Digitizing Simplified you ask we scan Marc Holtman

Digitizing Simplified you ask, we scan Marc Holtman Amsterdam City Archives IS&T Archiving 2008 Bern

How it started Putting our inventories online Since 2004 all inventories of the SAA are available online All content is indexed by search engines like Google This system is being used intensively immediately after its launch So, are we happy now?

The Archiefbank “show me the images!” From the point of view of the user, online inventories are limited Consulting the documents still means: come and visit our reading room But our users asked: “where’s the button to view the images”?

The Archiefbank “show me the images!” In 2006 we decided to develop that button We called it the “Archiefbank” (or “Archivebank”) A system for consulting and downloading archival documents using the web The first version was launched in July 2007 At this moment it contains 3. 000 images And we have 25. 000 Archiefbank users accounts

You ask Digitizing at the customer’s request The SAA handles 32 km of archives So, where do we start? Research into statistical data learns: there’s no such thing as public favorites How long does it take to scan 32 km of archives? Therefore, in the Archiefbank the user prioritizes the digitizing 1 meter archive = 7. 000 scans Within a production of 10. 000 scans a week All public inventorynumbers can be requested for digitizing via the search system it will take 431 years to scan every piece of paper If we can process all customer requests, it can be concluded that we have no backlog in digitizing and the objective that the customer is able to consult items has been achieved

You ask Digitizing at the customer’s request The SAA handles 32 km of archives So, where do we start? Research into statistical data learns: there’s no such thing as public favorites Therefore, in the Archiefbank the user prioritizes the digitizing After digitizing the scans are available User friendly bach-office system to for all requests / communication manage alusers All public inventorynumbers can be requested for digitizing via the search system Users can buy 1 or more scans The more scans you buy, the cheaper If we can process all customer requests, it can be concluded that we have no it gets (1= € 0, 50 ps / 1000 = € 0, 25 backlog in digitizing and the objective that the customer is able to consult items ps) has been achieved

You ask Short delivery time Delivery time has to be as short as possible This can only be achieved when two requirements are met: 1. A streamlined, efficiently organized, large-scale reproduction process 2. A very fast and fool-proof web application

You ask Low costs Archival research easily runs into dozen or even hundreds of documents That means, our customers will need a large amount of scans An ordinary Xerox copy form a document in our reading room has to be the benchmark for the price of a scan This means that the production costs for digitizing have to be very low We developed a entirely new reproduction process with: Large scale and quality geared to the objective as its key concepts

We scan For archival research via the web The scans made in this process are made for the purpose of research by the user And NOT as a substitute for the originals Conservation of the original remains the major concern Still it has a conservation function, because the originals no longer have to leave the repository

We scan A broad spectrum of all kinds of archival documents The documents that are being digitized can have the following forms Small and large size (up to a newspaper) Bound and loose-leafed entities Card indexes Old and modern material Low and high contrast documents Text alone, text and image together Hybrid forms The same quality standard applies to all documents

We scan With a quality standard based on the purpose of the scans To keep costs as low as possible we let the basis for the standard of quality follow from the purpose of the scans: Consulting digitized documents for archival research from the screen or in print The standard of quality can therefore be stated as follows: Information legible in the originals must also be legible in the scans Do we use the Metamorfoze guidelines for digitizing on request? Price comparison (estimates) At this moment: No. We would like to, but costs are still to high. Archival, Metamorfoze: 0, 65 Euro Photo : 2 – 4 Euro Archival, Archiefbank: 0, 20 Euro

We store No separate, uncompressed images In order to keep storage costs low we store no uncompressed masterimages At this moment we use lossy (JPEG PSD 10) compression (300 dpi real size, 24 bits) The result is an image with: A relatively small filesize Excellent legibility on screen and print enough flexibility Images are stored internally and on our webserver (all images are online available) On the webserver derivates are made for zooming and contrastfilter

Straightforward approach The reproduction process Developing the reproduction process Always the complete unit is scanned Material preparation as simple as possible The originals will not be numbered Guidelines and best practices for digitizing often take no account of enormous number of scans to be produced when digitalizing documents No fragile material or material packaged in an unconventional way We developed and implemented a reproduction process that can manage an average amount of 10. 000 scans every week Constant production Contracting out the scanning to outside Mass digitizing on project basis also follow this reproduction process partners To mage the process we developed in-house a workflow system that dovetails perfectly with our existing systems and procedures

Conclusion Large scale digitalization of archives at low costs and quick delivery is possible if The image quality, compression and fileformat is based on user requirements The reproduction process is streamlined, efficiently organized and automated wherever possible The back- en frontoffice systems are user friendly and fit the demands and needs the customer and employee put on it optimally

And to end with… The Archiefbank is nominated for an important Dutch ICT award (Computable Awards) The other nominees are very big companies The one who gets most votes wins So, if you want to help us to win this award. . . http: //www 2. computable. nl/computableawards/stem/