Sound and Practical Unanticipated Reuse of Software (SPURS):
BIRS Invitational Workshop and
Special Issue of the Journal of Software: Evolution and Process


Position papers/short proposals due: 17 November 2016
Workshop: 7--9 April 2017 in Banff, Canada (near Calgary)
Full manuscripts due: 17 June 2017

Classical software reuse demands that we predict well the future needs of software systems so that some of their functionality can be modularized in an easily reused form; this is anticipated reuse, including such standard industrial technologies as object-oriented inheritance, software components, and software product lines. Since our ability to predict the future is less than perfect, there occur cases where a developer cannot find a modularized, reusable artifact that meets their needs. To deal with these cases, other options that support unanticipated reuse are also needed.

The most obvious approach supporting unanticipated reuse is copy-and-modify: for example, when a software developer is faced with a development problem, they will often search the web for coding solutions that appear to solve their problem, and then they will modify this code to fit their particular needs. Due to the high costs associated with other alternatives, copy-and-modify is often seen as a viable, pragmatic choice in industry.

Copy-and-modify has benefits: customizability is not limited to an interface provided by a 3rd party and there is no need to depend on unwanted functionality. But copy-and-modify also has drawbacks: the reuser can make poor decisions about what to reuse; the lack of tool support for these tasks makes them error-prone; and there is little assurance that constraints on the originating system remain in force in the target system. On a large scale, the result can be high error rates, poor comprehensibility, and poor performance. Improvements are needed to make this industrially relevant process less risky.

Special Issue of Journal of Software: Evolution and Process

This special issue solicits original contributions presenting approaches that are both practical (meaning, potentially scalable to industrial-sized systems; potentially usable by industrial developers in a realistic setting) and soundness-aware (meaning, either formally sound or that limit the developer’s attention to portions of the work that are not provably sound, requiring manual intervention) for any or all of the classic phases in the reuse process: artifact selection; adaptation to and integration with the target environment; and verification that the final system conforms to pertinent constraints on the original system.

To provide a common basis for comparison of techniques, two motivational examples and four case studies are provided in which some feature is to be extracted from an industrial system and reused within a target system. Authors should demonstrate how their work is practical and soundness-aware with respect to at least one of these case studies.

Position Papers/Short Proposals

Authors are encouraged to first submit a position paper as a short proposal for their manuscript (maximum 5 pages in ACM format). Each position paper should outline the idea of the approach, how it addresses the goals of the special issue, what its novelty will be, how the work will be connected to the case studies, and the potential for integrating the work with other work (e.g., covering other phases). Authors of the best proposals will be invited to a limited-attendance workshop, to discuss the ideas, to obtain feedback, and to seek commonalities.

Invitational Workshop

The invitational workshop will be held at the Banff International Research Station for Mathematical Innovation and Discovery (www.birs.ca) on 7–9 April 2017, in Banff National Park, Canada (near Calgary). Attendance is limited to 21 people. There is no direct cost for attendance. Participants are responsible for their own costs of travel and meals. Further details are available on the organizational website.

The goal of the workshop will be to improve the manuscripts to be submitted, to tie them together more closely, and to foster a collaborative community involving both empiricists and theoreticians.

Full Manuscripts

Full manuscripts will be submitted shortly after the workshop. Manuscripts whose authors have not been involved in the proposal and workshop process can expect to be at a disadvantage, but will be treated equitably.

Timeline

17 Nov 2016: Submission deadline for position papers/short proposals
17 Dec 2016: Feedback on proposals; invitations to workshop sent out
17 Jan 2017: Workshop initial schedule posted
31 Jan 2017: Workshop attendance confirmations required by this date
17 Feb 2017: Workshop final schedule posted
7–9 Apr 2017: The workshop will take place in Banff, Alberta, Canada (near Calgary). An initial session will likely be held on the Friday evening (7 April) and informal discussions will likely be held on the Sunday morning (9 April) to allow for people with tight travel arrangements. [Dates are firm.]
17 June 2017: Submission deadline for full manuscripts to be considered
17 Sept 2017: Notification of first round results (Major Revisions or better, else rejection)
17 Oct 2017: Submission deadline for revised manuscripts
31 Oct 2017: Notification of second round results (Minor Revisions or better, else rejection)
07 Nov 2017: Submission deadline for re-revised manuscripts
17 Nov 2017: Manuscripts ready for copy editing

Organizers

Prof. Robert J. Walker, University of Calgary (Canada), walker@lsmr.org
Prof. Agostino Cortesi, Ca’ Foscari University of Venice (Italy), cortesi@unive.it