OntologWiki: OntologySummit2010 PresentRequirements Synthesis

OntologySummit2010: Present "Requirements" - Synthesis of the Discussion (25I6)

This is the workspace for the co-champions to synthesize the discussion on this track. (25I7)

Track Label: Requirements - Subtrack Label: Present (25I8)

Mission: To survey the existing roles filled and activites performed by working ontologists, along with the skills required and where they typically get them. (25IA)

Pertinent questions: (25IB)

Where do ontologists currently work? (25IC)
What organization types (companies, departments, institutions, agencies, etc.)? (25ID)
What project types (long/short term, part of larger system/program, focus on IT|KM|search|interop|metadata|other)? (25IE)
What fields or industries? (25IF)
What folks are doing KOS work that is within or heading toward ontology, but may not be labeled as such? (25IG)
What tasks do working ontologists perform and as roughly how much of their time (requirements gathering, elicitation, informal modeling, formal modeling, formal testing, user testing, "gold standard" and/or regression testing)? (25IH)
What skills do working ontologists typically acquire on the job, and what via formal training? (25II)
What training do they wish they had, but don't have? (25IJ)
What skills do employers expect? (25IK)
What formal training do employers expects? (25IL)
How do employers select ontologists when hiring? (25IM)

Synthesis: (25IN)

Requirements Panel (2BHM)

Requirements Survey (2BHN)

The 2010 Ontology Summit community is in general agreement that the most valuable and urgently needed training will be informed not only by theoretical considerations but also, and centrally, by the needs of ontologists seeking employment and employers seeking quality, useful ontologists. Our understanding of those needs, however, has been scattered and divergent. Several discussions, a Panel session, and a survey were undertaken with the aim of improving this understanding. (2BHW)

We wanted to developing a richer and more clear picture of the requirements of ontologist employability (that is, those trained should well-prepared for the available jobs) and deployability (that is, those who hire trained ontologists find them ready and able to perform the needed work). To that end, we aimed to bring a strong "end-user" voice to the conversation. Each of our panelists embodied one or more end-user perspectives. Most had some combination of experience as a working ontologist, an employer, a manager, an evaluator or on-the-job trainer of ontologists, and a gatherer of lessons learned from full life-cycle ontology projects. (2BHX)

The Requirements Survey was developed, in part, from the results of the discussions and Panel contributions. In addition to basic respondent information, the survey included two main types of questions. One set of questions elicited information about the respondents relationship to ontology, for example: as a working ontologist, as a technical person on a border between ontology and another field, as a potential future ontologist, as an employer, manager, or evaluator of ontologist, as someone otherwise dependent on ontology work and its outcomes. Another set of questions elicited information about the nature of ontology work from the respondent's perspective: skills and knowledge needed, depth of training needed, types of work included, prioritization of necessary knowledge and skills. ...how distributed... (2BHY)

Sixty-four individuals responded to the survey. Of these sixty-four, forty-one described themselves as very familiar with ontology and seventeen said they were somewhat familiar. Most respondents came from organizations in the Education (17), Government(12), Scientific and Technical Research and Development(10) and Information Technology (10) sectors, with addition representation in Cross-sector consulting (5), Manufacturing (3), and others (8).Almost half of respondents identified their primary function as Research (31). The next most selected functions were Teaching and Training (16), Technical Lead (development) (12), Project Management (9), Enterprise Level Management (8) and Knowledge Management (8). Fifteen other responses, including "other," were represented (45, combined). It should be noted that most questions in this survey permitted multiple answers, so respondents may be in more than one response category for these and most other questions. (2BHZ)

Most respondents were in some way directly involved in the ontology work, as developers (29), designers (31), evaluators (28), researchers (41), or past participants in these activities (7). Future ontologists were also well-represented, with eighteen respondents expressing the hope or expectation of future ontology development or design work. Three expressed uncertainty over whether the work they did made them ontologists or not, and ten did not personally do ontology work at all. The questions concerning areas of training, experience, skill, and knowledge were presented with a fairly long list of potential answers and one or more open-ended options. It was clear from the preceding Summit discussions and panel input that it would be infeasible to include all areas suggested by participants. It was important, however, to include at least a sampling across these areas, such that respondents would not be primed to confirm or disconfirm some particular prioritization over others (or at least, that the skewing potential of such priming suggestion would be minimized). To that end, we made a point of including choices that were both: research-oriented and applied; domain-specific and not domain specific; application-centric and ontology-centric; machine-centric and human-centric; methodology-centric and tool-centric; theoretical and hands-on; inside and outside of the most restrictive views of what constitutes ontology. We did, however, draw one boundary explicitly: that between the philosophical field of ontology and the field of applied, formal ontology as encountered in information and computer sciences. (2BI0)

