Thanks Emily, for giving us a word to cover the matrix of issues that we have been wrestling with here, where we see tensions among different aspects of the four dimensions that we already identified as desiderata. One is the tension between aspects of "useability" and of "extensibility/adaptability" -- especially between the desirability of developing a standard early enough in the emergence of some cooperative endeavor in order to be able to (build tools that) enable efficient corpus building, and the danger of codifying so early that the annotation conventions (or the tool) enforce an analysis that delays scientific advance when the phenomena are not amenable to being conceptualized within existing ontologies. (This was the tension that Mark also noted in his comments after our report yesterday.) A second is the conflict between aspects of interoperability and extensibility/adaptability -- e.g., between what we had been calling "top-down" (or "ISO-type") and "bottom-up" (or "organic") approaches to the development of annotation standards, which we know see is another species of what Working Group 6 called "the balance between central planning and bottom-up initiative" (linking in the Wikipedia page on the metaphor of http://en.wikipedia.org/wiki/The_Cathedral_and_the_Bazaar). If your group has thoughts about how to couch a discussion of these tensions and how to link to relevant passages of the "white paper" that y'all are writing, we would very much like to hear them!

 out of  found this valuable. Do you find this valuable?    

Yes, it is used. Richard, do you want to take charge of this? The issues that we might discuss / document here include:
(1) the fact that the Unicode symbols are UTF-16 rather than UTF-8 and so muck up FOS-wiki sites and can't be used easily in some systems such as EMU and the EMU library for R
(2) the various ASCII-fications that have been developed (WorldBet, SAMPA, etc.) to get around the problems of the standard symbols being UTF-16, and who uses them for what purposes
(3) the arguments pro and con on broad versus narrow tagging for various purposes such as tagging for speech synthesis versus speech recognition versus sociophonetic studies, etc.
(4) inter-annotator consistency rates for different degrees of "narrowness"
(5) the whole question about alignment and how good (for the different purposes) various HTK-based aligners are
(6) the literature on consistency and tagging practices with child speech, etc.

 out of  found this valuable. Do you find this valuable?