PharoJS is an open source infrastructure (framework + middleware + tools) that allows developing ... more PharoJS is an open source infrastructure (framework + middleware + tools) that allows developing and testing in Pharo Smalltalk, applications which ultimately run on a JavaScript (JS) interpreter. Historically, and up to Pharo version 9, PharoJS generated JS code following the ECMAScript 5 (ES5) standard. Since, ES5 has no explicit support for classes, PharoJS transpiler had to generate JS code that mimicked the Pharo class structures, behaviors and hierachies. After migrating PharoJS to Pharo version 10, we decided to adopt the ECMAScript 6 (ES6) standard. One of the major features provided by ES6 is a set of new constructs to explicitly support class-based programming. In this paper, we describe the changes we have made to PharoJS to switch from ES5 of ES6. We describe the consequences of adopting ES6 on generated code, and the positive impact on all peformance metrics.
Tail call elimination is a common optimization, particularly for functional languages, where it i... more Tail call elimination is a common optimization, particularly for functional languages, where it is a semantic requirement. The value is less obvious in an Object-Oriented language such as Smalltalk. In this paper we show that it has a significant execution-time improvement for typical workloads. It also has significant memory savings for certain programs, and enables novel (to Smalltalk) programming styles.
Proceedings of the 15th ACM SIGPLAN International Workshop on Virtual Machines and Intermediate Languages
Since the 1980s, Deutsch-Schiffman dispatch has been the standard method dispatch mechanism for l... more Since the 1980s, Deutsch-Schiffman dispatch has been the standard method dispatch mechanism for languages like Smalltalk, Ruby, and Python. While it is a huge improvement over the simple, semantic execution model, it has some significant drawbacks for modern hardware and applications. This paper proposes an alternative dispatch mechanism that addresses these concerns, with only memory space as a trade-off, that should demonstrate dynamic performance only slightly worse than the best possible with full type information for the program. CCS Concepts: • Software and its engineering → Runtime environments; Dynamic compilers; Just-in-time compilers.
: The software architecture of a software-intensive system greatly determines system quality. Eva... more : The software architecture of a software-intensive system greatly determines system quality. Evaluating that architecture for fitness of purpose before the system is implemented or undergoes a major modification is a cost-effective approach to uncovering deficiencies early and reducing risk. When used appropriately, software architecture evaluations can have a favorable effect on a delivered or modified government system. This technical note describes the application of the Software Engineering Institute's (SEI) Architecture Tradeoff Analysis Method (trademark) (ATAM (trademark)) to the U.S. Army's Warfighter Information Network-Tactical (WIN-T) system. The WIN-T system is being developed by a government-contractor team headquartered at the U.S. Army's Communications and Electronics Command (CECOM) in Ft. Monmouth, New Jersey. This technical note presents the WIN-T program context, the definition of software architecture, and the background of the WIN-T organization and...
Proceedings of the International Workshop on Smalltalk Technologies, 2015
Amber is a wonderful Smalltalk programming environment that runs on top of Javascript, including ... more Amber is a wonderful Smalltalk programming environment that runs on top of Javascript, including a browser-based IDE and compiler, as well as command-line support. The only challenge is that execution performance can be 1--2 orders of magnitude slower than native Javascript code. Amber-Direct is a series of modest changes to the compiler and some infrastructure classes and methods that bring most generated programs to within a factor of two of native Javascript. The challenge we faced was maintaining a seamless integration into existing Javascript classes while maximizing fidelity to Smalltalk execution semantics.
The component reliability model described in (4) requires that component execution frequency, rel... more The component reliability model described in (4) requires that component execution frequency, reliability, and trans- formations be analysed statistically based on parameters to the component. This paper describes work in progress to- ward that goal.
