Content digitizing techniques

The archives of the LDM separated into five parts: anthropology, geology, botany, entomology, and amphibious herpetology were collected by the NMNS.

Anthropology: historical relics of Yami tribe, 260 specimen s, 395 slides, 50 drawings, 3 tapes, and 50 manuscripts. The content include flying fish ceremony, piecing board boat, fairy stories and legend, costumes and decorations, religion, economic activities, and etc.

Geology: rock and mineral, 275 specimens, 542 slides, and 50 drawings. The content includes landform, volcano, marine resources, geological structure, and etc.

Botany: 90 specimens and 927 slides. The content included herbaceous plants, seaside plants, forest, shrubbery, wetland plants, and etc.

Entomology: over 100 specimens, and 991 slides. The content includes butterfly resources, beetles, grassland cline, monsoon forest cline, and etc.

Amphibious herpetology: 50 specimens, and 347 slides. The content included Green Turtle, snake, lizard, and etc.

The original archives were digitized into text, image, audio, and video files. Then, the LDM set a relational database management system to integrate the multiple data. The LDM proposed an easy way to produce exhibition elements from the digital archive. F or the content provider, the input is automatically incorporated into the associated database. The common component s in a hypermedia document include structural text, data series, graphic representations of functions and tabular data, geometric drawing, animations, video, etc. After the investigating, the LDM had completed about 50%

percent of all the collected archives.

In printing industry, the NSTM collected the printing artifacts (e.g., phototypesetting machine, typecasting machine, photoengraving machi ne, Chinese type writer, proof press, printing press) were belonged to 1930~1990. Moreover, the NSTM also introduced the history of printing industry based on manuscripts since the Shang Dynasty (1523-1028 BC, e.g., the seal). In the ancient locks, what th e ACMRC collected were since the Spring and Autumn Period (700 -403 BC). And the ACMRC

described the development progress of lock since the Stone Age that the people used Rocks to lock the cave based on the need of security. Then the collected data were cla ssified according to their era, region, purpose, type, special function, and etc.

Disappointedly, many printing machines moved to mainland China with the printery, and the new computer printer was invented since two decades. The exhibitions of techniques o f printing industry in TTRIT included the relief printing, plate printing, lithographic printing, and halftone printing. The exhibitions of artifacts of ancient locks were broad locks, pattern locks, necklace lock, shackle locks, combination locks, and etc . The digital data were made into referable and researchable explanatory data.

The TTRIT had completed about 90% of all the collected archives.

3. 2. The standards of metadata

The LDM let each webpage be artistic; the common database was established into the digital museum content management system (DMCMS) (Figure 1) to integrate the text illustration, image, classification, sounds, and archival information would work more fluency [4]. For the LDM was a prototype information system; the existed standards of metadata (e.g. OAI, SOAP, OPENURL) would not fit in the virtual exhibition system. For various archival types, standards of archives, increasing data, modifying data, and retrieving data, the DMCMS could integrate text description, audio and video files , archives link, and database into the dynamic programs library.

Such an editing system (DMCMS) for extracting multimedia objects from the database could quickly compose them into exhibitions. The administrators could input or choose a title, a set of pict ures, some text, an audio clip, and a particular style sheet (in XSL form), and an online exhibition was completed.

Then a dynamic page would be automatically converted from the XSL files into an XML document automatically.

Figure 1. Digital museum content management system

The management system of metadata of TTRIT used the Metairiti software based on XML. The Metairiti software provided several functions: input Document Type Definition (DTD), set database, define the schema of database, catalogue the me tadata, management and description of digital data, and etc (Figure 2). The advantage of the Metairiti software was the distributed management of database could provide a managerial module to integrate different modules of various users demand, so that the researchers located at different location could add, set, and revise the metadata together.

The Dublin code was used to be the standard of metadata of TTRIT. The metadata format of Dublin code was used as the common standard in the projects of National Digital Archives Program (NDAP, over 200 projects). The goal of use Dublin code was to form the union catalogs by connecting with other digital archives via the same metadata format. There were some referable examples of the record rules of each explanatory data to input every column, such as an image encoded based on its era, professional field, name, serial numbers, materials, size, purpose, and its details. The structure of the metadata was divided into four parts: data for basic demand, data for research, data for management, and data for children version.

Figure 2. The module of setting the metadata

3. 3. Retrieval function

The DAL developed natural language f requently asked questions system (FAQ) and collected over 100 questions about the Lanyü Island. The users could input oral question (e.g., what is the special insect in Lanyü Island?) as inquiring condition, then the question database would show the answer. If the system did not show the answer, then it would store the question and return to the s ystem manager by E-mail, so that the system manager could look for the answer from the experts and add into the database.

Based on the common database, the retrieval function was to create a knowledge database and the users could use keyword to gain the in formation. On the other hand, the users could explore the LDM in the tree structure of linear level representation in the main catalog and sub catalog was convenient and distinct. The FAQ system did not show the answer for students input questions but the retrieval result of using key work showed several hyperlinks connect to the web pages. For the difficulties of comprehension, the students did not find the relation of main catalog and sub catalog in the tree structure of linear level representation.

Retrieving in the TTRIT, the users could retrieve the information by input the catalog, denomination, era, or key words of archives (Figure 3); the result of retrieving information could show more precise than the result of using other searching engine (e.g., Y ahoo, Google). The key words of images were embedded in the text mining, and the key words of video files were related the file name. The key words were classified in metadata system for fast returning the retrieval result.

Figure 3. The interface of ret rieval functions

3. 4. Modules

For producing web pages constantly of the gradually increasing data, the LDM used XML and XSL to set the automatic producing webpage system that integrated image, text, audio, background, or other special effect to be multi -layer exhibitive modules. The multi-modules were flash-based (Made by Macromedia Flash program) and included four important modules were as follows (Figure 4):

Modularized Flash Media -Handler: the access and management of the flash -based media-handler that was a SWF file with embedded action script for presenting a specific type of dynamic media.

Multi-layer Presentation Composer: combined the XML elements with a set of flash -based media-handlers into an integrated flash animation and assigned each media -handler a different layer.

Exhibition pages: created the title of webpage, blocks, inline text, and media elements links of an exhibition page. The text was encoded in HTML format to produce various presentations.

Sitemap Organizer: for the content of page was independent to the navigation system, all the links were inter-page and specified inside the exhibition sitemap. The sitemap organizer would assemble and structure a sequence of exhibition pages into an exhibitions sitemap. Then the exhibition pages co uld be re-arranged and re-assembled into new exhibition sitemap without modifying the content of page.

There were three central issues for producing web pages flexibly and easily included in the methodology of LDM: (1) an easy way to produce exhibition ele ments from the digital archive ; (2) once an exhibition was produced, an easy way to create different presentation styles suitable for different users; (3) an easy way to re-arrange an existing exhibition to create new ones for users with different needs.

Figure 4. Flash-based exhibition modules

The TTRIT set three modules base on the demand of different users; the each module was as follows:

1. The demand of manager: the security, exchange archives with other museums, maintain and add database (Figure 5).

2. The demand of researcher: for researching, referring, and correcting the content of archive.

3. The demand of viewers: for learning, exploring, investigating, and accessing information.

During the process of output, the three modules were managed and set by the Metairiti software. The module of data management was designed to store automatically, that is, the system would store the explanatory data automatically in the temporary database and sent a message to the manager. The manager would consult othe r professionals to evaluate the accuracy and relevance of the explanatory data then bring into the update database. The researchers could log in the system and review the explanatory data or provide their comment. Finally, the finished data would become th e data of the viewer module to exhibit.

Figure 5. The managing module