Hi. I've been following this list for a while now, because as a data modelling kinda person I'm interested in formats, conversions, metamodels, their mappings, and the like. I'm afraid to say the slow development of DFDL has disappointed me. To me it seems quite a number of implementations of DFDL should have been out there already, yet they aren't. I think the main reasons are that people in this working group are: a) concentrating too much on the details of specific binary formats to be parsed/use cases instead of a generic parsing framework for common binary data as a whole, b) paying too much attention to the XML encoding of the language, instead of its semantics, or its possible alternative encodings, c) spending far too much time meeting, specifying, theoretizing, and downright bickering-in-committee, instead of piece-meal implementing the most useful parts of the standard, d) contrariwise, optimizing the specification prematurely, e.g. with the hesitancy to declare multidimensional arrays as a first class citizen even if there are unresolved issues, e) doing all of the real work over the telephone or in person, so that low budget, practical minded people cannot contribute meaningfully, f) failing to show other people what the worth of this whole effort is, and thus failing to liaise efficiently even with the Grid Forum, and thereby, finally, g) failing to engage the broader circle of developers, managers, curators, nerds, and whathaveyou, who might be interested in this sort of thing. Is it at all possible that this effort could somehow move into the mainstream in the near future? That outside forces, like the FOSS community, could speed up the transition? In all, that some of the above peeves of mine could be rectified? -- Sampo Syreeni, aka decoy - decoy@iki.fi, http://decoy.iki.fi/front +358-50-5756111, 025E D175 ABE5 027C 9494 EEB0 E090 8BA9 0509 85C2