lunduniversity.lu.se

Computer Science

Faculty of Engineering, LTH

Denna sida på svenska This page in English

Minding the trouble gap

2013-09-02

Elizabeth Bjarnason

Elizabeth Bjarnason knows the importance of communication. Collaboration and good communication can be a strong point and a safety net when running a software project. On the other hand, bad communication can cause big problems. Therefore, Elizabeth Bjarnason has developed two methods to support project teams in identifying troublesome gaps and gaining insight into good practices.

 

Her technical research topic is the role and impact of requirements engineering (RE) within software development. In particular, she focus on how a closer integration of RE can support a more efficient and effective development of software products. 

– Writing up my thesis I see ‘collaboration’ as a read thread that runs through both the initially investigative studies about challenges and practices for the alignment and coordination of RE with testing, and the two methods I have designed for supporting project teams in improving their development practices. 

Distances, or gaps, between people and between artefacts can cause problems with coordinating and aligning the various activities within a software project. For example, gaps in the communication between the business and development side of a company can result in that the development side working towards different goal than what the business side intended. Similarly, if there are semantic differences, gaps, between the artefacts used for specifying the requirements and the test cases, a different set of requirements are tested than what was specified. Both of these gaps, if left unaddressed, will result in the project developing software that doesn’t match the product intentions with delays and increased cost required to put it right.

Therefore, Elizabeth Bjarnason and her team developed two improvement methods; Evidence-Based Timeline Retrospective (EBTR) and Gap Finder. The methods have been found to support project teams in in identifying troublesome gaps and gaining insight into good practices that can alleviate these. 

– One of the core strengths of the EBTR method is that it enables project teams to learn from past events. That is done by providing visualised time lines of project history around which the project members can reflect and discuss on weak and strong points of the project. The Gap Finder method gives project members new insights by providing a view of various project gaps as factors that can explain various issues. Physical distance can increase the risk of misunderstandings and delays, for example when it is far between the developers and the person who has insight into customer requirements, says Elizabeth Bjarnason.

She has a range of experience of research collaboration with industry: multiple interview studies, general group meetings, project-specific group meetings, questionnaires, active collaboration on research design. 

– Ideally there is a regular interaction and exchange of research ideas and company challenges, from which studies can be designed that match interests on both sides. However, in my experience this is the exception to the rule. I have had this close partnership on one study, the design of EBTR method. For all the others the drive and design was all from the research side, she says. 

According to Elizabeth Bjarnason, for successful research collaboration with industry it is needed to secure contact people with:

 

  1. a genuine interest - not just assigned as part of the job 
  2. time 
  3. sufficient authority within the organisation to gain access to the resources needed for a study, e.g. interview participants, security clearance for data access etc. 

 

– Then, of course, this needs to be anchored further up the organisation, but from an individual researchers perspective this personal speaking partner is the key. In general, most practitioners are helpful and want to contribute but are very busy with their regular work. It might be possible to gain access to individual interviewees, but to perform a group meeting with a whole project team is extremely hard to realise; both to gain an okey and to practically schedule it in their calendars. It can be done but requires a lot of time and effort, says Elizabeth Bjarnason.

It is important to give something of value back to your industry partners. However, differences in perspectives and priorities between industry and academia in general can make it hard for a researcher to identify the winning points for industry. There are wide communication gaps caused by this difference in priorities. According to Elizabeth Bjarnason, it is up to the researcher to bridge these. 

What are your advices for other researchers who want to work with industry partners?

– Research and industry collaboration does not happen by itself, not anywhere. Collaboration with an industry partner requires the researcher to quickly pick up company terminology and insight into their world. Having previous industrial experience helps a lot with this, says Elizabeth Bjarnason.

 



EASE


Back to list.