A broad range of requirements were identified by significant numbers of participants, including: logic and formal semantics, working with subject matter experts, knowledge representation, Ontology mapping, Extracting knowledge from existing corpora, Ontology lifecycle, conceptual modeling, and others. However, the ranking of these requirements varied significantly across respondent characteristics. ...table... (2BMA)

Combined Summary (2BHO)

coming... (25IO)

Draft of Requirements Inputs to the Communique (2BHP)

The 2010 Ontology Summit participants were in general agreement that the most valuable and urgently needed training will be informed not only by theoretical considerations but also, and centrally, by the needs of ontologists seeking employment and employers seeking quality, useful ontologists. Community understanding of those needs, however, has been scattered and divergent. For this reason, the Summit incorporated discussions, a panel session, and a survey specifically aimed at developing a richer and clearer picture of current requirements for ontologist employability and deployability. We asked: what are the requirements for training programs that would produce ontologists well-prepared for the available jobs? What is needed if those who hire trained ontologists are to find them ready and able to perform the needed work? (2BRS)

Among the many interesting results produced by the survey, the following are of greatest significance for training and curriculum developers: (2BRU)

There is a large gap between training needs and training availability, in both quantity and type. Formal training opportunities are still scarce; most respondents have formal training in other fields and on-the-job or self-directed training in ontology. Furthermore, the Current Training survey found that most training opportunities exist within academic degree programs (for example, ontology courses within Computer Science curricula. However, there is substantial demand for training outside of such contexts, including single courses, professional certification programs, hands-on training with or without certification, and short familiarization courses as short as one week. This demand is perhaps best interpreted in conjunction with the fact that the majority of respondents expressing an interest in training already have some level of familiarity and indicate that their interest in ontology grows from their current work in one way or another. (2BRV)

Some areas included in current formal courses is also identified as a requirement by potential employers, evaluators, working ontologists, and potential trainees. Logic and formal semantics top this list. These are not, however, typically included in shorter-duration training offered to working professionals. There is therefore a significant need for a sort of training not generally available: substantial training, delivering not just familiarity but also technical competence, offered in some form accessible to those not pursuing a degree (beyond what they may already have). (2BRW)

Some areas largely absent from existing curricula are consistently identified as important by experienced, working ontologists, potential trainees, potential employers, professional collaborators, and senior professionals who manage, lead, evaluate and depend upon ontologists. High on this list is training and/or experience working with Subject Matter Experts, Ontology Lifecycles and components thereof (including testing, evaluation, and refactoring), Knowledge elicitation and extraction from document corpora, and applied methodologies such as ontology mapping and knowledge representation. (2BRX)

There is significant demand for training even among those already familiar with the field. (2BRY)

The Very Familiar possess and consider useful a considerable variety of background training an knowledge. That is, people who came with such backgrounds did not see ontology work, as unrelated to this background. This is an important measure of contributions from those fields. It contrasts, for example, with the lower assessment of relevance one sees in traditional career switchers the lawyer turned caterer, for example, or miner turned technician. This result supports the view that the field, as it now exists, is fundamentally interdisciplinary. It also suggests that there is much to be gained by designing curricula that incorporate these contributions and welcome students from many backgrounds, in contrast to focusing more exclusively on just a few feeder fields or departments. (2BS0)

The Requirements Survey produced some significant illumination of the gaps between current ontologist requirements and the areas in which current training delivers adequate preparation. That contrasts and surprises emerged underscores the current lack of information flow or understanding between those who need training and those who produce it. Given the usefulness of this understanding if the training and need are to be better aligned, it would be useful for the survey exercise to be refined and developed into some long-running activity or mechanism for gathering information about the activities, skills, knowledge, and expectations on working ontologists. This technically rough study can provide a jump start, and lessons learned, for the development of a more sophisticated elicitation mechanism, more amenable, especially, to statistical analysis and discovery of clusters. As the field of ontology continues to evolve and demand shifts with it, it would also be useful to curriculum designers and providers to have their picture of training needs regularly update. We hope that one or more organizations with an interest in the quality and availability of ontology training will therefore support development of such a long-running survey or other information-gathering mechanism. (2BRZ)

 This page is maintained by AmandaVizedom
 Please do not edit or modify it yourself; send any editing request to any one of the individuals named above.    (25IP)

OntologySummit2010 PresentRequirements Synthesis