The World Wide Web (WWW) [BLCL+94] increases in size and usage all over the world. In fact, WWW is one of the fastest growing services on the internet [Pit94]. We make our papers, articles, and other pieces of information available on the WWW marking up their contents (text, graphics, sound, movies) in HTML [BLC93]. We fill our WWW servers with huge amounts of information - but what about maintenance?
When making a longer article available in HTML we would like to give readers
a choice between: reading the whole paper, a section, a subsection,
or only the table of contents,
i.e. giving them a choice of different granularity.
Using the current HTTP [BL93] servers
we have to create separate files for each different choice!
To have an article in 
and make it available on WWW,
we need to convert it using a tool like 2HTML, which
creates yet another set of files from the original source.
A possible solution to the maintenance problem is to use software
like MOMspider [Fie94] which can be used to verify
the consistency of a local web (link verifications).
However it only operates on the information structure,
not on the information content.
We have a partial solution to the problems of maintenance, sharing and re-use of information, and viewing of information at different levels of granularity. Our first prototypical WWW front end to LINCKS [LSP93], enables readers to browse the contents of a LINCKS database using any HTML browser. The underlying features of LINCKS are able to solve a few of the mentioned problems.
In this paper we give an overview of our prototype front end, starting with a brief description of LINCKS' architecture as well as the data model. Then we give a more thorough overview of the ``document'' structure used in LINCKS before we describe the WWW front end's architecture. Finally, we discuss experiences learned from the realization of the prototype and possible future extensions.
For a demonstration of the prototype, Demonstration Front End.