Toward a Shared Vocabulary for Systems Engineering Paola Di Maio, University of Strathclyde/ISTCS. org WIMS 11, Norway 25 -27 May 2011
CONTENTS a. About this paper, what's new? b. How did it all start c. Worldview d. About Systems Engineering e. The trigger: wikification f. Entanglement g. Mixed Method Approach h. Text extraction i. integrating Dimensions j. The proposed architecture k. Future work WIMS 11, Norway 25 -27 May 2011
Whats new? Shared vocabularies are not new but it is surprising to find that a) in real world situation not many people use them b) not many people outside the SW community know about them c) some efforts started in 2009, when I joined INCOSE This paper introduces: - SEBOK and related challenges - An integrated, socio-technical perspective (exctracting the vocabulary is just a beginning) - Novel worldwiew and approach WIMS 11, Norway 25 -27 May 2011
SYSTEMIC PROBLEMS a. ARE INHERENTLY NON-QUANTIFIABLE • CONTAIN GENUINE UNCERTAINTIES • CANNOT BE CAUSALLY MODELLED OR SIMULATED • REQUIRE A JUDGMENTAL APPROACH • TECHNICALLY FLAWLESS SYSTEM, FAIL DUE TO HUMAN AND SOCIAL FACTORS WIMS 11, Norway 25 -27 May 2011
HOW DID IT ALL START Joined INCOSE in 2009, presented a paper at EUSEC 2010 Meet Art Pyster from Stevenson, invited to join SEBOK effort SEBOK is at a critical turning point to move to WIKI Involved in effort, and discussing approaches toward semantic wiki See 'challenges to wikification' for more details WIMS 11, Norway 25 -27 May 2011
WORLDVIEW (EPISTEMOLOGICAL PARADIGM) Postpositivism quantitative determination reductionism knowledge can never be found empirical observation and measurement (scientific method) theory verification cause-and-effect deductive 19 th century writers Conte and Locke Constructivism qualitative understanding multiple participant meanings (humans) social and historical construction theory generation open-ended questioning inductive Advocacy and Participatory qualitative (can lead to quantitative) political empowerment issue oriented collaborative change-oriented focus on marginalized or disenfranchised Marx, Marcuse Pragmatic consequences of actions problem-centered pluralistic real-world practice oriented (Creswell) TRANSFORMATIVE Delivers systemic change SYSTEMIC Tackles issues as a whole, takes the integrated perspective WIMS 11, Norway 25 -27 May 2011
SOCIO TECHNICAL SYSTEM CONSTANT CHANGE/EVOLUTION CAUSAL DEPENDENCIES INTERACTIONS AND TRANSFORMATIONS PSYCHOLOGICAL AND SOCIAL FACTORS. WIMS 11, Norway 25 -27 May 2011
ABOUT Systems Engineering The discipline of systems engineering (SE) has been recognized for 50 years as essential to the development of complex systems. [. . . ] SE is still treated primarily as heuristics learned by each practitioner during the personal experimentation of a career. The heuristics known by each differ, as shown by the fractured development of SE “standards” and SE certification. [. . . ] As a result of this heuristic understanding of the discipline, it has been nearly impossible to quantify the value of SE to programs. [Sheard 2000] [. . . ] The differences in understanding, however, just as typically result in disagreement over the level and formality of the practices to include. [Shenhar 1997] WIMS 11, Norway 25 -27 May 2011
Image Source: wikipedia
KNOWLEDGE SHARING Source: P. Di Maio WIMS 11, Norway 25 -27 May 2011
WIKIFICATION, CHALLENGES 1 Provide justification for what is essentially a cultural change from a ‘closed world’ perspective, to the open dynamic world, ’ Many implications. it reflects in the logic of the technical system being implemented, and in the organisational and management setup of development practices and documentation itself. WIMS 11, Norway 25 -27 May 2011
WIKIFICATION, CHALLENGES 2 What's the best way to migrate the content from hard coded documents to a SMWiki environment? What kind of structure should the wiki provide to enable meaningful representation of the content? What model of governance is suitable for a distributed multiauthors team that would allow both max flexibility to encourage contributions, and that would also support sufficient validation for the quality of the contribution? WIMS 11, Norway 25 -27 May 2011
ENTANGLEMENT/CIRCULARITY Loading the corpus to a wiki requires some decision in relation to the ‘structure’ which means, how to best organise the corpus depends on how the corpus structure is defined. Subsequent ongoing edits and editorial changes to body of knowledge after it has been opened up on the wiki, will be determined by what editorial process is in place. The editorial process in turn depends on what governance model is chose. These interdependencies between various steps are what we refer to as ‘entanglement’ Solving the entanglement can be done by a) identifying and mapping all the key factors and their dependencies b) constructing a the project schedule and a plan build around them, in what is essentially a very ‘agile’ style.
