Open Documentation Cataloging Project










Mailing List

SourceForge Project Page





What is wrong with the current documentation system?

Right now documentation on a typical Linux system is spread out in many different directories in the filesystem. It is generally in many different formats, languages, and styles. This makes it very difficult for a user to find the documentation they are looking for and extremely difficult for a help browser to interface all of the documents. Generally, the help browser cannot determine the title, language, or subject of a document.

What should a good help browser do?

A good help browser should be able to:

  • Show the user all the documentation on the system
  • List the document titles
  • Display documentation only for a specific language
  • Sort the documents into a subject-sorted Contents List

Ideally, it could also:

  • Search for documents on the Net
  • Determine the document version
  • Display a short description of the document

What does a help browser need?

The help browser will need to know:

  • Which documents are on the system and where they are at
  • The title of each document
  • The language of each document
  • The format of each document
  • The subject of each document
  • Where to place the document in the Contents List
  • (and generally a few other things)

With this information, a help browser will be able to present the user with a complete list of the documents on the system in a particular language, sorted into a contents list. It will also allow the help browser to provide more advanced features such as searching by document subject, description, or type. Also, since the actual document location is specified by a URI, the document itself does not need to reside on the local computer but could be anywhere on the Net.

What is the OMF?

The OMF is a simple 16-element metadata specification which allows document authors to describe various properties of a document, such as those listed above. This metadata will provide the crucial information the help browser needs to provide both simple and advanced features to the user.

Where does ScrollKeeper come in?

ScrollKeeper is a set of installation/uninstallation scripts and a library which help browsers can use to sort and search the metadata. By extracting these tasks out of the individual help browsers, we avoid duplication of work and guarantee compatibility between various help browsers.

SourceForge Logo