Andrei Varanovich

Modeling the linguistic architecture of software products

Understanding modern software products is challenging along several dimensions. In the past, much attention has been focused on the logical and physical architecture of the products in terms of the relevant components, features, files, and tools. In contrast, in this paper, we focus on the linguistic architecture of software products in terms of the involved software languages and related technologies, and technological spaces with linguistic relationships such as membership, subset, or conformance. We develop a designated form of megamodeling with corresponding language and tool support. An important capability of the megamodeling approach is that entities and relationships of the megamodel are linked to illustrative software artifacts. This is particularly important during the understanding process for validation purposes. We demonstrate such megamodeling for a technology for Object/XML mapping. This work contributes to the 101companies community project.

101companies: a community project on software technologies and software languages

101companies is a community project in computer science (or software science) with the objective of developing a free, structured, wiki-accessible knowledge resource including an open-source repository for different stakeholders with interests in software technologies, software languages, and technological spaces; notably: teachers and learners in software engineering or software languages as well as software developers, software technologists, and ontologists. The present paper introduces the 101companies Project. In fact, the present paper is effectively a call for contributions to the project and a call for applications of the project in research and education.

A Framework Profile of .NET

We develop a basic form of framework comprehension which is based on simple, reuse-related metrics for the as-implemented design and usage of frameworks. To this end, we provide a framework profile which incorporates potential reuse characteristics (e.g., specializability of types in a framework) as well as actual reuse characteristics (e.g., evidence of specialization of framework types in projects). We apply framework comprehension in an empirical study of the Microsoft. NET Framework. The approach is helpful in several contexts of software reverse and re-engineering.

Axioms of Linguistic Architecture

In documenting software technologies (e.g., web application or modeling or object/relational mapping frameworks) and specifically when discussing technology usage scenarios, one aims at identifying and classifying the involved entities (e.g., languages and artifacts); one also aims at relating the entities (e.g., through conformance or I/O behavior of program execution). In this paper, we present a logic-based axiomatization (an emerging ontology) for the underlying types of entities and relationships, thereby formalizing recurring documentation idioms such as ‘a software system (e.g., a Java application) to use a technology (e.g., a test library)’ or ‘a technology (e.g., a web application framework) to facilitate a certain concept (e.g., the MVC pattern)’. The axiomatization is illustrated by examples applying to the Eclipse Modeling Framework. The inclusion of types of entities and relationships is driven and thus validated by a literature survey on megamodeling.

The 101haskell Chrestomathy: A Whole Bunch of Learnable Lambdas

The paper describes the 101haskell chrestomathy---a collection of Haskell programs implementing features of a hypothetical information system in a manner to represent knowledge about functional programming useful for learning (and teaching). The programs are enriched with documentation, metadata, and links to other knowledge resources such as Wikipedia and Haskell textbooks. The underlying ontology is informed by a process of knowledge integration which derives a consolidated vocabulary mainly by text mining and summarization from textbooks. The usefulness of 101haskell for teaching is demonstrated with a functional programming course that is directly based on 101haskell.

Method and Tool Support for Classifying Software Languages with Wikipedia

Wikipedia provides useful input for efforts on mining taxonomies or ontologies in specific domains. In particular, Wikipedia’s categories serve classification. In this paper, we describe a method and a corresponding tool, WikiTax, for exploring Wikipedia’s category graph with the objective of supporting the development of a classification of software languages. The category graph is extracted level by level. The extracted graph is visualized in a tree-like manner. Category attributes (i.e., metrics) such as depth are visualized. Irrelevant edges and nodes may be excluded. These exclusions are documented while using a manageable and well-defined set of ‘exclusion types’ as comments.