Relational Databases

This week's reading exposed me to something that I use every day, yet know little about how it works. In a normal day, college students are exposed to databases which handle everything from the purchase of coffee in the morning to the scheduling of courses throughout the day. In the AccessScience assigned reading, the author discusses some of the hardware and software workings of databases, as well as some of the information architecture logistics. Finally, the author discusses the relational model of database mangement that E.F.Codd proposed in the early 1970s. As I began to read other websites about databases, I found the wikipedia page for the software giant, Oracle Corporation. In the histotry section of this article, it discusses how Oracle's founder, Larry Ellison was inspired from this same paper about relational database management systems, such that he created the multi-billion dollar software enterprise. It seems like few people have heard of E.F. Codd, yet many have heard of Larry Ellison, as he was able to use database technology to propel himself to the 5th wealthiest person in the world (Source: Forbes).


Nonetheless, I chose to read up a little more on relational model theory. This theory describes a database as a collection of predicates (true/false or boolean functions) over a finite set of predicate variables, describing constraints on the possible values and combinations of values. At the heart and core of building information modeling exists this precise definition; a BIM is simply a database by which building elements are represented by the collection of predicates, enhanced by a unique set of constraints. Embeded in the database are entities which allow for the user to define a building and its components by its unique features. Each entitity carries information, as well as constraints. An example of the relational model is shown below, where each pice of data is linked together by a common "key:"

The second article I read this week was called BIM Interoperability and Relational Databases. This article presents a way in which contract specifications ("specs") can be produced more efficiently and accurately using relational databases. Traditionally, specs are drafted in a word processing software program, producing the text in a flat word processing file. Sections are not linked together, and functions such as data formatting can be cumbersome to implement throughout the document. However, using relational databases can link sections together, which would increase productivity, coordination, and reduce errors.