----- Original Message -----
Sent: Wednesday, June 25, 2008 1:58 PM
Subject: ISIS3 - Message from Russia - Part II
Dear Colleagues
I am sending the attached document mentioned in our previous message.
This document was sent by
Dr. Vladimir Yu.Zitserman, PhD, Head of data bank laboratory
Dr. Michael S.Trachtengerts, PhD, Senior researcher
Thermophysical Properties Center, Joint Institute for High Temperatures of the Russian Academy of Sciences, Moscow, Russia
Regards
Ernesto Spinak
The general strategy on development of ideas and the technologies in information system ISIS
1. General view. The concept of development of an information product should be based on: (1) available experience of its use in various fields and (2) detailed studying of new challenges and needs of information community.
2. The analysis of practical use and results has shown efficiency of system ISIS. It took place in quite certain niche among vast family of the different products that apply for storage and processing of information. The system is effective as a DB instrument for descriptions of any kind and purpose documents. The tag system set allows for user to identify a document with arbitrary attributes. Final result for the user ¾ brief data about a document and its address (physical, server, URL etc). Self-restriction of ISIS developers to bibliographic kind of data and patronage by UNESCO have provided unprecedented advantages of system:
herefore, it may be desirable to keep these advantages in new versions. It seems for us important, that the following improvements should not go beyond outlined circle of problems and opportunities. Even essentially updated version should remain no more than metadata base, that are detailed descriptions of the documents which itself are placed on external sources (electronic, paper and so on). Attempt to overcome these restrictions would lead to significant complication of the system without obvious competitive advantages in comparison with the products, that applied now for storage of full texts, structured documents, digital and graphic data, etc. As it seems to us, the author of development ABCD adheres to this direction too.
3. Challenges from information community
While remaining in the limits specified above (metadata base for all kinds of documents) new versions of system should consider essential trends in technology of storage and distribution of the information. We shall specify only some of them:
A dominating role of electronic documents;
INTERNET as the basic environment of storage of the information;
Integration of different sources of information, requirements on interoperability and so forth;
Many kinds of standards and initiatives on formats for data storage and data exchange;
New concepts and technologies, for example, the concept of electronic libraries.
The analysis of these challenges, and also real needs of libraries and information centers, allow to choose directions of work on new versions of system ISIS. Certainly, here they are outlined in enough general view that assumes in the further detailed discussion and study of each of the chosen directions.
4. Possible directions of work
4.1. First two items specified in section 3, assume that the new system should be capable:
to integrate both electronic and paper sources of the information as equal, in both cases specifying their physical addresses, means for access and so forth;
to integrate all kinds of documents on a server, local network computers and documents with the remote access, located in elsewhere in the INTERNET;
(3) to provide access (opened or authorized) to a DB through the INTERNET. It is necessary to study also an opportunity of simultaneous work with the several DBs based on ISIS technology, but being presented in various formats and located on different servers.
4.2. Support of various international formats of metadata
Work in this direction as it is represented, should take a special place in development of system ISIS. The standardized metadata allow to solve many problems: identification and a finding of resources; definition of rules of access and conditions of use of the document; detailed consideration of content aspects, and so forth. Besides the traditional standards which have appeared in library community, there are existed some alternative formats:
Text Encoding Initiative, (http://www-tei.uic.edu/orgs/tei/) that is developed in the Center of electronic texts of Virginia in 1989 as a tool for process to prepare digital texts which identifies an electronic resource and its printed source by means of the metadata placed inside of the same electronic resource
Encoding Archival Description, (EAD, http://lcweb.loc.gov/ead/) is a set text SGML metadata that was developed for archive needs and used for standardization and classification of unique archival materials. It is used in archives and libraries of manuscripts for indexing corporate records and personal documents. Metadata EAD versions 1.0 are supported by the American archival society and Library of the Congress of the USA.
Dublin Core Metadata Initiative, (DCMI, http://purl.org/dc/) - is the simple and easy in use set of metadata based on format HTML. The Dublin kernel represents an absolute minimum of metadata necessary for search in a network.
Special formats are offered for various subject fields in view of their specificity: archival; geographical; distant education; medical; museum; scientific; news.
It is desirable to study an opportunity to integrate metadata and references that is may be of interest in science and publishing. There is interesting experience by Open Journals Project, http://journals.ecs.soton.ac.uk , The Open Citation Project, http://opcit.eprints.org/ , in linking of references. Similar work is going on in the project " Open Archives Initiative " (http://www.openarchives.org/). Introduction of metadata in links between related documents has a huge potential for search of information.
4.3. Structurization of the document
Apparently, representation of ISIS document in the form of a set of equivalent fields limits opportunities in some cases for adequate representation of the information. It seems reasonable to introduce at least two kinds of documents:
1. Documents with semistructured data;
2. Compound documents.
With the first one it is necessary to provide an opportunity of change in a set of fields (structure of metadata) for various records. It is essential, that scheme for semistructured data will be never considered as aprioristic. Variations and expansions of structure are possible at occurrence of new documents so the scheme of the description is a posteriori.
In the second case it is necessary to study an opportunity of the description of the compound documents represented in the form of the hierarchical scheme. For example, serial and multivolume editions, collections of the documents prepared in different formats can be considered as those of second kind. System of metadata should describe in this case both the head document, and documents of the bottom level, providing mutual references.
Each of the specified kinds shows necessity of some new features in development of the system. Certainly, in the further possible requirements to system and ways of a data structure should be detailed.
The separate question which is a subject for exploring is work with XML-documents. Though XML becomes the real standard of data presentation, it is necessary to estimate, how it is reasonable to connect its problems with ISIS.
5.Applications, limits and restrictions
New functionality and an opportunity of integration of the DB prepared in various formats, will wide circle of applications for ISIS. This means both ¾ fields of subjects and the typology of documents. Libraries of a various structure and scale are traditional users of ISIS. Special consideration is needed for opportunity of ISIS use to administration of electronic libraries. This term (electronic library or digital library) now applied to large repositories of the text and multimedia information with corresponding structures and means to access. Besides the mass media collections containing digitized books, films, and musical documents, digital libraries appear in scientific fields: data of bioinformatics, results of geographical or astronomical researches, etc. As a rule, these collections based on rather expensive software. A question arises: is it possible to organize a similar collection with ISIS? Certainly, its functions should be limited to document descriptions, to the descriptions of their structures and to opportunity to get link of access for each document at any structured level.
At the successful fulfillment of this problems practically any field of scientific activity, in social area (libraries, museums, monuments), in publishing and media fields, etc., may find use of ISIS. Of coarse, first of all these applications can be effective in management of descriptions of documents.
23rd June, 2008
Thermophysical Properties Center, Joint Institute for High Temperatures of the Russian Academy of Sciences, Moscow, Russia.
Dr. Vladimir Yu.Zitserman, PhD, Head of data bank laboratory
Dr. Michael S.Trachtengerts, PhD, Senior researcher
mailto:trachtengerts@mtu-net.ru
No hay comentarios.:
Publicar un comentario