For a typical software system, it is generally consid- ered infeasible to calculate system reliab... more For a typical software system, it is generally consid- ered infeasible to calculate system reliability from the reliabilities of its constituent components because soft- ware systems, unlike hardware systems, tend to violate the underlying independence assumptions inherent in the usual reliability calculations. We present a set of component design and interaction rules which, if fol- lowed in software development, can produce systems with the highly independent components necessary in order to legitimately calculate system reliability from component reliability. We present a system which fol- lows our rules, and show that in this case system relia- bility calculated from component reliabilities was very close to the true system reliability.
ons will be clearly stated on the lab. Individual labs are worth very few marks, so if they are n... more ons will be clearly stated on the lab. Individual labs are worth very few marks, so if they are not handed in on time, no opportunity will be available to hand them in later, and they will not cause an INC grade. The departmental assignment policy applies. You are expected to have read and understood it. If you have any questions about this please come and talk to me. All labs and assignments must have complete header documentation, including a statement of authorship. Labs Lab time will be used for working through formal labs with example problems. These labs will be written up and handed in at the end of the lab. There will be 9 labs to be handed in (none for weeks before assignments are due, although teaching assistants and the professor will be available for questions, or for November 6 when the lab time will be used for an on-line test). Course Outline The text for the course (and for CPS209) is The Schematics of Computati
Program paths (equivalently program input subdomains) are interesting for a variety of purposes i... more Program paths (equivalently program input subdomains) are interesting for a variety of purposes in software engineer-ing, including testing, reliability, and estimating of resource requirements. The full potential of paths is often not real-ized because it is difficult to generate the most important paths from the – potentially infinite – set of possible paths for a program. This paper introduces two probabilistic algorithms to gen-erate paths based on an operational profile describing the probability distribution of the possible inputs. The first is a theoretically clean, but implementationally difficult algo-rithm based on program continuations. The second is a more practical, but less accurate algorithm based on Monte Carlo techniques. Prototype implementations of both algorithms are briefly examined. Smalltalk provides the ideal testbed for these ideas because of its metaprogramming capabilities. These implementa-tions utilize: dynamic code generation, full object orienta-tion, ...
This paper describes the use of WatIF, a simple compiler intermediate form based on the λ-calculu... more This paper describes the use of WatIF, a simple compiler intermediate form based on the λ-calculus, for writing compiler front-ends for a wide range of programming languages. Somewhat similar intermediate forms have been previously used for single-language compilers, but applicability to such a range of languages has not been previously reported. Examples of code generated by front-ends for several languages show the large set of application-language features supported by WatIF.
Proceedings of the 3rd international workshop on Free composition - FREECO '12, 2012
ABSTRACT We are creating a visual programming environment for userprogrammers (who are not traine... more ABSTRACT We are creating a visual programming environment for userprogrammers (who are not trained programmers) that they will use a few hours per week.. Because of this sporadic usage pattern, traditional code structuring approaches are both overkill and difficult to learn. Therefore, the environment is designed to support very fluid relationships among code chunks, in terms of their relationship and evolution, as well as very fluid interactions among code chunks when processing data.
Series on Component-Based Software Development, 2004
Software components are today the most promising approach to dealing with the complexity and unev... more Software components are today the most promising approach to dealing with the complexity and uneven quality of software systems. The design-using-components paradigm has been extremely successful in almost every engineering field, with its benefits of rapid, routine, reliable system construction. The central dilemma of software design using components is that component developers cannot know how their components will be used and so cannot describe component properties for an unknown, arbitrary situation; but if the component customer (system designer) must determine relevant properties of each component before using it, component-based development loses much of its appeal. In technical terms, component behavior depends on the operational profile the component sees when in place in a larger system; in turn, that profile depends both on system usage and the internal structure of the system, neither of which can be known to the component developer. A solution to the dilemma is presented in a new theory of component-based design. The component developer performs measurements and provides a component description in such a way that the component buyer can later factor in usage information without repeating the measurements. The heart of the theory is a method of calculating how an operational profile is transformed by one component to be presented to the next component in a system. The theory originated in an investigation of system reliability to be calculated from component reliabilities. It then became apparent that similar methods applied to other system properties, such as its run time and its security properties. Each of these properties can be measured for components by conventional testing and these measurements enter into calculations of the composite system properties. The system designer does not construct or experiment with a trial system, but uses a CAD tool to make predictions of how the system will behave. Preliminary validation experiments support the theory.
Online marking of assignments can lead to improved marking consistency and integrate well with on... more Online marking of assignments can lead to improved marking consistency and integrate well with on-line mark reporting. It can also be easier for the marker. Unfortunately, most such systems do not include good feedback mechanisms for the students. This paper describes an environment that provides online marking with convenient, structured and detailed feedback.
