I have not been following this closely, so let me make a guess at what the basics are, and in so doing through my little 2 kopecks worth in. If there were a combined, heavy library with both circuit element symbols and footprints contained within a single grouping for a single device, part number, group of related parts, so that from either gschem or pcb you could search for the root name of that part/device and be given a list of possible symbols to use in gschem (I know Atmel Atmega parts have differing numbers of pins based on DIP, TSOP, etc) and then the matching footprint (or footprints, if several package/mounting styles or sizes could still fit) would be stuffed into the element attributes. Or, if you were working directly in pcb, you would access the same library, search for the same root device name, and be given a list of all known footprints for it ( and possibly stuff the corresponding symbol in the element attributes incase you ever wanted to go backwards to gschem). This group, bundle, set of data might also include spice model data for the part (possibly once again linked to a package style). In either case, you could open up the linked data in either gschem or pcb to 'heavy up' the item you are working on to export the data to some other tool flow (does pcb to spice to verify the design make sense?) As for names, data pack would be similar industry standard data sheets. Technically, these would be data nodes or data branches or data clusters. You could even call them device seeds because they seed desired data into whatever tool path they have data for and you desire. Mike
_______________________________________________ geda-user mailing list geda-user@moria.seul.org http://www.seul.org/cgi-bin/mailman/listinfo/geda-user