Considering Academia-Industry Projects Meta-characteristics in Runtime Verification Design
Runtime verification, with its practical applicability and myriad of theoretical challenges it still poses, has the potential to bridge the gap between academic research in the field of formal methods with the software industry. In order to facilitate thi
- PDF / 235,369 Bytes
- 10 Pages / 439.37 x 666.142 pts Page_size
- 23 Downloads / 207 Views
Abstract. Runtime verification, with its practical applicability and myriad of theoretical challenges it still poses, has the potential to bridge the gap between academic research in the field of formal methods with the software industry. In order to facilitate this, it is useful to extrapolate success patterns from previous projects: Are certain characteristics of an industry-academia project a determining factor in the project’s success? How can runtime verification design decisions take into considerations project characteristics to improve the chances of success? This paper attempts to shed some light on these questions by reflecting on five projects with two partners over the past ten years. A number of lessons emerge, perhaps the most poignant one being the need to think long term in setting mutually beneficial goals from which a strong working relationship can emerge.
1
Introduction
Although various underlying notions from runtime monitoring and verification, albeit in limited form, have long found themselves in standard quality assurance practice in industry, its adoption as a first class element and building block of the system being built is still rare. Much of the use of runtime verification techniques in industry, thus still stems from projects in conjunction with academic partners interested in exploring scalability and industrial-relevance of these techniques. In the literature reporting these projects, the focus is invariably the effectiveness of the techniques used on the system under scrutiny. What is usually not reported (being beyond the scientific scope of such publications), is the process of adoption, the context of the project, the logistic challenges encountered and its longer term impact in terms of adoption of techniques beyond the scope of the original project e.g. Was it an industry- or academia-led project? Was runtime verification being engineered retrospectively on a legacy system or one being developed from scratch? It is worth noting that some, although not all, of the observations we make are relevant to any project with partners from both industry and academia, and not limited to runtime verification. c Springer Nature Switzerland AG 2018 T. Margaria and B. Steffen (Eds.): ISoLA 2018, LNCS 11247, pp. 32–41, 2018. https://doi.org/10.1007/978-3-030-03427-6_5
Considering Academia-Industry Projects Meta-characteristics
33
In this paper, we present anecdote-based observations based on our experience from five academia-industry projects1 , discussing and reviewing major design decisions in runtime verification engineering. An implicit assumption throughout the paper will be that the goal of the academia-industry collaboration is the benefit to both parties. Hence, in the next section we attempt to describe success from the two points of view. In Sect. 3 we describe project metacharacteristics which might affect the engineering decisions discussed in Sect. 4. We bring everything together in Sect. 5 by reviewing our decisions in five projects and report on their success. The last section concludes
Data Loading...
