H. L. Ossher. A mechanism for specifying the structure of large, layered systems. In B. D. Shriver and P. Wegner, editors, Research Directions in Object-Oriented Programming, pages 219--252. MIT Press, 1987.  Home/Search   Document Not in Database   Summary   Related Articles   Check

....subsystems, and interfaces; dependencies among components such as supplier client, composition, and control flow relations; and attributes such as component type, interface size, and interconnection strength. The structure of the system is the organization and interaction of these artifacts [1]. This paper comments on the deficiencies in traditional documentation techniques and describes an approach to supporting software evolution through documenting software structure via reverse engineering. In particular, it focuses on how the approach facilitates understanding the structure of ....

H. L. Ossher. A mechanism for specifying the structure of large, layered systems. In B. D. Shriver and P. Wegner, editors, Research Directions in Object-Oriented Programming, pages 219--252. MIT Press, 1987.

....the so called connectors [5] In our case, on the other hand, the law can impose global properties i.e. regularities on a system. The enforcement of constraints over the structure and behavior of systems is not entirely new, however. An early, but isolated, attempt to do so is that of Ossher [7], who built an environment which allows for the specification and enforcement of a very special kind of structural laws , that regulate module interconnection in layered systems. One should also mention in this context the very interesting work on computational reflection, such as the metaobject ....

Ossher H. L. A mechanism for specifying the structure of large, layered systems. In Bruce Shriver and Peter Wegner, editors, Research Directions in Object-Oriented Programming, pages 219--252. MIT Press, 1987.

....domains using program transformation are discussed in Chapter 6. 2.3 Techniques for Representing Structure By understanding the ways structure can be expressed with modules, the usefulness of a transformation can be shown by demonstrating how it changes modules to modify structure. Harold Ossher [Ossher 87] described the basic structural properties essential to organizing information in a concise fashion, allowing substantial reuse and localization of changes. Grouping identifies a set of program components as being part of an aggregate component, perhaps with a name. Denoting a group is to denote ....

H. L. Ossher. A mechanism for specifying the structure of large, layered programs. In B. Shriver and P. Wegner, editors, Research Directions in ObjectOriented Programming, pages 219--252. MIT Press, Boston, 1987.

....such 2 Rigi is named after a mountain in central Switzerland. as supplier client, composition, and controlflow relations; and attributes such as component type, interface size, and interconnection strength. The structure of a system is the organization and interaction of these artifacts [23]. For a large software system, the reconstruction of the structural aspects of its architecture is beneficial. This process may be termed structural redocumentation. It involves the identification of the software artifacts in the subject system and the organization of these artifacts into more ....

H. L. Ossher. A mechanism for specifying the structure of large, layered systems. In B. D. Shriver and P. Wegner, editors, Research Directions in Object-Oriented Programming, pages 219--252. MIT Press, 1987.

....such as procedures, modules, and interfaces; dependencies among components such as client supplier, inheritance, and control flow; and attributes such as component type, interface size, and interconnection strength. The structure of a system is the organization and interaction of these artifacts [2]. One computer aided technique of reconstructing structural models is reverse engineering. The process of reverse engineering identifies the system s current components, discovers their dependencies, and generates abstractions to manage complexity. This understanding can then improve subsequent ....

H. L. Ossher. A mechanism for specifying the structure of large, layered systems. In B. D. Shriver and P. Wegner, editors, Research Directions in Object-Oriented Programming, pages 219--252. MIT Press, 1987.

....actually be employed, after every update of the system, and that any discrepancies thus detected would be immediately corrected. The enforcement of constraints over the structure and behavior of systems is not entirely new, however. An early, but specialized, attempt to do so is that of Ossher [15], who built an environment which allows for the specification and enforcement of a very special kind of structural laws, that regulate moduleinterconnection in layered systems. One should also mention in this context the very interesting work on computational reflection, such as the metaobject ....

Ossher H. L. A mechanism for specifying the structure of large, layered systems. In Bruce Shriver and Peter Wegner, editors, Research Directions in Object-Oriented Programming, pages 219--252. MIT Press, 1987.

....type, interface size, and interconnection strength. The structure of a system is the organiza Scale Extensibility Applicability Monolithic End user programmable 10KLOC 10MLOC Task specfic General purpose Figure 1: Reverse engineering design space tion and interaction of these artifacts [29]. 4.1 Key requirements To be considered successful, a reverse engineering tool must be flexible with respect to its applicability to multiple domains, it must be extensibile with respect to its functionality, and it must support the analysis of large (O(10 6 ) LOC) programs. These three ....

H. L. Ossher. A mechanism for specifying the structure of large, layered systems. In B. D. Shriver and P. Wegner, editors, Research Directions in ObjectOriented Programming, pages 219--252. MIT Press, 1987.

....procedures, modules, and interfaces; dependencies among components such as clientsupplier, inheritance, and control flow; and attributes such as component type, interface size, and interconnection strength. A software system s hyperstructure is the organization and interaction of these artifacts [Oss87] This level of abstraction is called the software architecture level [PW92] Classical architecture has concepts that are desirable for flexible software architecture, including multiple views and architectural styles. For example, a building architect would provide one representation of the ....

