Location

§ 5 We first contacted the owners of two large private archive collections in order to introduce our project and gauge their possible reactions to it. The response was positive and led to other contacts. Our classes at the university provided us with another possible source of interesting documents; in the discussions of the project, some of our students told us of the existence of other small individual archives belonging to wealthy farmers, who kept these documents as a guarantee of their rights. We also heard from owners of historical documents who were keen to have these materials published. As a result, we can divide our materials into two kinds: well-established, well-organized archives, and single documents. At present the Arquibanc project is working with 33 archives at various stages of development.

Recovery

§ 6 In some cases the archives are well preserved and well arranged. In these cases, the recovery process involves digitization and little else, whereas in the case of documents in poor condition their recovery entails a great deal of work. In the latter case we also suggest better ways of preservation to owners.

Arrangement

§ 7 Some archives were already very well organized. The owners of the Fontcuberta archive, for example, an important heritage archive with seventy linear meters of documentation from the tenth to the twenty-first centuries covering the counties of Osona, Alt i Baix Empordà and Vallès Occidental, had already added indices, master books, and so on. But in other cases we have organized the collections following the indications of the ISAD international system (Conseil International des Archives/International Council on Archives 2000) specially designed for heritage archives (Fernandez 1991; Gifre, Matas, and Soler 2002).

Dissemination

§ 8 In addition to the distribution of documents in printed format, online databases constitute the main tool at our disposal. Online databases can provide access to materials preserved in small or privately-owned archives which are hidden or difficult to track down. We currently have two databases: Scripta, which deals with the vast Fontcuberta archive, and comprises three sets of documents, and Memoria, the Arquibanc database, which includes documents from a variety of sources.

§ 9 The databases contain public, semiprivate and private registers and fields for the different kinds of materials or collections. They are hosted on a University of Barcelona server, where they are properly maintained, and can be accessed through the Arquibanc research project website: http://www.ub.edu/arquibanc/home.html.

First objective

§ 10 To fulfill our first objective – the location and evaluation of private documentary sources – we contacted, first of all, the owners of those archives which were already arranged in order to draw up cooperation agreements to cover the study, classification, description and cataloguing of the materials. As a result of these first contacts other owners asked us to consider including their documents in our project. Some of our university students also alerted us to the existence of new archives covering a surprisingly long chronological period, from the Middle Ages to the twentieth century. Because of the huge volume of the material available, we were obliged to make a selection. Among the materials we chose was the Fontcuberta archive. We also decided to compile and study materials in which we considered the risk of deterioration or disappearance to be high.

§ 11 One of the most important tasks in this first stage was the appraisal of the dimensions of the materials and the difficulties they presented in order to plan the work teams and the time necessary for their study.

Second objective

§ 12 To fulfill the second objective – the recovery of documents at risk – we digitized all the materials to prepare them for study and also to preserve the ones at clear risk of deterioration or loss. In many cases, we advised the owners of fragile and poorly preserved documents to entrust them to the National Archive of Catalonia, if they themselves were unable to supervise the restoration needed for adequate preservation.

Figure 2: Some of the documents collections are in very bad condition.

Organization

§ 13 With regards to organization, there is a clear distinction to be made between two sets of materials. The first is the Fontcuberta archive, extremely well organized and described according to the owners’ criteria of use, which is to oversee and manage their patrimony. In this case, no intervention on our part was required. The second is the body of materials that had not previously been classified or described. Their inclusion in the database only partially relieves this situation, but is a great help in carrying out a classification of the documents in compliance with the ISAD(G) (General International Standard for Archival Description) regulations. Once the process is concluded, the documents can be accessed using the database’s search engine or through the documentation classification chart.

