Practical session of the postgraduate course on Information Visualisation. A considerable part of the course consisted of practical exercises in collaboration with other students. Source: Victor Pascual Cid. Click on the image o enlarge it

One of the reasons (albeit not the only one) for the long silence of this newsletter has been the process of the elaboration and starting up of the first postgraduate course, to our knowledge, on Information Visualisation taught in Spain. Hosted in the Instituto de Educación Continua (IdEC) of the Universitat Pompeu Fabra in Barcelona, Spain, it has been directed by Profs Josep Blat and Juan Carlos Dürsteler and coordinated by Dr. Victor Pascual Cid.

This course arises from the longstanding conviction that  for the advancement of Information Visualisation it's necessary to educate the increasing critical mass of professionals that are more and more in need of, and make use of, visualisation techniques to analyse and understand data.

There are, in different parts of the world, a certain amount of introductory and advanced courses on the topic, mainly addressed at professionals of computer science (see for example the Infovis-wiki.net site). We are not aware,however, of a course of transversal character, suitable for any type of professional, joining a solid theoretical and academic foundation together with a more practical part oriented towards critical judgment that promotes the making of eminently useful visualisations. (We'll thank the readers for any links in that sense).

The Syllabus

With the purpose of offering such a type of course we created a syllabus based on the process that is explicit in the diagram of Information Visualization due to Engelhardt and Dürsteler as a methodological basis to produce a practical approach to this discipline.

Consequently we established 3 main knowledge blocks following the above mentioned diagram:

InfoVis diagram Source: Image by the author. Click on the mage to enlarge it

• From data to information: that includes the definition of the problem and its associated creativity techniques, the conversion of data into information by means of metadata, the treatment, cleaning, generation of derived data and a long list of topics related to the transformation of raw data into high quality information.
   
• From information to visualisation: The core of the course where all the aspects and techniques that enable the designer to convert the already generated information into a graphic visualisation are treated. This includes visual variables, syntax and grammar of the graphic language, use of visual objects to create advanced graphics, composition rules, visual metaphors and a wide catalogue of "hands on" sessions oriented towards acquiring the methodology and criterion needed to build effective visualisations.
   
• From Visualisation to Understanding: If the user doesn't understand the visualisation, it is meaningless. Hence its indispensable to acquire the knowledge that helps to comprehend how understanding is built from the moment our senses perceive the visualisation until the cognitive models that intervene in the creation of knowledge are elaborated. Here we review concepts related to vision sciences, Gestalt, cognitive psychology, mental models, usability evaluation and interaction design, among others.
   

On the other hand, with the objective of complementing this core of information visualisation basic knowledge, we established a set of advanced seminars that included music visualisation, augmented reality for data visualisation, geovisualisation, advanced interaction design, practice with commercial tools like Quadrigram, elements of graphic programming in R and Processing languages, visual journalism and evaluation of graphics, among others, taught by professionals that work on a daily basis on this subjects.

Development of the course

The course ran from October 21st until June 1st, when the visualisation project was presented. The project, together with the exam that took place in March, once the theory and practical sessions ended, served as the evaluation of the students.

The very transversal students' profile, was made up of 5 journalists or communication professionals, 4 software engineers and/or multimedia programming experts, 1 marketing expert, 1 architect, 1 politics science professional, 1 master in business administration, that made up the 13 students in total.

• In the first phase, up until Christmas, the basic principles of the three blocks of the diagram were explained.  This task was carried out by the 3 main teachers of the course. A total of 68,5 hours were taught of which approximately 40% were practice sessions.
  
   
• In a second phase that covered after Christmas until March 17th, a total of 72 hours more were consumed. In this case the classes were devoted mainly to the advanced seminars mentioned above, that were taught by 14 invited professors, all of whom with a wide range of experience in their respective areas.
  
  It's important to note here that for the practice sessions the professors used tools that do not require any particular computer science knowledge along with the teaching of the basics of graphic programming using some of the languages that are widely used nowadays.
  The objective was that even the non programmers could see and understand the possibilities that programming offers, irrespective of whether the programming is made by themselves or ordered from third parties
  
   
• The third part of the course was devoted to the elaboration of a non trivial visualisation project by the students. For that purpose a data set with a raw file with 7 million entries corresponding to sales transactions worldwide during 2008, was kindly provided by a local company. 
  
  Three groups were formed, each one with a multidisciplinary profile set, in which there was at least one programmer, one expert in communication and one representative of the remaining professional profiles.
  
  Each group was asked to define a problem to solve on the dataset taking into account the needs of the company and create a complete visualisation project applying the methodology and knowledge acquired during the course.
  
  From March to June a total of 3 control sessions were done with each group.  In them the groups presented the advances and difficulties found so far and received suggestions and guidelines from the 3 main professors and contributions from all the other students.

Results

Three projects, that we will comment on in more depth in the next articles, were presented. It should be noted that the results of the projects pleasantly surprised the panel of professors evaluating them by the quality of the work and its potential impact, even though they were present in the control sessions and knew in advance the preliminary evolution of the work.

We have extracted the following conclusions:

• The methodology taught appeared to be a solid foundation that served the students as a strong guideline and enabled them to perform an exhaustive study in the different areas of the diagram. This translated into maturity and balance in the treatment of the data, the generation of visualisations and the attention to the cognitive aspects of the same, that is difficult to find in many of the visualisations you can find nowadays on Internet.
  
   
• The quality of the projects has been very high specially taking into account that many of the students didn't have previous knowledge of Information Visualisation and were not used to creating charts or graphic representations.
  
   
• We should reinforce the number of practical and "hands on" sessions where the students presented their solutions to proposed problems in front of the constructive criticism of professors and specially classmates, since it has demonstrated to be a powerful tool for the learning and understanding of the theory and the building of criterion.
   
• This type of course helps to create an awareness of the power of the application of Information Visualisation in transversal sectors and to generate a critical mass (today still incipient) of highly qualified professionals

These conclusions  will be incorporated in the next edition of the course, that can be consulted in the website of IdEC.

Links of this issue: