using a [tracecall] object you can trace back the original cause of messages.
This is fairly esoteric to me. Could you give a "real world" example of what such an object would be used for? What need does it fulfill in your own patching strategies? Why did you make it?
I would mostly use this object to make a test patch to find whether that object works. Seems pretty circular so far. I would make such a test patch because it also tests a part of DesireData's internals that [tracecall] uses. Basically, [tracecall] itself is just an internal component. I am exposing it to patches based on the belief that internal components should be testable "externally" (as much as possible using .pd files). This itself is based on the belief that programmes should be tested by other programmes.
I will have a presentation at pd-convention about testing methodology and other connected topics such as type systems, documentation, XP, OAOO, etc. So far the title is:
"A Type Theory for the Documentation of PureData"
_ _ __ ___ _____ ________ _____________ _____________________ ... | Mathieu Bouchard - tél:+1.514.383.3801, Montréal QC Canada