- Information Is Interesting Stuff
- Information and Structure Are Inseparable
- Formal Languages Are Easier to Compute Than Natural Languages
- Generic Markup Makes Natural Languages More Formal
- A Brief History of the Topic Maps Paradigm
- Data and Metadata: The Resource-Centric View
- Subjects and Data: The Subject-Centric View
- Understanding Sophisticated Markup Vocabularies
- The Topic Maps Attitude
The Topic Maps Attitude
The topic maps paradigm is a step along the road to global knowledge interchange. It may well turn out to have been quite a significant step. Nonetheless, it is very obviously not the last step. If it successfully moves our species forward toward global knowledge interchange, the topic maps paradigm will owe much of its success to the fact that it is resolutely responsive to current technological, economic, and anthropological conditions, and just as resolutely responsive to certain philosophical values and attitudes. Some of these values came from the comparatively young traditions of the markup languages community.18 Other values are derived from much older traditions. What follows is a summary of the values and perspectives that I find most remarkable.
We must recognize that civilization is what makes it possible for us to have breakfast every morning, and civilization's increasing ability to develop and exploit information resources is generally correlated with the richness and quality of life available to each human individual living on our planet. Global knowledge interchange is important to every single living human being.
We must cherish diversity by giving diverse worldviews the ability to be expressed and exploited alongside and in federated combination with all other worldviews. This includes respecting communities of interest, encouraging their formation, and not coincidentally causing them to provide themselves with usable interfaces for use by other communities of interest.
We must understand that worldviews provide essential contexts for communication and that communication rests on our intuitive ability to cross the chasm between symbolic expressions and reality. We must work to provide computers with increasing sensitivity to (that is, apparent awareness of and ability to act upon) diverse human contexts.
We must accept partial solutions and partial expressions, demanding neither comprehensiveness nor perfection. There never will be any such thing as a "complete" topic map, or one true ontology suitable for all contexts, or a holy grail of "knowledge." A single human being or organization can accomplish something only within some limited scope. Providing a way for incomplete, imperfect utterances to contribute, in some useful way, to the ongoing intellectual life of the human species is essential.
We must understand and adapt to the fact that different subjects of conversation have different kinds of reality, for example, an information asset is real in one sense, the Statue of Liberty in another, shoe-ness in a third, and Minnie Mouse's high-heeled shoes in a fourth. At the same time, we must understand and exploit the fact that all subjects are, in some sense, the same, in that we humans seem to find them worthwhile to discuss.
We must provide a way for ordinary people to quickly and easily gain a superficial understanding of global knowledge interchangea way that does not compromise a deeper level of abstract simplicity and power.
We must abandon "simplifying assumptions" that actually interfere with our ability to manage and maintain our increasingly complex civilization (for example, the resource-centric view of metadata and the idea that the interchange structure of information should always be the same as the structure of the information itself).
We must provide technology that is suitable as a foundation for business models that, in the aggregate, make many significant contributions to global knowledge interchange and the general availability of knowledge.
We must recognize infoglut as the single most formidable remaining enemy of global knowledge interchange in a world where the connectivity problem is already well on the way to being permanently solved.
We must recognize that subjects of conversation are the true axis points of information, even though they are not addressable by computers. Creating addressable information resources to represent nonaddressable subjects allows the addressable resources to be used as public "hooks," called published subject indicators (see Chapter 5), on which topic relationships, names, and relevant information can be "hung."
We must acknowledge that generic markup is the most natural and most economically conservative way to interchange and archive valuable information assets whose future exploitability cannot be completely predicted (that is, practically all information assets).
We must accept that markup (whether generic or procedural) will always be too rigid or otherwise inadequate for all applications. Thus we must support the ability to impose arbitrary structure on arbitrary information by means of external, independently maintained metadata.
We must understand the need for markup and other metadata to be described, even as they themselves describe other data.
We must recognize that the federation of knowledge assets is an ongoing activity that must account for the evolution of the knowledge assets to be federated, without losing the value of investments in previous federating activities.