H. Ossher and P. Tarr. Using multidimensional separation of concerns to (re)shape evolving software. Communications of the ACM, Oct. 2001.  Home/Search   Document Not in Database   Summary   ACM   TOC   Related Articles   Check

....that is spread across one or more modular units. Such code is often called tangled or crosscutting code. Crosscutting code results in software that is less reusable, less readable and harder to maintain. The research community is responding to this problem by proposing multiple programming tools [KLM 97, BA01, OT01, LOO01] to extend existing programming languages to support modularization of such crosscutting concerns and hence, achieve a more advanced level of separation of concerns. The language extensions often provide some syntactical construct for modularizing these crosscutting concerns. Many of these ....

....language extensions often provide some syntactical construct for modularizing these crosscutting concerns. Many of these programming tools for advanced separation of concerns are built by creating or modifying the compiler of the existent languages that weave the crosscutting concern in the code [KLM 97, BA01, OT01, LOO01]. Another approach to achieve advanced separation of concerns is through the use of reflection [Sul01] The advantage of reflection over a modified compiler is that it allows one to dynamically weave and unweave the crosscutting concern into or out of the application. This dynamism of adding and ....

[Article contains additional citation context not shown here]

Harold Ossher and Peri Tarr. Using multidimensional separation of concerns to (re)shape evolving software. Communications of the ACM, 44(10):43--50, 2001.

....Separation of Concerns. Recent e#orts investigating multidimensional separation of concerns (MDSOC) 20] have made progress in handling the case of conflicting modularizations. MDSOC deals with the multi dimensional hyperspace of concerns that software must accommodate. Approaches like [11] which a#ord MDSOC enable software to be developed under multiple system decompositions that can exist simultaneously. Similar to the concept of code weaving, these independently developed system slices are automatically bound into a consistent whole. Component Containers. Another approach ....

H. Ossher and P. Tarr. Using multidimensional separation of concerns to (re)shape evolving software. Communications of the ACM, 44(10):43--50, 2001.

....serious drawbacks in terms of re usability. 1.4 Related Reading This project began with the idea of trying to describe performance as an aspect, especially for numerical applications. The introductory reading of the Communications of the ACM issue dated October 2001 [EFB01, EAK 01, LOO01, OT01, BA01, KHH 01, PC01, MWB 01, CKF 01, NEF01, GBNT01, Sul01, NEF01, GBNT01, Sul01] and of [CE00, chap. 8] was used so as to learn the principles and de nitions of AOP. This led to the reading of the papers cited in the remainder of this thesis. 1.5 About this Dissertation This thesis ....

HAROLD OSSHER AND PERI TARR. Using multidimensional separation of concerns to (re)shape evolving software. Communications of the ACM, 44(10):4350, 2001.

....macro examples (whose programs exceed 30K lines of code) Our work provides strong evidence that SWR scales to synthesis of large systems. 1 Introduction Multi Dimensional Separation of Concerns (MDSOC) is an approach to simplify the specification, development, and design of software [14][20] 15] MDSOC asserts that a program can be decomposed or modularized in many different ways: by classes, by features, by aspects, by functions, etc. Each partitions the space of primitive software artifacts, called units, in a different way. Thus, partitioning software by features ....

H. Ossher and P. Tarr. "Using Multi-Dimensional Separation of Concerns to (Re)Shape Evolving Software." CACM 44(10): 43-50, October 2001.

....the Algebraic Hierarchical Equations for Application Design (AHEAD) model that generalizes equational specifications to multiple programs and multiple representations. AHEAD is related to other methodologies, such as Aspect Oriented Programming [18] and Multi Dimensional Separation of Concerns [22], and thus our results are not GenVoca specific. We review a tool set that supports AHEAD. As a demonstration of AHEAD s viability, we have bootstrapped AHEAD tools, generating over 100K LOC of Java (and other non code artifacts) solely from equational specifications automatically, a task that was ....

....currently lacking in the AHEAD tool set. Gray has shown how aspects apply to non code artifacts [12] AHEAD shows how both code and non code artifacts can be refined simultaneously in collection hierarchies. Multi Dimensional Separation of Concerns (MDSC) is another program refinement technology [22]. We have built GenVoca generators using Hyper J. Layers correspond to hyperslices, and GenVoca equations correspond to compositions of hyperslices. Further, MDSC advocates that the techniques for assembling customized code from hyperslice compositions should also work for non code artifacts as ....

H. Ossher and P. Tarr. "Using Multi-Dimensional Separation of Concerns to (Re)Shape Evolving Software." CACM 44(10): 43-50, October 2001.

