Our database approach is very aggressive. To explain this I’ ll provide some background. One way to build a database is to scan pages and then convert the text using optical character recognition (OCR) into searchable fields. At first, we attempted this method, but OCR is not perfect, particularly with older texts and archaic type faces. The result is that scanning works well if you want to create a picture of a page from a book, but not so well if you want to create a sort-able, searchable database of information from that book.

Another alternative, and one accessible to anyone who uses ABE, is to employ a series of fields, typically up to six, and to input data. These fields are searchable and these searches are very fast. You need to know what you are looking for. When you do, you can find it quickly. Our approach also employs fields. We parse our data into 49 fields and manually enter the data. This is both expensive and bibliographically complex. That is why the majority of our team are bibliographers with backgrounds in library science and special collections work. But what becomes possible, over time and with patience and the help of our members, is the grand slam of bibliographical projects the reconstruction of the past. Here is what we are going to try to do. We begin by developing a comprehensive list of bibliographical sources. We then add their information to the database. Parsed into very small pieces, we then use a complex set of search combinations to extract from the data information that essentially has been lost to view to all but a few people. How complex? We provide 20 searchable fields all of which are searchable individually and in any combination with other fields, some of which accept ranges. The number of theoretical possibilities is enormous. Over time, we will develop screens for the most popular methods of searching. In the beginning, members will use our tools and their imaginations to develop their own sorting parameters.

Having said this, I need to quickly tell you that it will take time for our bibliographers to elaborate the fields in our database to meet our high standards. Here is one immediate possibility. Search a date range: say, 1775 to 1830 and by a particular printer who may have printed material in 3 or 4 locations. Instantly, you obtain a rich list of material this printer produced. Youll be able to bring up everything we have and then develop a list of potentially collectable books. With this list created, you can then upload these book details to a variety of listing sites as a wants list. Many don't charge and will let you keep the list posted for a year. You may or may not buy what you find but it is certain that material will come up. In doing this you will be pursuing a personal goal known only to yourself. Over time, material will emerge from unexpected places.