ICSE 2010 - SUITE 2010: 2nd International Workshop on Search-Driven Development - Users, Infrastructure, Tools & Evaluation

Goals and Motivation

As software development is a process of both information creation and information gathering, software developers are constantly searching for the right information and the right person to solve their problems at hand. This workshop will focus specifically on exploring the notion of search as a fundamental activity during software development. The general theme of our workshop is to adress topics related to ”understanding and fulfilling various information needs of software developers”. The goal of the workshop is to bring researchers and practitioners with special interest on these topics together. Participants will have broad range of expertise in topics ranging from building software tools and infrastructure, information retrieval, user studies and HCI, benchmarking and evaluation.

Topics

The workshop is concerned with all aspects of the search experience of software developers.

Search-driven development involves a broad range of activities and topics: from finding reusable software assets to finding experts within an organization; from using the mighty grep to using internet-scale search engines dedicated to software artifacts; from finding matching code fragments based on a simple text query to based on a given unit-test. At the first edition of this workshop we identified suitability and trustability of search results as key factors in software search. For example, that software search results must not only be relevant to the search query but also suite into the current source code under development; and that source code alone is often not sufficient to assess the trustability of a search result. Similar factors apply when searching for external artifacts, whole projects or developers expertise.

Topics of interest include, but are not limited to:

Approaches and tools for searching software artifacts.
Approaches and tools to search for developer expertise.
Behavioral and executables specifications (eg unit tests) as search queries for automated search and reuse.
Crawling and parsing of internet-scale code repositories.
Empirical studies of search and navigation in software development.
Experience reports on setting up and running large software search engines.
Information retrieval techniques and machine learning approaches to search software.
Integration of search engines with IDEs.
Just in time comprehension tools for developers.
Leveraging Web 2.0 and social networking techniques for developer’s information needs.
Methods of integrating indexed data from various sources and histories.
Natural language processing to support software development.
Novel argumentation models for searching in software message boards and email archives.
Pragmatic reuse of components, i.e. through code transformation rather than via frameworks.
Programming by example.
Program analysis techniques in code search. For eg; automatic extraction of API usage patterns for program comprehension.
Query languages for software search.
Ranking strategies and heuristics for code search.
Searching aids for debugging and profiling.
Searching the computation space, i.e. runtime object graph / memory.
Slicing and generative techniques for search results extraction and synthesis.
Standards and benchmarks for the evaluation of search tools.
Suitability and Trustability of search results.
The use of visualizations to support software search.
Understanding information needs of software developers.
We are looking forward to exciting and stimulating discussions about all these ideas ? and much more ? at the workshop!

Intended Audience

The workshop is intended for:

Researchers with interest in enhancing search technology for developers.
Tool builders that want to improve the search experience of developers.
HCI Researchers with interest in better understanding of the search needs of - Search-engine providers from the industry targeting the needs of developers and software industry
Each participant is requested to submit a position paper in advance and each participant is requested to read all the submitted material, so that the workshop itself can be devoted to discussion instead of presentations. Submissions will be made electronically to facilitate the rapid exchange of information.

The upper limit for the number of participants is 30 and the participants will be selected on the basis of the submitted contribution. We actively seek a format which emphasizes fruitful interactions and discussions. This involves brief (5 minute) presentations of position papers; break-up sessions in discussion groups, and plenary meetings to discuss results.

Submission Guidelines

Submissions can span one or more of the following general categories:

Human Factors
Industry/Experience Report
Software Analysis
Software Information Retrieval
Tool Building
Be electronic. Submit your position paper (no more than 4 pages) in PDF, using the workshop’s Cyberchair website, so that we can collect and publish all of the submissions on the website. Accepted submissions will be included in the ACM digital library. Submissions should be made in the ACM Proceedings format. Please follow the style guidelines as prescribed by ICSE, also detailed in the ACM web page.

Be short. Please keep all position papers under 4 pages.

Be innovative. It is okay to propose a recent idea that still has some unfinished sides to it. It is supposed to be a workshop, not a mini-conference.

Be a rebel. Neglect these guidelines if you feel that your idea needs a special treatment in some way.