Hal forces a GetResponse, which has been setup in these plugins to be the start of the response, then it switches to Balabolka for the rest.
so I say "please research animals", and hal has to give a response from the hal engine (mandatory for Hal program), so i made it something short that it can handle. a typical response would be Hal : "Linking to research database". Then for the meat of the research response, (which is too long for Hal to handle) Balabolka starts (minimized), and reads the research about animals in the background.
the whole response could be given in the Balabolka program, but that would leave us still needing a mandatory GetResponse from Hal. what then?
If the GetResponse is something other than Hal speaking a line (such as opening a program or something) then Hal will do your GetResponse and add it's own Spoken line to it. I wanted the line that would have to be spoken from Hal to elude to the fact that it is doing research, and not a random GetResponse about research; such as -
Hal: "People like to study about cultures" or "NASA is a research facility"
which would (i think) sound strange in front of the research text about animals.
If you have any ideas about getting Hal to give a long response without using a method like this then please share your ideas with everyone. we would love to hear them. the problem is: no one has come up with a way to do this that i know about.
Of course "the cleaner the better", so i have always been looking for a way to do this only within Hal to no avail. Any help would be appreciated. Basically Hal would need to pass it's response to Sapi or Haptek one line/word at a time like a tts program does. What it does now, is dump the whole response at once, which overloads Sapi.