PharoJS is an open source infrastructure (framework + middleware + tools) that allows developing ... more PharoJS is an open source infrastructure (framework + middleware + tools) that allows developing and testing in Pharo Smalltalk, applications which ultimately run on a JavaScript (JS) interpreter. Historically, and up to Pharo version 9, PharoJS generated JS code following the ECMAScript 5 (ES5) standard. Since, ES5 has no explicit support for classes, PharoJS transpiler had to generate JS code that mimicked the Pharo class structures, behaviors and hierachies. After migrating PharoJS to Pharo version 10, we decided to adopt the ECMAScript 6 (ES6) standard. One of the major features provided by ES6 is a set of new constructs to explicitly support class-based programming. In this paper, we describe the changes we have made to PharoJS to switch from ES5 of ES6. We describe the consequences of adopting ES6 on generated code, and the positive impact on all peformance metrics.
Tail call elimination is a common optimization, particularly for functional languages, where it i... more Tail call elimination is a common optimization, particularly for functional languages, where it is a semantic requirement. The value is less obvious in an Object-Oriented language such as Smalltalk. In this paper we show that it has a significant execution-time improvement for typical workloads. It also has significant memory savings for certain programs, and enables novel (to Smalltalk) programming styles.
Proceedings of the 15th ACM SIGPLAN International Workshop on Virtual Machines and Intermediate Languages
Since the 1980s, Deutsch-Schiffman dispatch has been the standard method dispatch mechanism for l... more Since the 1980s, Deutsch-Schiffman dispatch has been the standard method dispatch mechanism for languages like Smalltalk, Ruby, and Python. While it is a huge improvement over the simple, semantic execution model, it has some significant drawbacks for modern hardware and applications. This paper proposes an alternative dispatch mechanism that addresses these concerns, with only memory space as a trade-off, that should demonstrate dynamic performance only slightly worse than the best possible with full type information for the program. CCS Concepts: • Software and its engineering → Runtime environments; Dynamic compilers; Just-in-time compilers.
: The software architecture of a software-intensive system greatly determines system quality. Eva... more : The software architecture of a software-intensive system greatly determines system quality. Evaluating that architecture for fitness of purpose before the system is implemented or undergoes a major modification is a cost-effective approach to uncovering deficiencies early and reducing risk. When used appropriately, software architecture evaluations can have a favorable effect on a delivered or modified government system. This technical note describes the application of the Software Engineering Institute's (SEI) Architecture Tradeoff Analysis Method (trademark) (ATAM (trademark)) to the U.S. Army's Warfighter Information Network-Tactical (WIN-T) system. The WIN-T system is being developed by a government-contractor team headquartered at the U.S. Army's Communications and Electronics Command (CECOM) in Ft. Monmouth, New Jersey. This technical note presents the WIN-T program context, the definition of software architecture, and the background of the WIN-T organization and...
Proceedings of the International Workshop on Smalltalk Technologies, 2015
Amber is a wonderful Smalltalk programming environment that runs on top of Javascript, including ... more Amber is a wonderful Smalltalk programming environment that runs on top of Javascript, including a browser-based IDE and compiler, as well as command-line support. The only challenge is that execution performance can be 1--2 orders of magnitude slower than native Javascript code. Amber-Direct is a series of modest changes to the compiler and some infrastructure classes and methods that bring most generated programs to within a factor of two of native Javascript. The challenge we faced was maintaining a seamless integration into existing Javascript classes while maximizing fidelity to Smalltalk execution semantics.
The component reliability model described in (4) requires that component execution frequency, rel... more The component reliability model described in (4) requires that component execution frequency, reliability, and trans- formations be analysed statistically based on parameters to the component. This paper describes work in progress to- ward that goal.
