Content Management for Standards
|Home -> HL7 -> CMS|
Traditionally standards were written using simple word processing software. Even well designed documents only use spell checking and style guides during the process of editing. Since word processors are designed as single user tools, no or only poor support for concurrent editing and version control exist. This approach may be sufficient for small and simple documents. Since standards of today consist of a large amount of pages and complex structure new techniques have to be developed.
By the use of databases techniques HL7 performed a first step to consistent documentation of the standard. For now this database exists as a supplement to the printed standard for version 2.x and provides easy to use navigation through the standards meta data as well as the textual document.
The basis of this database is a meta model of the standard. All kind of information has been assigned to appropriate entities. Afterwards the information can be transferred out of the documents into the database. This database is available by Health Level Seven, Inc.
The information can be used to
create HTML files for quick navigation
create XML files for computation
compare versions and identify changes
add site-specific information
store additional information like translations and comments
create documents for negotiating interfaces
support the development of applications
This kind of presentation database still leaves the process of editing to word processors with all their insufficiencies. The next improvement on the standard documents has to be in the field of editing. By use of content management techniques this process can be improved significantly.
The main improvements will be
consistent layout by separation of content and design
consistent use of meta data by introduction of references
version control and change logs
easy production of different formats for presentation
The first approach to create a database for content management is to break down the standard document into small pieces of text. The HL7 data like tables, segments, items, datatypes is extracted and replaced by references to the corresponding entities in the database. This insures a consistent use throughout the whole document. The remaining text is stored in the database as is. By linking text and references the structure of the original document is preserved.
The user can navigate through the database by the help of a browser and simple CGI scripts. The HL7 data can be edited by the use of CGI scripts and HTML forms. The database is responsible to keep track of the changes. These techniques can be used for editing the text as well but are still quite limited. Superior maintenance information links the different fragments to ballots, draft and final versions.
The next step will be to use standard toolkits for implementing a web enabled editor to perform more complex editing tasks. This will provide the HL7 community with the opportunity of concurrent editing and rapid documentation of the evolving standard.
If you are interested in the presentation please feel free to download the PowerPoint-Show.
Last Update : September 1, 2001