Harold L. Ossher. A mechanism for specifying the structure of large, layered systems. In Bruce D. Shriver and Peter Wegner, editors, Research Directions in Object-Oriented Programming, pages 219--252. MIT Press, 1987.

....such as procedures, modules, and interfaces; dependencies among components such as client supplier, inheritance, and control flow; and attributes such as component type, interface size, and interconnection strength. The structure of a system is the organization and interaction of these artifacts [26]. One class of techniques of reconstructing structural models is reverse engineering. Using reverse engineering approaches to reconstruct the architecture aspects of software can be termed structural redocumentation. The University of Victoria s work is centered around Rigi [27] an environment ....

H. L. Ossher. A mechanism for specifying the structure of large, layered systems. In B. D. Shriver and P. Wegner, editors, Research Directions in Object-Oriented Programming, pages 219--252. MIT Press, 1987.

....subsystems, and interfaces; dependencies among components such as supplier client, composition, and control flow relations; and attributes such as component type, interface size, and interconnection strength. The structure of the system is the organization and interaction of these artifacts [6]. This paper describes an approach to supporting software evolution through reverse engineering. In particular, the paper focuses on extensions to an existing reverse engineering environment to make it flexible and user programmable, and hence domainretargetable. By providing a user programmable ....

H. L. Ossher. A mechanism for specifying the structure of large, layered systems. In B. D. Shriver and P. Wegner, editors, Research Directions in ObjectOriented Programming, pages 219--252. MIT Press, 1987.

....the structure may take many forms. In this dissertation, I use the term, structure, in the same broad, flexible sense defined by Ossher: Any system consists of parts, such as modules, procedures, classes and methods. The structure of the system is the organization and interactions of those parts. [Ossher 1987, p. 219] To give a sense of the reason for this broad definition, consider the variety of structural information in the artifacts of two example software systems the Chiron user interface system [Keller et al. 1991] and the GNU groff document formatting system. These two systems were chosen as ....

....with the intended structure. Ossher built such a checker to support a comparison of the structure as expressed in a system s source code with a GRID description of the intended structure [Ossher 1984] A GRID is a formalism developed by Ossher to describe the structure of large, layered systems [Ossher 1987]. This kind of checker was also built as part of the Software Landscape environment that supports a visual module interconnection language [Penny 1993] Similar to Ossher s approach, in the Software Landscape, relations between program entities could be extracted and compared directly to relations ....

[Article contains additional citation context not shown here]

Ossher, H. 1987. A Mechanism for Specifying the Structure of Large, Layered Systems. In Research Directions in Object-Oriented Programming, B. Shriver and P. Wegner Eds. The MIT Press, Cambridge, MA, 219--252.

....by their class made the job 74 of retargeting to Ada somewhat easier, since retargeting the source code output was simply a matter of editing a single file and changing the keywords that were generated. This orthogonal grouping is reminiscent of the grid organization proposed by Ossher [Ossher 87] and indicates more of a syntaxdirected organization than an object oriented one. The benefits of this organization suggest another type of source code transformation that could be applied to a software system by a meaningpreserving restructuring tool. Although this type of modification would ....

H. L. Ossher. A mechanism for specifying the structure of large, layered programs. In Research Directions in Object-Oriented Programming, B. Shriver and P. Wegner, eds.. MIT Press, Boston, MA, 1987: 219-252.

.... layers , together with the restriction on communication between modules which allows messages to be sent only within a layer, or down one layer. This regularity, frequently employed for large systems, provides a useful framework within which a system can be constructed, managed and understood [19]. A single system may have many different regularities, each with its own specific role to play. As an example of a special purpose regularity, consider the following token based protocol which might be employed by a distributed system S in order to ensure mutual exclusion with respect to a given ....

....architecture for general software systems, of the kind described here. The following are the main such efforts known to the author: Perhaps the earliest attempt to provide an explicit global law for general software systems (not for some specific kind of systems, like databases) was by Ossher [18, 19]. He built a mechanism that imposes a specified layered module interconnection structure on a given system, which bears some similarity to our example law discussed in this section. The Meta system of Marzullo and Wood [7] has a global set of policy rules about the 10 interaction between ....

Harold L. Ossher. A Mechanism for Specifying the Structure of Large, Layered Systems. In Bruce Shriver and Peter Wegner (editor), Research Directions in Object-Oriented Programming, pages 219--252. MIT Press, 1987.

No context found.

H. Ossher. A Mechanism for Specifying the Structure of Large, Layered Systems. In Bruce Shriver and Peter Wegner, editors, Research Directions in Object-Oriented Programming, pages 219--252. MIT Press, 1987.

Online articles have much greater impact   More about CiteSeer.IST   Add search form to your site   Submit documents   Feedback  

CiteSeer.IST - Copyright Penn State and NEC