For a typical software system, it is generally consid- ered infeasible to calculate system reliab... more For a typical software system, it is generally consid- ered infeasible to calculate system reliability from the reliabilities of its constituent components because soft- ware systems, unlike hardware systems, tend to violate the underlying independence assumptions inherent in the usual reliability calculations. We present a set of component design and interaction rules which, if fol- lowed in software development, can produce systems with the highly independent components necessary in order to legitimately calculate system reliability from component reliability. We present a system which fol- lows our rules, and show that in this case system relia- bility calculated from component reliabilities was very close to the true system reliability.
ons will be clearly stated on the lab. Individual labs are worth very few marks, so if they are n... more ons will be clearly stated on the lab. Individual labs are worth very few marks, so if they are not handed in on time, no opportunity will be available to hand them in later, and they will not cause an INC grade. The departmental assignment policy applies. You are expected to have read and understood it. If you have any questions about this please come and talk to me. All labs and assignments must have complete header documentation, including a statement of authorship. Labs Lab time will be used for working through formal labs with example problems. These labs will be written up and handed in at the end of the lab. There will be 9 labs to be handed in (none for weeks before assignments are due, although teaching assistants and the professor will be available for questions, or for November 6 when the lab time will be used for an on-line test). Course Outline The text for the course (and for CPS209) is The Schematics of Computati
Program paths (equivalently program input subdomains) are interesting for a variety of purposes i... more Program paths (equivalently program input subdomains) are interesting for a variety of purposes in software engineer-ing, including testing, reliability, and estimating of resource requirements. The full potential of paths is often not real-ized because it is difficult to generate the most important paths from the – potentially infinite – set of possible paths for a program. This paper introduces two probabilistic algorithms to gen-erate paths based on an operational profile describing the probability distribution of the possible inputs. The first is a theoretically clean, but implementationally difficult algo-rithm based on program continuations. The second is a more practical, but less accurate algorithm based on Monte Carlo techniques. Prototype implementations of both algorithms are briefly examined. Smalltalk provides the ideal testbed for these ideas because of its metaprogramming capabilities. These implementa-tions utilize: dynamic code generation, full object orienta-tion, ...
This paper describes the use of WatIF, a simple compiler intermediate form based on the λ-calculu... more This paper describes the use of WatIF, a simple compiler intermediate form based on the λ-calculus, for writing compiler front-ends for a wide range of programming languages. Somewhat similar intermediate forms have been previously used for single-language compilers, but applicability to such a range of languages has not been previously reported. Examples of code generated by front-ends for several languages show the large set of application-language features supported by WatIF.
Proceedings of the 3rd international workshop on Free composition - FREECO '12, 2012
ABSTRACT We are creating a visual programming environment for userprogrammers (who are not traine... more ABSTRACT We are creating a visual programming environment for userprogrammers (who are not trained programmers) that they will use a few hours per week.. Because of this sporadic usage pattern, traditional code structuring approaches are both overkill and difficult to learn. Therefore, the environment is designed to support very fluid relationships among code chunks, in terms of their relationship and evolution, as well as very fluid interactions among code chunks when processing data.
Series on Component-Based Software Development, 2004
Software components are today the most promising approach to dealing with the complexity and unev... more Software components are today the most promising approach to dealing with the complexity and uneven quality of software systems. The design-using-components paradigm has been extremely successful in almost every engineering field, with its benefits of rapid, routine, reliable system construction. The central dilemma of software design using components is that component developers cannot know how their components will be used and so cannot describe component properties for an unknown, arbitrary situation; but if the component customer (system designer) must determine relevant properties of each component before using it, component-based development loses much of its appeal. In technical terms, component behavior depends on the operational profile the component sees when in place in a larger system; in turn, that profile depends both on system usage and the internal structure of the system, neither of which can be known to the component developer. A solution to the dilemma is presented in a new theory of component-based design. The component developer performs measurements and provides a component description in such a way that the component buyer can later factor in usage information without repeating the measurements. The heart of the theory is a method of calculating how an operational profile is transformed by one component to be presented to the next component in a system. The theory originated in an investigation of system reliability to be calculated from component reliabilities. It then became apparent that similar methods applied to other system properties, such as its run time and its security properties. Each of these properties can be measured for components by conventional testing and these measurements enter into calculations of the composite system properties. The system designer does not construct or experiment with a trial system, but uses a CAD tool to make predictions of how the system will behave. Preliminary validation experiments support the theory.
Online marking of assignments can lead to improved marking consistency and integrate well with on... more Online marking of assignments can lead to improved marking consistency and integrate well with on-line mark reporting. It can also be easier for the marker. Unfortunately, most such systems do not include good feedback mechanisms for the students. This paper describes an environment that provides online marking with convenient, structured and detailed feedback.
Uploads
Papers by Dave Mason