PHP, eXist, XQuery

The development of the Map of Early Modern London website was broken up into three distict sections: establishing a standardized encoding format, creating a queryable repository for the encoded documents and providing an infrastructure for users to access these documents in a variety of ways.

Text for the website was marked up in TEI P5 format and then stored in an eXist XML database. These two technologies work very well in tandem to provide an extensible, standardized and robust way of storing the text for this project. PHP along with XML-RPC was chosen to provide the infrastructure for the website due to it's ubiquity on TAPoR servers and its rapid prototying/development time.

This project has created an extractable infrastructure for providing rapid access to data stored in an eXist database. It also provides concepts useful for creating a web of links between disparate articles. This allows users of the website to traverse information in a way that is natural and intuitive. The final stage of this project will be the creation of a system for handling submission and review of proposals and will necessitate the addition of an administrative utility.

XQuery, eXist, Cocoon, Tomcat The development history of the Graves' Project is documented in detail at the web site URL above. The present phase is to make the transcribed Graves Diary documents searchable online. To do this, the project team has deployed the open source native XML database system eXist, with general development focused on improving the speed and efficiency of information retrieval. Additional development is focused on stabilizing and standardizing a format for the rendering and placement of enclosures (some multi-page) alongside their digital images. The primary outcome is a programming template based on standardized formats that will automate the process of rendering and placing enclosures accompanying digital images. This tool will be developed through Open source encoding, and made available for general purpose use.

TAPoR generated image mark up tool Dr. Carlin and Martin Holmes developed an XML model which allows for both the text of the annotations, and the image areas to which they correspond, to be encoded in a single document. The markup of text is based on a subset ot he TEI P5 guidelines. The TEI encourages customization of its guidelines to accommodate for a wide range of implementations, an approach this project demonstrates. The markup of the images has been achieved with the Image Markup tool, an application progam. This project has been the vehicle driving the development of the Image Markup tool (see full description of the mark-up tool). The project team and researcher will continue to work together creating features and modifying its utility based on the demands of the project.

The file format is TEI P5 XML, with the image data and annotation areas defined in a block of SVG code inside the document (making a multiple-namespace document). The files produced so far have been assembled into an eXist database, which allows them to be displayed in a browser. The user can click on annotations in the text and see the relevant part of the image highlighted. We intend to to do further work on various publication and display formats based on the eXist database.
The tool, developed exclusively using open source code libraries, is an open source application. The project has been the vehicle driving the tool's development, and will continue to be the primary application for the project, though it has already been applied to other UVic TAPoR projects (such as VI history) and likely will be used in others as well.

Other significant technical developments associated with the project: 1) saving the state of the Image Markup tool on closing so that on re-opening the application the program returns to an identical state, 2) detection of installed browsers, 3) use of reflection on the application structure to support translation of the interface.

XML technologies, Cocoon, PosgreSQL The ISE website provides an opportunity to showcase the integration of many advanced technologies, since Shakespeare's works are unrivaled in their combination of complex textuality and the multimedia experience of the performance of his plays.

It is a principle of the ISE to employ only open-source software.

At the core of the project is Cocoon, a "pipeline" that transforms XML and XHTML files into dynamic web pages. The site is at present a combination of some static pages, and a large number of pages dynamically generated from a range of web applications. In future the static pages will themselves be delivered dynamically. The design of the site will thus be fully generalized, so that it can be modified or added to without further development of software. The major web applications themselves employ advanced technology:

1. Shakespeare in Performance is a PosgreSQL database delivering text graphics, sound, and video records of Shakespeare productions around the globe. Fully relational in structure, the database resembles the International Movie Database, with the addition of actual artifacts of performance as well as data on those taking part. The front end of the database is PHP, which delivers XML to Cocoon. Under construction is a back end that will permit error-free data entry from a web browser.
2. The Text Viewer is an application that combines multiple XML files of the texts, their annotations, and associated collations into a single viewer interface. Since Shakespeare's texts appeared in multiple formats (Quarto, Folio) and are being re-edited (Modern), the XML files necessarily involve the use of overlapping hierarchies of information. This technical hurdle has been overcome by an ingenious adaptation of the concept of XML namespaces, such that the separate trees can be extracted elegantly from a single master file. In due course we will generate conventional TEI-conformant XML from these same master files.
3. The Facsimile Viewer is an extension of the Text Viewer that permits multiple views, browsing, and searches of facsimiles of the plays. It is also based on a PosgreSQL database. This feature provides more power and flexibility than any other related site on the web.
4. The Links Database provides links to over 700 external sites of interest to readers of Shakespeare. This is an XML database, and is the central repository of all links referenced in on the site.

