WWW94: ALIWEB - archie-like indexing in the web

presented by martijn koster, NEXOR Ltd, UK

while most approaches to create automated indexes attempt to search all documents on the whole web, ALIWEB wants the indexes to be distributed and reside on the server that provides the information. the idea is that each WebMaster should manually maintain an index file on her or his server and that one system on the web periodically retrieves all these index files and builds up a global database. in order to support an automated collection of the index files, these files must follow a certain standard. fortunately people at NEXOR did not invent a new standard, but they use the templates proposed by the Internet Anonymous FTP Archives (AIFA) working group which is similar to RFC822 mail headers.

once a server administrator has created an index file, the server has to be registered at ALIWEB using a form provided on the web. during the registration process, the existence of the index file is checked. the administrator can specify how often the index file shall be retrieved. the collection of the index file is known as "harvesting".

the global database can be searched through a search engine, either via a simple interface in the web or via other databases, e.g. the CUI WWW catalog.

this paper is no longer available on the web.

2nd_day_aliweb / 28-jan-1999 (ra) / reto ambühler