Broadening the Database Field for VLDB2000

As is traditional, the primary objective of VLDB2000 is to provide the highest quality conference program that presents the most recent advances and challenges in data management to both those attending the conference and those who do not. The second objective of VLDB 2000 is to contribute to the VLDB Foundation goal of broadening the field of database management. This section describes the broadening objective and how VLDB2000 will contribute to it.

The Challenge

Over its 40-year history, database research has made major contributions to developing database management systems technology, which now lies at the heart of every conventional application. In principle, the database field could contribute to the full scope of data management, its applicability, its challenges, and its future directions. In practice, this has not happened. The database field now faces two challenges. First, fundamental data management assumptions do not apply to the data requirements of the next generation of applications. Long-standing data application challenges, such as semantic interoperability, inhibit data-centric solutions, thus leading to solutions in other domains. New computing environments, such as the WWW, require a rethinking of core database technology in all areas, e.g., architecture and query/search techniques. Second, new application domains typically do not appeal to the database field to address new data management problems. Over the past decade, data-intensive problems (e.g., the role of data in the Web, e-commerce, digital libraries, and knowledge management/discovery) have been addressed by other areas or newly created areas. As the Internet, gizmos, e-applications, ubiquitous computing, and other trends revolutionize computing, which is accompanied by an explosive growth of data and transaction volumes, database technology is relegated to its conventional forms. The database field is being narrowed to core database technologies used by conventional applications. 

As we embrace the largest revolution in computing history, the database community should face the full scope of data management challenges in the next generation of computing, that is, all forms of data in all applications. Data, which is one of the three pillars of modern computing, is critical to current and future computing. Data volumes and database transactions are growing to astronomical levels. Yet data and its management is not a first-class citizen with process/logic, communications, or presentation. New areas with critical data components do not turn to the database community for ideas, advice, or technology. In turn, the database community does little to reach out to these new domains. The database community has much more to offer the current and next generation of computing than is currently offered or requested. We want to change this. We want the database community and its technology to take its proper role in the next generation of computing.

The VLDB2000 Response: Encouraging Broadening

In response to the above situation, the VLDB Foundation and VLDB2000 encourage contributions to both (1) core database technology; and (2) topics that will broaden database technology to address all relevant data management challenges. We rest assured that many strong core database technology contributions will be received. We intend to accept as many as in past VLDB conferences. We strongly encourage broadening papers, those on non-core database topics. Assuming quality criteria are met, we intend to accept more broadening papers than in the past. VLDB2000 is introducing a fourth parallel session for this purpose. VLDB2000 will use quotas to ensure the acceptance of contributions that substantially contribute to this broadening as long as VLDB quality criteria are met.  

The change is not dramatic. All past VLDB conferences have accepted both types of paper. However, core database technology papers have always dominated in both volume and quality. Following this VLDB tradition, VLDB2000 would consist largely of contributions to core database technology, including many high-quality "delta" papers. VLDB2000 will make a small but concrete step toward a greater number of contributions that will broaden the database field. 

VLDB2000 encourages broadening by soliciting papers that are (1) on a broader range of topics than those considered by previous database conferences; (2) on riskier and more novel challenges, as opposed to incremental improvements on existing results; (3) from a broader range of contributors (e.g., from across the spectrum of developing and deploying database technology and from those outside the field who pose new requirements and challenges); and (4) in novel formats such as reports on case studies, systems development and testing, and product evaluations relative to new application requirements.

Broadening Defined

Intuitively we in database research all know if a paper addresses conventional database topics or if it reaches beyond to novel issues and challenges. The following presents the criteria that will be used to make this distinction. When VLDB 2000 program committee chairs and members must evaluate "if the paper is on a conventional database topic or if it goes beyond conventional databases," the following considerations will apply.  

A paper will be considered as contributing to broadening the database field if it addresses issues beyond conventional database topics and technology. It must contribute to expanding database technology or methods beyond conventional databases and applications to the full scope of data management, its applicability, its challenges, and its future directions, as discussed above. 

Papers can contribute to broadening to varying degrees. A paper on database technology in government would contribute to broadening but not as much as one on WWW-based database technology in government. 

Some topics are more likely to contribute to broadening than others. For simplicity, we distinguish (1) core database technology topics from (2) broadening topics, as is done in the VLDB 2000 Topics of Interest (see WWW page). However, a paper on any topic could contribute to broadening or to core database management. Typically, a paper on core database technology would not contribute to broadening. However, a paper on core database technology to support WWW-based approximate or evidence accumulation queries would contribute to broadening. Typically papers on non-core database topics such as advanced languages, methods, or applications will contribute to broadening. However, papers on conventional database applications, such as databases in government or telecommunications, may not. 

Generally, research methodologies for conventional database topics are well established. This leads to high-quality standards for the papers on those topics. Correspondingly, research methods for broadening papers, those that go beyond conventional topics, are much less settled. This leads to novelty and innovation and a lack of adequate quality measures. 

Some contribution categories are more likely to contribute to broadening than others. Research papers have tended to address core database technology. However, we are hoping to have more research papers on broadening topics. Vision papers, by definition, must contribute to broadening. Finally, Experience and Application papers are likely to contribute to broadening unless they deal with conventional applications or systems.  

VLDB2000 will use quotas to ensure the acceptance of contributions that substantially contribute to this broadening as long as VLDB quality criteria are met. Although core database technology will constitute the majority of papers and topics in VLDB2000, there will be a quota for the number of sessions devoted to core database technology research papers. In the past, papers submitted on advanced application areas were rejected since they did not match the quality established by papers on core database technology. The quotas are intended to resolve that problem by allocating sessions for topics that will contribute to our broadening objectives.