25th International Conference on   Very Large Databases  Edinburgh - Scotland - UK  7th - 10th September 1999

History 

VLDB has been the primary international conference on database management for over 22 years, since it's inception at Frammingham, Massachusetts in 1975. An indication of the high quality is that research papers are considered on a par with journal papers for tenure decisions, worldwide. Hence, a significant part of the audience, generally 350-500 attendees, will be some of the leading database researchers in the world! 

Conference Structure 

The conference consists of a mixture of research papers, applications papers, vision papers, tutorials and panel sessions. Research papers present the results of academic research, applications papers report on solutions to problems found in applying database technology within commercial and industrial environments and vision papers attempt to provide a road map for future developments in database systems. Tutorials are half-day sessions where participants can learn about significant developments in database technology. Panel sessions permit an exchange of views between leading academic and industrial representatives working with database management systems. 

VLDB '99 

In September 1999, the twenty fifth VLDB will take place in the Edinburgh International Conference Centre (EICC). This is only the second time the conference has come to the United Kingdom and the first time it has come to Scotland. The venue brings the conference to an area around which many of Scotland's leading financial institutions are centred. This presents the Scottish finance sector with an opportunity to participate in an event that deals with a topic that is central to their I.T. strategies.  

Industrial Track, VLDB’99 

Over the past 5 years VLDB has made a strong effort to foster a close collaboration between the applied and theoretical database communities. We anticipate 60% researchers from academia and industry and 40% database product users and vendors. In order to continue this trend the organisers of the Edinburgh conference are seeking to put together a strong industrial track. The goal for the industrial track is to improve the correlation between the problems that are currently subjects of research activity and the industrial problems that actually occur in a commercial environment. However, the industrial and academic communities have very different modes of operation and motivations. Some of the hardest problems in computer science are the practical ones. Researchers often avoid practical problems since they don't have access to the real information and it takes a long time to get it and understand it. Industrialists often "solve" hard problems by brute force since they lack the understanding, resources, or time to pursue existing, or worse yet for them, not yet existing solutions that are more effective, elegant, efficient, etc. The research communities and industrial communities have a lot to gain from each other. But technical transfer between the communities seldom happens smoothly. Industry faces many technical challenges to which research could contribute if only communication was improved. A mechanism is being sought by the VLDB'99 organising committee to bridge this gap. 

VLDB’99 Industrial Strategy 

VLDB'99 will have an industrial advisory board chaired by Michael Brodie, GTE Laboratories working in conjunction with the programme committee which is responsible for the final selection of the conference programme. The job of this board is to solicit good contributions from industry or with an industrial interest. This includes application-oriented and product-oriented contributions. If needed, it will assist the programme committee in reviewing industrial papers.  

The industrial advisory group will include local industrial advisors. The companies with the most incentive to take part in VLDB'99 are probably those companies who are close to the venue for the conference. The local advisors will seek to maximise local industrial participation in the event. There may also be some benefit to the local VLDB'99 team to be in more direct contact with industry. 

The organising committee would like to hear industry views on current practices, challenges, and requirements; on the future of computing; the role that you see for database technology; and what real users want from database technology. For this dialog, we want to hear both the technology pull view (what users and applications want) and the technology push view (plans of the capabilities of next generation database technology).  

The following VLDB program events could be influenced by industrial concerns: 

• research
• applications / experience: lessons learned
• products / market / trends
• vision
• case studies, ...

Below are some initial ideas for topics or themes of interest to industry. 

Example Session Topics 

• Future Corporate Computing Architectures and the Role of DBMSs
• Legacy IS Migration: The Role of DBMSs
• Schema Integration, Evolution, and Migration: Semantic Interoperability
• Special Purpose DDBMS
• What Industry Needs From the DBMS Community
• DBMSs (technology) in future information / computing architectures
• Challenging future applications (e.g., multi-media, WWW, medical)
• Are Conventional DBMSs Dead? What are the requirements of future DBMSs

• "Academic Research Is Increasingly Irrelevant To Real Problems"
• Technology pull: "Industrial Use of DBMSs: Breaking the Myths of Researchers"
• Technology push: "Next Generation DBMS Technology and What it can do for Industry"

At the end of each day plenary industrial talks with real world lessons, challenges, etc. e.g., the current Microsoft/SAP/WWW Strategy could be held. 

Industry Day 

An Industry Day would be one day of VLDB'99 that has a significant number of sessions of interest to industry so that industrialists who can't spare 3-4 days at VLDB'99 could come for one day. This could also be sponsored by local industry. Industry Day could have at least one high-powered professional speaker, to aid in drawing the industrial community. It could also to feature a number of applications-oriented presentations. 

Industrial Theme 

Edinburgh has many financial and insurance industries. VLDB'99 could involve them by creating events (e.g., panels, tutorials, etc.) to suit their interests, etc. This might be done in conjunction with soliciting support or sponsorship (see sponsorship deals). 

Possible themes include: 

• The Role of DBMS Technology in Telecom, Finance, and Insurance
• Industrial Strength DBMS Requirements, Challenges, and Solutions
• Technology Transfer: What Industry Needs From Research

Dos: 

• tell us your unique experience that the research community (or others) are not likely to have and would benefit from)
• speak from your industrial perspective and experience to research folks. Go at least 1/2 way towards them.
• speak your mind, we hope this is a real opportunity
• tell us the misconceptions that you perceive that the research community has (e.g., about your application domain)
• state your essential message in two sentences at the beginning, middle, and end. Make it so compelling that one year later they will remember that sentence (and perhaps you).
• give lesson’s learned - what worked, what didn’t, what would you do now?
• what short cuts did you take ?
• what problems did you address ?
• open problems that you feel the research community might help with
• pose questions to us (without fear that you need to provide an answer)
• key application pulls (as opposed to technology push)

Don’ts 

• sell your product or your company – do sell the advanced technology concepts and the challenges they meet
• present your research achievement or research agenda unless you are giving a keynote with this purpose
• present your industrial research result (e.g., the paper you didn’t get accepted)
• use a talk / slides created for some other purpose or if so, please tailor them to the VLDB audience. If you want help on this, ask the person who invited you to give the presentation. 
• tell us what your system and your team did (e.g., history) except to serve a DO (above)
• talk in generalities or bit level detail - hit the problems square on