Dissemination

§ 14 Our priorities are the needs and desires of each owner. Owners do not usually object to the processes of digitization, systemization and description, but on the question of providing free access to the documentation, their opinions vary widely: some stipulate that the publication should be exclusively in printed form, while others opt to restrict access in different ways. We must not forget that we are talking about private archives situated in private homes. To deal with this situation, the owners and the University of Barcelona sign agreements stating that access will be via the Internet, and will be supervised by the database’s administrators. Our main dissemination tool is the project’s website (http://www.ub.edu/arquibanc/home.html), where the databases containing the materials can be accessed. However, we also plan to publish much of the material in printed form, and in fact two projects of this kind are currently underway (Cantarell, Comas, and Muntaner 2011; Baiges, Cantarell, Comas, Piñol, and Soler).

§ 15 This is where the value of the database is particularly evident. As we noted in the introduction, a large amount of historical material remains unpublished. We feel that any instrument that enables the scientific community to consult these materials is valuable, since it may lead us to reformulate or qualify our previously held ideas. But our aim was not to create an instrument to replace the critical editions of documentary corpora, which we consider to be essential. Indeed, in many cases, so much of the documentation is unpublished that these editions cannot be carried out because of the technical difficulty and economic cost involved. The challenge, then, was to design a tool that was in a way comparable with traditional editions and could at the same time complement them and help to provide new insights.

§ 16 So, our starting point was the belief that this online database should not be limited to a repertoire of document images, but should contain the standard elements of diplomatic editions of documentary corpora. At the same time, it should be able to generate indices to aid consultation. Finally, another key feature of the database, thinking in particular of large-scale research projects, is its ability to promote teamwork. Large teams of researchers would be required for projects of this size.

Design

§ 18 Each document is given a register comprising all the fields necessary for identification, description and classification. These fields can be easily created and defined by the database editors without the need for specific training. Therefore, without the aid of computer technicians, researchers can adapt the structure of the database to meet the particular characteristics and objectives of each project.

§ 19 A special field in each register contains the image(s) of the digitized document in a readable, downloadable format. If there are multiple images, they can be consulted sequentially and as thumbnails. In fact, the size of the images was the first problem we encountered in the project. Especially in the case of the parchments, the size and the state of preservation of many of the documents generated a digital image that was too large to be included in the database. It has been very difficult, and in many cases impossible, to obtain high-quality, high-resolution images of an acceptable size (not more than 1 Mb).

Figure 3: Screenshot of the database interface showing the image thumbnail.

§ 20 Initially, this caused certain problems for browsing, but these have been largely overcome thanks to the improvements in web technology in recent years; however, we still have the problem of server space. The small number of large format documents in the Memoria database do not present difficulties, but for larger projects such as Scripta, which contains the documentation of the Fontcuberta archive, we have hundreds of large format documents, which increases the required storage space and prevents the proper functioning of the tool due to the size of the images of these documents. In these cases, we supply researchers with a high resolution digital copy on request and, if they wish to publish the facsimile of the document, we can provide an image that meets the printer’s requirements.

Indexed fields, data searches and the generation of indices

§ 21 The database editors can create all the indexed fields they need for the objectives of each project. In the case we are describing, the fields are the following: signature (single and required field); date and place; document type in the case of public fields, and collaborator, image control, and state of revision in the case of private fields, that is, fields accessible only to the database editors and administrators. Private fields are useful for internal control of the state of each register, because they provide answers to important questions: who is the author of the description, has it been revised and if so by whom, who is in charge of the digitization, compression and publication of the image, which registers contain images and which do not, and what is the level of access of each document. They also provide access data such as the date of creation or modification, number of visits, and so on.

§ 22 All these fields can be explored using the browse option and can generate the corresponding indices. All the fields, with the obvious exception of the image, can be explored with the search option. As we noted above, the search engine of the databases allows us to identify all the elements located in any of the fields and produces a variety of lists according to the search results obtained. This means that each and every one of the words contained in the abstract of the document, whichever language was used (Catalan / Spanish / English etc.) can be found by the search engine. This requires that each document has attached to it an abstract worded as broadly as possible, in order to enable the user to refine the search at maximum. Searches can be made for proper names (name and surname, or name and rank) using the search for adjacent words option, political or administrative posts, and so on. The search engine also allows the use of dummy characters (? or *).

Data entry and teamwork