Looking for some recommendations on commercial s/w to specify the functional requirements for applications. The applications I'm dealing with are for a wireless wearable voice-activated data collection device. Its inputs are commands spoken by the user, and received data tables via TCP/IP. Its outputs are "spoken prompts" (text-to-speech through a headset), and processed data strings via TCP/IP. The spoken prompts can intermix fixed text and variables (based on any data input), in any way. The processed data strings, likewise, can be any mixture of inputs in any format (fixed-length fields, delimited variable length fields, with or without fixed text). The processing flow involves a script of spoken prompts, with a pre-set list of accepted responses to each prompt, and pre-set data transmit/receive points in the script. The processing flow can branch, loop, and do other logic based on any input (whether spoken command or received data). I've looked at a couple of UML packages already, but they're basically glorified Draw programs, which isn't what I need. What I do need is something that lets me actually model out this kind of data flow & processing application -- define the type and handling of each input & output element -- and model a chronology of user interaction and processing. The data element modeling needs to be fairly detailed, going all the way down to length, padding, and justification. It needs to tell me when I have a loose end or a conflict in the model. Anyone have any recommendations, or even good leads? -- JDM Properly speaking, there's no such thing as a "bruised" ego, or "injured" pride. The correct term in either case is "lanced."

go to office max and buy a reem of paper and maybe a mechanical pencil, those work wonderful. Personally i like the one with the rubber grip. Sorry man thats the only thing i can come up with other then telling you to write the program yourself then use it, and sell it to me :)

For the intensive communications requirements, I don't think that you will be able to get by without having a computer on each unit. There are software with similar functions for telephone systens. But most use the dial tones to direct the decision tree. Voice recognition is not quite good enough to do what you want without extensive training. I think there is software to evaluate the decision tree, but you are the one that have to decide whether the outcome of the answers given is valid.

Wizard-fred, can you point me to the software you mentioned that would evaluate the decision tree? It sounds like it might be what I'm looking for. The device and its applications are already a reality -- I work for Vocollect, we've built and shipped over 75K of the boxes, and created several dozen custom applications to run on them. -- JDM Properly speaking, there's no such thing as a "bruised" ego, or "injured" pride. The correct term in either case is "lanced."

Sorry don't know the name of the specific software. Since you make the boxes and some of the the applications, you should have some knowledge of the branching capabilities. You don't want any dead ends. May be the ability to back up and try another branch. Maybe the paper and pencil approach is better to limit the choices of the system.

No problem then -- and thanks, having the term "decision tree" to search on, led me to some helpful trial-demos. -- JDM Properly speaking, there's no such thing as a "bruised" ego, or "injured" pride. The correct term in either case is "lanced."