TD4DASS 2020

Workshop on Technical Debt in Data Analytics Software Systems


March 16-20, 2020 - Salvador, Brazil

Software engineering has been traditionally focused on systems built on general computing platforms, be they centralized or distributed. However, contemporary software systems have large numbers of distributed and purpose-built components deployed in specialized platforms, like the Internet of Things (IoT), Industry 4.0, Intelligent Cities, Cyber-Physical Systems, or Context-Sensitive Systems, among others. These kinds of software systems usually collect large amounts of data that will be later analyzed to extract information and produce knowledge. Even though data analytics solutions are of great interest in such software systems, most existing design solutions for data analytics are developed ad-hoc, which raises many technical challenges for software engineers. In particular, it is possible to observer that software architecture techniques are not yet exploited in data analytics software systems (DASS). Therefore, it is argued that DASS exhibits technical debt very often, in the sense that decisions are driven by functional or technology concerns with short-term objectives and not long life usage.


This workshop aims to bring together researchers and practitioners to discuss the typical problems and quality gaps that appear in the development of DASS, both from the perspective of software architecture and technical debt techniques. Also, the workshop seeks to foster the discussion of techniques, tools, and methodologies for identifying, measuring, and managing design and code quality in DASS. Some interesting questions that we attempt to address in the workshop include:

  • Which types of quality problems often arise in DASS?
  • How does technical debt items manifest in DASS? What are the challenges?
  • Is technical debt in DASS different from the technical debt appearing in "standard" software systems?
  • Do the existing techniques for technical debt apply to DASS?
  • In case there might be differences (in DASS), what are they, and which techniques are necessary to manage the debt?


The end goal is to put together a group of researchers and practitioners interested on the engineering of DASS and to develop a research agenda on topics related to the engineering of high-quality DASS.