The Whiteboard

Yes. Check out http://oeg-lia3.dia.fi.upm.es/oops/. The authors looked through the literature and discovered 29 common pitfalls that arise when building ontologies. Code has been written to automatically find 21 of the 29 pitfalls. Try it out and give the authors feedback. The whole community will benefit.

Workers, professionals and managers are all becoming acutely aware of the discrepancy between the performance of consumer based technology and their internal IT department. The gap between the information and services they can get for ‘free’ on the web vs. what they get from IT is creating a high level of frustration and dissatisfaction between IT and its constituents. Managers are pushing for easier to use IT solutions, service delivery to next generation mobile devices and better and more. These requests seem reasonable in light of the fact that many of these services are already available to the consumer. Unfortunately, the entire enterprise IT structure and process is set up to make delivering on these demands very difficult. Generally, enterprise IT departments are centralized, highly controlled and operate under a top-down hierarchy. Technology decisions are made at high levels and then pushed down the organization in slow and unpredictable ways. The slow pace of adoption, traditional project management process, endless systems integration problems, IT backlogs, security concerns, and other issues work against even the best intentions to improve IT’s performance. Semantic technologies offer the IT a unique opportunity to solve many of these problems by preserving investments in legacy data, applications and infrastructure and providing better, easier to use services that are delivery-independent. Thanks to a semantic approach, an enterprise can build a development roadmap that enables them to both preserve the legacy systems that are still working efficiently and provide a higher level of data and systems integration. Because a semantic or messaging systems sits on top of the legacy applications and uses the data independently of the system it facilitates system integration between legacy, cloud and new applications without having to recreate the entire system. The following section discusses how Semantic Technology can help IT organizations improve IT productivity by:

• Reducing complexity
• Improving productivity
• Reducing Total Cost of Ownership.

gist is a minimalist upper ontology designed to aid the production of business oriented ontologies. It does not stand for anything; it is a word, meaning “a general understanding.” gist is freely available at www.semanticarts.com/gist and is licensed through the Creative Commons Share Alike license. It has been under development by Semantic Arts for over six years. It consists of just under 100 classes and just under 100 properties. One design goal was completeness: it is meant to cover nearly all the concepts that come up in real world enterprise ontology development. In our last several ontology projects we have found that fewer than 5% of the concepts uncovered are not directly expressible using the gist primitives. Another goal was to promote disambiguation. We have avoided highly abstract concepts, as they tend to be ambiguous, and we have gradually culled out ambiguous business terms. gist is very highly axiomized, and as such when used it is highly self-correcting.

Semantic technologies will create incremental value by making us more perceptive, intelligent and collaborative learners.

• The knowledge modeling capabilities of semantic technology will encourage us to create abstract models of our ideas, run simulations to evaluate and improve them and then implement the new models faster. This will enable us to solve problems faster, implement new ideas more effectively and innovate more often. We will become more perceptive in anticipating and responding to change and managing it successfully.
• The intelligence building properties of a semantic model will help us become smarter. Intelligent agents driven by semantic models will help us search, find and analyze vast information resources, including structured and unstructured data in private and public data repositories. They generate relevant and timely information that we can use to solve complex problems, communicate more effectively and create novel solutions.
• Semantic technologies facilitate a computer’s ability to learn from past cases, new information and user interaction. The more a semantic model is used, the smarter it gets. The smarter it gets, the more people want to use it. It becomes a self-improving system encouraging collaboration and learning and helping us become smarter over time.
• Semantic technologies help us create symbiotic relationships with everyone who shares the system. It realizes the value of social networking by making it easy to share information and build on each other’s knowledge. It helps us evolve towards more collaborative and democratic use of resources and power.

In conclusion, while semantic technologies initially seem complex, they are actually a way to help us dramatically simplify information systems. Since they help us reorganize our concepts into organized models that a many computer applications can access, they provide us with bridges between disparate systems. This empowers us to tackle new problems, create innovative services and make better informed decisions in ways that have never been possible before.

Semantic technology makes us more effective, efficient and strategic. Effectiveness. A system that can easily be modeled to reflect the way you think or the processes you manage, increases your effectiveness.

• You can solve problems faster because you can quickly apply your problem solving process to a greater range of structured and unstructured data and you can analyze information in ways that has not be predetermined.
• You can make better decisions because you have access to more and better data, and ‘what-if’ scenarios can be spontaneously applied to help you see the potential impact of your choices.
• You can plan more productively because semantic models facilitate the ability to find new ways to organize, track and report on data without having to rely on programmers to set up new systems.

Efficiency. Semantic technology increases efficiency because it increases end users’ control over information, so a job can be done faster with fewer resources.

• Intelligent access to detailed information that is constantly changing helps you solve problems faster, so you can minimize the negative impact of the problem.
• Flexible models run by intelligent algorithms enable a wider range of people to make consistent decisions in real time.
• More finite and flexible control over key business processes, such as inventory management or customer service, improves organizational planning and operating efficiency.

Strategic advantage. Semantic technology facilitates change and encourages innovation because it enables the people responsible for the success of the organization to easily use information resources to ask questions, project scenarios and invent new solutions.

• Since it is possible to extend the relationships identified in a semantic model without impacting its integrity, cross discipline collaboration is encouraged, which encourages people to tackle complex, multi-dimensional problems.
• Semantic modeling enables us to move from idea to implementation much more quickly than traditional IT technology, increasing our organizational agility and ability to quickly respond to change.
• The flexibility inherent in semantic models invite us to imagine the future. Since it is easy to create new relationships, we are encouraged to challenge the status quo and ask cross-disciplinary questions.

They can be the same software. That isn’t the essential difference. There are many exceptions and edge cases but the central difference in moving from on premise to SaaS is that you are moving from a large capital budget item that takes a long time to approve to the steady drip of transaction or usage fees from here on out, which can be funded from the operating budget.

It mostly depends on whether you are the cloudor or the cloudee. If you are consuming cloud services theoretically there isn’t much difference, but practically there is. You will likely be paying more for a private cloud (not so much because it is inherently more expensive, but because there isn’t any competition) and consuming the private cloud will generally be more secure. If you are the cloudor (the provider of the cloud service) it is working a bit against you . You are forced to have excess capacity that you must pay for in order to provide for the potential surge.

A good ontology simplifies your information management systems because it creates an enterprise-wide definition of what things are called and their interrelationships. In our experience, even the complex requirements of multi-billion, global companies can be organized into ontologies of about 1000 concepts. Because ontologies reuse properties they can represent the real world complexities of a business using many fewer properties than logical models would need.

The world of traditional information technology is black or white. If something isn’t the same, then it is different. Every new distinction requires the creation of a new table. This creates a problem because once a new table is created the concept is considered new and unique from every other concept, causing redundancy and confusion. In the semantic world, shades of grey are tolerated. Once you formally define a concept, the semantic model creates a ‘web of similarity’ enabling the inference engine to associate the new concept to other classes that are similar. Queries are able to search not only for explicit information defined in the black and white world, but also for the ‘grey’ information found in the inferred subclasses. This enables users to ask much simpler questions and get much more complete information in return.