Personal libraries > Technological handling of the project

Technological handling of the project

 Editorial system Hyperion, which was adapted for needs of the project is built on Java technology using technologies such as J2EE, Jakarta Struts and XML. It´s based on aplication server JBoss as well as well-established and efficient relational database PostgreSQL. It provides items classicication from many points of view and possibility of multiple ranging of them.

Web aplication of „personal libraries" is settled on server of firm MathAn s.r.o. which maintain that editorial system. In future it should be removed to server Jinonice what is the server of Charles University.

Structure of the digital library and the user interface were created according to special needs of providing an access to information from personal libraries.

For each of the personal libraries was made in the editorial system a special collection where were put digital copies of part of the documents from the personal libraries together with description of the original documents and the digital copies themselves.

In each collection there are also texts dealing with processing of the collection of the particular personal library, its history, its characteristics and its context. Static texts as „About the project" and "Actuality" are in a isolated section. Another isolated section has a list of „Some other personal libraries".

Metadata formats for making available and preserving special informations from personal libraries

After considering use of Dublin Core metadata set was chosen for description matadatas as more extended and therefore more suitable MODS (Metadata Object Description Schema) format and for proper consistance of digital objects and their good communication now and in future was implemented format METS (Metadata Encoding and Transmission Standard). As technical metadata was used recommendation of metadata set taken from schema PREMIS (Preservation Metadata) made for use of participants of Czech significant system for digitized archival documents „Kramerius". This recommendation was created by Jan Hutař, head of Departement for digital preservation of National Library of Czech Republic in 2008. Into the METS were embeded also some elements of schema MIX  (NISO Metadata for Images in XML Schéma/Technical Metadata for Digital Still Images Standard)  dealing with hardware and height and weight of the digital copies.

For metadatas dealing with law information were used METS Rights schema and Creative Commons Licence.

Implementation and use of format MODS, METS and other schemas and formats were processed according to guides, manuals and examples on websites of Library of Congress.

Long time archiving and their communication thru OAI-PMH protocol will be processed thanks to cooperation of Repository of Charles University, which uses Digitool of firm ExLibris. Digitool enables replication from METS/MODS format (a sample of metadata records METS/MODS exported from the editorial system was successfully tested for harvesting by OAI-PMH protocol). Successful was as well testing of convertion MODS to Dublin Core format (one more function available by Digitool). In case needed there is also a possibility to adjust such records according to the particular needs.

METS records of all the documents were imported into the repository of Charles University. This way there will be assured long time preservation of the digital copies which were made and their description in case the original collection of the personal library would be destroyed. As a strong institucional repository is the Repository of Charles University able to keep some development and use of emerging technologies developed to preserve digital content into future.

Drawing up the digital copies and their processing

Copies were made on scanner Epson, Canon CanoScan CS4400F (HP Scanjet 5500c in case of the Z.Fibicha library and CanoScan 5200F in case of the K.H.Borovsky library - as the scanners available on the places) and digital photo camera Canon EOS 400 (except those few copies which were processed by institutions itselves). The copies were made as JPEG format, with resolution 300 a 400 dpi, with ussual JPEG kompression). In JPG format they are also available to users in the web aplication.

From amount of copies were chosen the best quality ones which were after edited by Adobe Photoshop7 and Microsoft Office Picture Manager.

Metadata description

There was chosen such a set of description metadata that is able to preview to the user the most important information about the book and also about that copy and the significant items hold in it. There are ussual data taken from the ussual classic book description: title and subtitle of the book, number and title of part of a book with more volumes, name and surname of the author and if needed other contributors: translator, illustrator, author of typography, book-designer, graphic-designer and printer. There is aslo the original size of the book available.

Very important is information about location of each book (town, institution holding the book, name of the collection the book is in). As much as this is important transcription of inscription in the book or eventually any other contextual information.

At periodic there were added number and issue. Call Number which is filled in the editorial system is not available to the users.

Down to this description there are three level of resolution of digital copies with transcription up to the highest one.

Except these datas were filled according to the rules of MODS format into formular in editorial system another datas as: format of the digital copies, language of the document, subject key words (the most significant are such as: „personal library", name of the original owner of the collection (eventually another personalities noted in the inscription), kind of inscription). Another metadatas: name of the scanner and type of document. Each digital copy of significant page is described by title of the copy („cover", „inscription" etc.), type of object (still image), local identificator identifiing the copy in the system. Note on level of the book or each page were filled in field mods:note.

Organisation of content of the website

It´s possible to browse all documents of each of the six collections by the first letter of surname of author of the document (by the first letter of the title at those documents which are without author) or choose one letter and than follow the names of authors.

Searching is made by: browsing in list of individualized books of one personal library or lists of all the books of the library (made in Word of Excel) or by putting a request into a Search field in the upper left corner of the website. By the filter is possible to search in a particular personal library or all the texts in the website.

While display particular book user sees these information about the book: surname and name of its author, year and place of edition, publisher, number of pages, edition. Down to these are a few thumbnails of copies of significant parts of the document. After click on one of these is the user offered by larger thumbnail with information about the document and its original size, if needed also about other contributors.

If user doesn´t click on the description of the book but just on a page as a thumbnail, the page opens in full resolution.

At the full resolution of pictures there is a transcription of the inscriptions for feeling users more comfortable.