Any thoughts?

[cload]

hi Quartinium
the individual that was working on the Gretta plug-in lost his serving site account.
So I decided to host all of his plug-ins on my sky Drive for everyone to download.
The pitcher retrieving plug-in should be there, as well as the newest upgrade to the Mark five plug-in.
https://skydrive.live.com/?cid=8eaf5f2f7fa962c9#cid=8EAF5F2F7FA962C9&id=8EAF5F2F7FA962C9%21112
Sincerely, a data munching cruncher, I hope I helped, use the link above to get to my sky Drive.
C load.

[cload]

sorry! about the no link above, it's all good now.