The Lore project was declared a success in the year 2000 and is now pretty much out of business. These pages represent a snapshot of the project at some time in the past. We hold no responsibility for the currency (or lack thereof) of their content.

Introduction

Lore is a database management system (DBMS) for XML, a simple and increasingly popular data model. Nested, tagged data is the essence of XML.

XML is well-suited for loosely structured (or semistructured) data, since the standard puts no restrictions on the tags or nesting relationships. No "schema" is required, since XML data is self-describing: structure and data are intertwined in one format. For rapidly evolving data, such as data on the Web, XML gives users the freedom to change their data without constantly updating an associated schema. In other situations, for data whose structure changes less often, XML optionally supports Document Type Definitions (DTDs) for restricting the tags and nesting rules. In either case, XML is ideal for exposing and exchanging a simple and convenient view of data.

As more and more data is exchanged and ultimately stored in XML, users and applications alike will benefit from the ability to query XML data. End-users in an interactive setting will expect to perform keyword searches or specify queries through a convenient GUI. Further, a declarative XML query language (and query processor) will be as important to XML applications as SQL and relational databases have been to traditional business applications. A declarative query language simplifies application design, modification, and extensibility by pushing data processing to a tuned query engine rather than relying on custom code to handle the data.

Lore Overview

The Lore project focuses on defining a declarative query language for XML, developing new technology for interactive searches over XML data, and building an efficient XML query processor.

Lore has been under development at Stanford University since 1995. Lore was originally developed for a data model we named OEM; OEM is essentially equivalent to XML, but OEM does not support the notion of DTDs. Hence, much of our original work was geared toward the assumption that data is semistructured and need not conform to a particular schema. Recently we have migrated Lore to fully support XML; see From Semistructured Data to XML: Migrating the Lore Data Model and Language. We have a fully-functional prototype DBMS, complete with query language, multiple indexing techniques, a cost-based query optimizer, multi-user support, logging, and recovery. Lore also includes novel technologies such as DataGuides, management of external data, and proximity search. Online Lore demos are available to the public. The Lore executables are also available.

Site Index

Here is a summary of the information at this site:
  • Research: A more detailed discussion of the research issues we are exploring
  • Demos: Online Lore demonstrations
  • People: A list of past and present contributors
  • Publications: Citations for all Lore publications, including links to PostScript and Pdf copies
  • Get Lore: Instructions for downloading Lore
  • Related Work: A short (and outdated) discussion of related projects

Funding

The Lore project has received research funding in the past from DARPA and the Air Force Rome Laboratories. Currently Lore is funded by NASA and by the National Science Foundation under grant IIS-9811947.