Say I have about 20 tags to find in a document and manipulate. What is going to be the faster approach, reading through the entire tree and handling the matching tags as it comes across them, or doing a #find on the document for each tag. Thanks.