NoSQL and the Database Curriculum
Dr Jim Paterson, Glasgow Caledonian University
Teaching strategies for NoSQL and Big Data will continue to be a work in progress but the view of many educators is that these topics are now a core part of the database curriculum. The aim of traditional relational database courses is essentially mastery of the relational model and SQL. In the case of NoSQL, however, students must learn about the data models of a diverse range of datastores and be able to adapt their knowledge to deal with different systems. Teaching NoSQL offers opportunities to develop a different set of skills in critically evaluating the needs of an application scenario and making an appropriate selection of datastore from the wide range of available systems, including, of course, relational databases.
This will require a greater emphasis in teaching on requirements analysis, and the introduction of example scenarios which reflect the developments in the industry which have led to the NoSQL movement. In order to make appropriate decisions students need a strong appreciation of the nature and capabilities of different categories of datastore. Practical examples are important in developing this appreciation. However, NoSQL datastores typically have their own specific query mechanism, and it will often not be practical in a course to explore or practice these in any significant depth. It is also difficult in teaching examples to provide hands-on experience related to the issues of scale, and some thought and imagination needs to be given to this.
These issues were among those discussed at the Teaching, Learning and Assessment of Databases workshop in the UK in July 2014, and you can view the proceedings at here (link to .pdf)