This is a paper on applying database techniques to support VLSI design activities. The environment for such activities is completely different from the environment in which, for instance, transactions on bank accounts take place. Those differences and their effect on system requirements are discussed in the paper.

Engineered objects--like VLSI design--are built by teams, simultaneously working on different parts of an overall design. It is therefore proposed that the design database be viewed as a library: design subparts are checked out to designers, who return them when done. In the paper, attention is also paid to aspects like concurrency control and resiliency to system crashes. Although a bit superficial, the paper is well written and contains some noteworthy remarks.