ENTANGLEMENT 1 MADE UP OF DIFFERENT PROBLEM SPACES: TECHNICAL COGNITIVE ORGANISATIONAL/ POLICY
ENTANGLEMENT 2
MIXED APPROACH TEXT EXTRACTION MORPHOLOGICAL ANALYSIS INTEGRATION OF VARIOUS ASPECTS OF THE COMPLEX PROBLEM
Automated Text Extraction Approach (ATX) STEPS: - Parsed Chap 2 using different generic tools (Opencalais, Alchemy) - Compared the output with the glossary created manually by the editors with the output of ATX OVERVIEW OF RESULTS - ATX yields more granular results and in some cases provides some categories (structure) to the terms RECOMMENDATIONs - Editors of the glossary could take into account additional granularity and categorization structure. - New ad hoc categories for SE could be created. - Categories could serve as metadata and could be encoded using SKOS or other suitable supporting schema MORE INFO: http: //tinyurl. com/2 vwq 5 wh
Morphological Analysis a. From classical Greek (morphe) : and means shape or form b. Morphology is the study of the shape and arrangement of parts of an object, and how these parts "conform" to create a whole or Gestalt. c. The "objects" in question can be physical objects (e. g. an organism, an anatomy, a geography or an ecology) or mental objects (e. g. word forms, concepts or systems ofideas). A methodological framework for creating models of systems and processes, which cannot be meaningfully quantified • Extended typology analysis was invented as early as the 1930’s by Fritz Zwicky, professor of astronomy at the California Institute of Technology – the famous Caltech in Pasadena http: //www. swemorph. com/pdf/it-webart. pdf
What is MA used for? - Complex issue which is not well formulated or defined; (”wicked problem”) - Well formulated/defined issue, but with no single solution (different solutions depending on…) - Well defined problem with aspecific solution which can be worked out. a. Mess • Problem • Puzzle (Russell Ackoff: Redesigning the Future, 1974; Michael Pidd: Tools for Thinking, 1996. )
PROCESS in answering the question: What's the best way to migrate content from a hardcoded document to a SMWiki environment, and then to maintain the future updates? 1. to upload the working version of a Body of Knowledge to a wiki after it has been finalized. This can be done as a single step, or a series of steps, for example one chapter at the time. (SINGLE STEP, FEW TEAM MEMBERS) 2. to allow users to make ongoing edits and changes throughout the life of the BOK In our description, a process consists of a series of dynamic steps (as opposed to a model, which is a static representation’). The following processes are required to deploy a body of knowledge to a wiki: (EDITORIAL PROCESS, ALL AUTHORS)
STRUCTURE “What kind of content structure should the wiki provide to enable meaningful representation of the content? ” It may be useful to define further 'structure'. It is envisaged that a minimum of two levels of structure can be easily inferred from the hard version of the corpus 1. content structure, for example a TOK (Table of Contents), including chapters, topics and/or subtopics, can be considered as a first level of content structure, 2. 'semantic' structure , essentially what a semantic wiki environment supports – refers to the allocation to each term (glossary entry) of a value and properties (is a concept , or is it a relation, either, or both, for example) as well as some lexical definitions for each term (a description of the meaning).
GOVERNANCE A governance model (gm) = 'who' does 'what' (possibly also 'when' and 'how'). characteristics of the technological environments influence and shape directly the governance model. The GM required for a large distributed project would necessarily have to take into account the 'virtual' and 'distributed' nature of the team. A governance model can be built by establishing a) what tasks constitute the editorial process b)what roles (team members and other users) have what permissions (what tasks in the editorial process can carry out). c) what sequence/priorities/rules should be applied to the execution of the tasks that. A workflow can then be built, and tags (metadata) can be derived/associated with the content at different stages of the process.
INTEGRATING DIMENSIONS (D. Bedford) www. dtic. mil/mlw/2004/presentations/Bedford. ppt
The proposed architecture outline
EVALUATION - GOLD STANDARD (Grefstette) Compare and integrate the manual glossary with the ATX glossary - Heuristic evaluation, competence questions - Possible integrated evaluation of various factors, including 'cost' (people hours, cognitive load etc) may need to be developed
Coverage/Scope s the vocabulary capable of representing all of the concepts used in the chapter? Does the vocabulary have the terms necessary to represent the full range of issues? Does the vocabulary encompass the terminology used to describe the various procedures? Does the vocabulary use terms that are commonly used by SE? Specificity s the vocabulary specific enough to accurately represent the many aspects of SE reality? Does the vocabulary capture information in sufficient detail? Structure Are the vocabulary hierarchies logical and complete? Are the meanings of terms clearly defined? Does the vocabulary contain redundant terms? Are there explicit rules for combining terms, or for combining terms and qualifiers? Does the vocabulary allow for multiple classification of terms, that is, can terms appear in more than one hierarchy? Maintenance Does the vocabulary have ongoing institutional support? Does the institution or body that developed the vocabulary have stable funding? Useability s the vocabulary mapped to other vocabularies used in the practice? Does the vocabulary meet the needs of a range of end users? Does the user interface facilitate optimal use of the vocabulary with minimal training?
CONCLUSIONS This paper presents a socio-technical view of the challenges associated with developing and implementing and managing a shared vocabulary for SEBOK, and analyses various aspects of the problem space using innovative worldview and integrating various analysis and knowledge representation techniques. It proposes an integrated 'emergent' architecture to combine vocabulary creation and management, with other aspects of the wikifcation task, including synchronizing the vocabulary development to the structuring of the SEBOK and the governance of the process.
FUTURE WORK Currently the SEBOK team is moving to a wiki, semantic wikification is going to be in a subsequent phase, The proposed architecture is intended to serve as guidance to capture emergence, and will have to be tested, refined and modified according to usage A fully fledged text extraction exercise could result in a 'gazeteer for the systems engineering lexicon' to support text extraction in the SE domain, using open source tools like GATE
Thanks This research is supported by EPRSC Grant EP/D 505461/1. Thanks to University of Strathclyde (Alex Duffy), for funding and academic freedom
Get in touch paola. dimaio@gmail. com
Скачать презентацию Toward a Shared Vocabulary for Systems Engineering Paola
65ad8b699e64a02ce9170f9b95fd536f.ppt