Hi, This raises many questions about the nature of what people are trying to do with EAD. First, are you trying to encode metadata about a collection or are you trying to encode a representation of a document (the extant finding aid) that contains information about a collection? Unfortunately, for a variety of political reasons, including a lack of common agreement on descriptive practice in the archival community, EAD is really an encoding scheme that provides a description of a document that describes a collection - rather than directly describing a collection (in art they call that derivative). In that sense it is one step removed from the collection. Is this what the archival community really wants to do when trying to provide information on the web to its end users? There seems to be an uncommon fondness for the finding aid as extant document in the archival community. Not an archivist, I would argue that a finding aid should be viewed only as metadata about a collection - and therefore is not, in and of itself, a document of instrinsic value - whose representation must be moved forward into the future (as perhaps we might want to move Shakespeare into the future). We can still capture the intellectual contribution of an archivist who has written a finding aid without capturing it in its original form. If we instead view the finding aid as a means by which important information was gathered about a collection in a form (the document) that is commonly used what really matters is the data contained within the finding aid. The representation of the extant document (the description of the description of the collection) is less important that capturing the metadata within the finding aid in a way that can be usefully manipulated and shared. If you want to take full advantage of the possibilities for machine processing, (search, retrieval, reuse in a variety of forms) then you want to focus on the content of the document rather than its form. You structure the data in ways that help capture semantic meaning (and thereby give clues to the computer for retrieving the sorts of information that you want). Really, the finding aid isn't the artifact that you are trying to represent. It is the collection, correct? The analogy, in a library setting: Rather than standardizing data when moving from a card catalog to a online system, you make a representation of the card from the catalog, idiocyncratic across various libraries. Certainly, there is intellectual effort in making a card catalog entry. It is however, the metadata about the book that takes precedence and not the intellectual effort of the cataloger (though what metadata is available is certainly her intellectual heritage). The idiocyncracies of the card catalog that take a back seat (Before anyone jumps down my throat, I know that finding aids often contain considerable intellectual content and so the analogy is imperfect). So, I would argue that thinking about an extant finding aid as anything more than the convenient container in its era for the information about the collection will inhibit the full potential of migrating data to a common electronic format. Secondly, the wonderful thing about XML is that you can do transformations that excerpt part of a document, reorder them , reuse in various formats. Bascially you can slice and dice your data in infinite ways - including making it look like the old finding aid. Given that you can do this with XSL, it seems that, regardless of the order in which you enter data, you can output it in any order. Thus, the order within the ead instance should reflect ease of data entry and machine processing (things like templating that allow a data entry person to quickly move through the process and identify missing information). And in fact, as a systems person, I would argue that a more rigid structure in EAD would have provided those trying to build access systems a number of simplified handles on the data. So, sure, if you view a finding aid as a document that in and of itself is an artifact, then changing the order would violate its representation. But, if instead, you think of it as a container (of its time) for information about the collection, then capturing the original order is secondary. What rises to prominence is the collection description -whatever form that takes. Liz Shaw Amy McCrory wrote: > Has anyone, in attempting to re-engineer legacy finding aids for EAD > compatibility, encountered resistance on the basis that re-ordering of > elements into series and other hierarchical levels would violate the > creator's original order? > > Amy McCrory > Archivist > Cartoon Research Library > Ohio State University