....took place during the early 1990 s, though C the language was developed between 1983 5 [2] and the first release version of Smalltalk was made in 1980 (with roots stretching much further back) 3] The first general purpose language version of AspectJ [4] was 0.2, released in 1999. Hyper J [5] was released in 2000. Both have roots going much further back. By the very dangerous process of extrapolation, that means we could start to see widespread adoption of aspect oriented programming sometime around 2005 6, and aspectorientation as the dominant paradigm by 2010. How do we get from ....

....those that retain source level compatibility with every legal Java program) are quite happy to specify aspects composition rules interceptors in configuration files, and then use byte code engineering to modify the byte codes produced by the Java compiler to achieve an end. Both AspectJ and Hyper J[5] for example can be viewed as taking legal Java programs, adding additional information in text files accompanying the program, and producing a set of Java byte codes . Obviously, I am not a supporter of random changes to the Java language. I do believe though that aspect aware language ....

Ossher, H. and P. Tarr, Using Multidimensional Separation of Concerns to (re)shape Evolving Software. Communications of the ACM, 2001. 44(10): p. 43-49.

....modular reasoning to the AspectJ language. Future work will also investigate the relevance of our proposal to other aspect orientation programming languages and techniques, such as composition filters [4] adaptive methods [16] and multidimensional separation of concerns as embodied by Hyper J [22, 25]. ACKNOWLEDGMENTS We would like to thank Yoonsik Cheon, Todd Millstein, Markus Lumpe, and Robyn Lutz, for their helpful comments on a draft of this paper. The work of Leavens was supported in part by the US National Science Foundation grants CCR 0097907 and CCR0113181. The work of both authors ....

H. Ossher and ih Tarr. Using multidimensional separation of concerns to (re)shape evolving software. Commun. 44(10):43 50, Oct. 2001.

No context found.

H. Ossher and P. Tarr. Using multidimensional separation of concerns to (re)shape evolving software. Communications of the ACM, Oct. 2001.

No context found.

Ossher, H., Tarr, P.: Using multidimensional separation of concerns to (re)shape evolving software. Communications of the ACM 44 (2001) 43--50

No context found.

H. Ossher and P. Tarr, "Using multidimensional separation of concerns to (re)shape evolving software," Communications of the ACM, vol. 44, no. 10, pp. 43--50, 2001.

No context found.

Ossher, H. and P. Tarr, Using multidimensional separation of concerns to (re)shape evolving software. Communications of the ACM 44 (2001), pp. 43--50.

No context found.

H. Ossher and P. Tarr, "Using multidimensional separation of concerns to (re)shape evolving software," Communications of the ACM, 44(10):43--50, Oct. 2001.

No context found.

H. Ossher and P. Tarr. Using multidimensional separation of concerns to (re)shape evolving software. Commun. ACM, 44(10):43--50, Oct. 2001.

No context found.

H. Ossher and P. Tarr, "Using Multi-Dimensional Separation of Concerns to (Re)Shape Evolving Software," Comm. ACM, vol. 44, no. 10, pp. 43-50, Oct. 2001.

No context found.

Ossher, H. and Tarr, P. Using Multi-Dimensional Separation of Concerns to (Re)Shape Evolving Software, Communications of the ACM, vol. 44, no. 10, 2001.

No context found.

H. Ossher and P. Tarr, \Using multidimensional separation of concerns to (re)shape evolving software," Communications of the ACM, vol. 44, no. 10, pp. 43-50, 2001.

No context found.

H. Ossher and P. Tarr, "Using multidimensional separation of concerns to (re)shape evolving software," Communications of the ACM, 44(10):43--50, Oct. 2001.

No context found.

Ossher, H. and Tarr, P. "Using Multidimensional Separation of Concerns to (Re)Shape Evolving Software". Communications of the ACM, 44 (10) 2001, pp.43-50.

No context found.

Ossher, H. and Tarr, P. Using multidimensional separation of concerns to (re)shape evolving software. Communications of the ACM, 44(10):43-50, Oct. 2001.

No context found.

H. Ossher and P. Tarr, "Using Multidimensional Separation of Concerns to (re)shape Evolving Software," Communications of the ACM, vol. 44, pp. 43-49, 2001.

No context found.

H. Ossher and P. Tarr. Using multidimensional separation of concerns to (re)shape evolving software. Commun. ACM, 44(10):43--50, Oct. 2001. Special Issue on AspectOriented Programming.

No context found.

H. Ossher and P. Tarr. Using multidimensional separation of concerns to (re)shape evolving software. Commun. ACM, 44(10):43-50, Oct. 2001.

No context found.

H. Ossher, P. Tarr. Using Multidimensional Separation of Concerns to (Re)Shape Evolving Software. Communications of the ACM, October 2001, vol 44, num 10.

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