Jon Reynolds' post serves, for me, as a caution. I have no idea (until SAA this fall) what programmatic decisions are driving the conversion of Georgetown's finding aids, so these comments aren't in direct response to Jon, just prodded by it. I think that the wholesale scanning of existing finding aids in any repository is a tactical error. Even if current OCR technology made it a breeze to do it, it seems to me a wasted opportunity for a repository to simply take paper finding aids and convert them to a digital format. There are so many issues concerning access, what future developments might hold for the use of digital delivery technologies in archives, whether or not our "audience" is the same when we make our finding aids available in digital format over the Internet, and whether or not we can approach that altered "audience" in exactly the same manner we've been approaching the researchers who happened to walk through the doors of our institutions to look at our finding aids in the past. A digitization program probably, IMHO, ought to be looked at as a long-term commitment by a repository to re-examine thoughtfully how collections are and might be used, what sorts of "virtual" use might collections support. Maybe there ought even to be some sort of feedback mechanism where we ask for and listen to users of our collections and finding aids--traditional users as well as those for whom use is made possible by this new delivery medium--and incorporate that feedback into our programmatic vision. In just the repositories in the area in which I currently live and work I can think of a half-dozen or so totally different kinds of finding aids--different in basic approach to the organization of information about a collections, as well as in style and layout. Within many repositories there are often radically different types of finding aids, and this depends only on when they were created and how frequently the collection they represent gets used. These sorts of things might be fine when the archivist is physically present to explain to a researcher/casual browser what different things mean; however, in an environment in which the two aren't in the same physical space, the finding aid has to do a lot more work on its own than many repositories currently expect it to do. These list discussions on the topic of the BFAP are exciting to me. I'm learning a lot about how the rest of my colleagues approach information and the provision of access to it. I, too, would extend the invitiation to Daniel Pitti, or someone from the BFAP team, to participate in the discussions. Perhaps after the Mellon-fellowship team convenes in Ann Arbor this summer some list participants who are involved can move the discussion onto this list. I would hope that interested archivists are given a chance to provide some input into the process during this crucial phase. It seems to me that it would provide some basic understanding of SGML, the developing DTD, and how they will contribute (and perhaps, who knows, revolutionize) the delivery of information about archival collections and,eventually, the collections themselves. In addition, it might make SAA's job of promoting the use of SGML for finding aids among archivists much easier once the DTD is accepted by the international standards community. The more we talk about it now, the less of a mystery it'll be in the future. [log in to unmask] Bill Landis Cranbrook Archives Cranbrook Educational Community Bloomfield Hills, Michigan [log in to unmask]