None cited to date

XML Technologies

XML technologies, TAPoR image mark-up tool The prototype phase of the project involves selecting less than a dozen plan documents and digitizing those to image files. It also involves compiling a list of elements and attributes to use in the markup based on the image files and then estimates of the amount of work required to provide markup appropriate to various audiences (scholars, construction professionals, general public).
The research phase of the project will involve reviewing, selecting and cataloging the relevant archives for the project. The digitizing and markup of the drawings will be based on the procedures and schema developed in the prototype, likely via the tools or coding routines developed to facilitate this process. Simultaneously, we will be working on the front-end interfaces appropriate to the intended audiences to allow them to interact with the material. Finally, we will populate a server-based database and test the front-end interfaces.
Some funding (by way of digitizing commitment by a commercial operator) is available for the prototype phase of the project. Further funding would need to be identified for the research phase. The prototype phase of this project focussed on transferable coding schemas and tools to facilitate the encoding of the images and produced useful information to guide the subsequent phase of the project. Some of that information is specific to building plan documents, and other information is of wider applicability. This project clearly addresses the UVic TAPoR node's mandate to explore multi-media projects.

MySQL, PHP, XHTML The database system developed in 2004 for phase 2 of the project was designed to emulate the real-life work flow for each document that is transcribed and ultimately published on the website. The system, composed of php scripts and a MySQL database, incorporated user authentication, and a document submission function made up of steps that intuitively grouped required elements for each document (such as citation elements) and the ability to upload a soft copy of the transcribed document that ultimately passes on to subsequent steps. Document processing is made up of listings indicating documents available for certain tasks in the appropriate processing order (such as translation, or markup), the ability to work on documents with a check-out and check-in process in place, and the ability to restrict skill-specific tasks to users with the right credentials. An automated script within the system is also setup to generate the project's websites on the fly by laying out each site's directory structure and creating HTML files containing text from within the database, thus eliminating the manual steps that are required to perform this task every time there is a revision made to the sites' contents.

To make the system complete, a flexible search function allows users to sift through thousands of documents by titles, authors and keywords; a custom administration panel that allows for user administration, progress reports for each website complete with document and word count as well as work flow status statistics; and the ability to override certain aspects of the work flow in order to facilitate work among project members are integral parts put in place.

Currently, in phase 3, the database continues to expand to accommodate the new projects and the needs of the Great Unsolved Mysteries in Canadian History team, including revisions, enhanced automation of tasks and finer tuned search and retrieval capabilities.

Files must be marked up to XHTML manually and uploaded to the database, and this work continues daily at HCMC. Additionally, the project has design and image editing needs, which are also met mainly by staff working in HCMC and elsewhere.

Currently, the project is using TAPoR resources, which are listed as in-kind donations in partnership with the Canadian Heritage grant funding. HCMC and TAPoR staff are providing some technical supervision and guidance. HCMC staff were key partners in the initial design, development and implementation of the database.

The database has some cross application for similar projects of this scope and purpose. The database will continue to be upgraded and expanded throughout the lifecycle of the "Mysteries" projects, now in the 3 phase of development, with 3-5 phases more to follow.

XML technologies, to be determined

XML technologies, Cocoon, Javascript, XHTML The schema has been developed and Oxygen is being used for the XML mark up, and an eXist database will generate statistics about the data, and present it through a website.

The mark up focuses on distinguishing Hirigana, Katakana, and Kanji characters via seg tags and place attributes. Currently work is being done on the feature structure array, and so far 6 types of documents have been developed with 10 projected. All of the original sources (newspapers, magazines) have been transcribed into Unicode, and the project focus is on the mark up of the data and the creation of the database and search engine.
The coding libraries can be applied to other projects potentially using non-western scripts in a Unicode format. The transferability lies mainly in the knowledge gained and its application to future projects involving the creation of word counts and frequency lists in Asian scripts. It serves as a prototype for potentially very large databases and the mark up of Asian script materials.

The second outcome involves the project's role as a test bed for identifying and documenting problems with the underlying eXist platform. The Katakana project has assisted us in pinpointing eXist platform development issues at a macro scale, with the end goal to create a stable eXist platform on which to build projects such as the above. Currently, TAPoR UVic is contracting an eXist developer to implement necessary changes (see eXist development for more details) for this project's (and others') future development.

PosgreSQL, PHP, MapServer, ka-Map Existing data has been ported from an ASP/MS Access database to a php/postgresql database, with added functionality. Database development focussed on creating an abstracted set of php functions that can be re-implemented in many different contexts that use php/postgresql. The most recent development demonstrates the use of a map interface to query the Postgres database. We are working on various features in that user interface to allow grouping and filtering of results. The objective is to allow the user to specify what kind of information she or he wants and to click on or select an area in the map and get back census and similar information related to the mapped location. Transferability to any project that intends to gather vector data as ancillary information and use it to filter results. Its potential use at the University of Victoria has been identified for the following projects currently under development or considered for future development: "London Map Project", "Towards Virtual Victoria & 3D History: A prototype for multidimensional historical representation", and "Creating an historical Geographic Information System [GIS] of Victoria, 1901 - 1902", and it may be suitable for the Maclure project as well.

Other outcomes include documented research into:

• porting of proprietary databases into open source databases, including integrating multiple databases into one, and bridging steps for a more streamlined process.
• creation of an annotation tool that allows outsiders to comment on the data and make their comments readable to the world in a wiki-like fashion.
• phonetic spelling engine for the queries within open source programs such as PosgreSQL and PHP.
• creation of tools for functionality for handling binary objects (still images, sound files and so forth)
• user interfaces for the editors and public to allow them to work effectively within an open source environment