Which semantic web

Citation: Catherine C. Marshall, Frank M. Shipman (2003) Which semantic web. Proceedings of the fourteenth ACM conference on Hypertext and hypermedia (RSS)
DOI (original publisher): 10.1145/900051.900063
Semantic Scholar (metadata): 10.1145/900051.900063
Sci-Hub (fulltext): 10.1145/900051.900063
Internet Archive Scholar (search for fulltext): Which semantic web
Download: http://www.csdl.tamu.edu/~marshall/ht03-sw-4.pdf
Tagged: semantic web (RSS)

Summary

Quote (reformatted):

"The Semantic Web is portrayed as:

1. a universal library, to be readily accessed and used by humans in a variety of information use contexts;
2. the backdrop for the work of computational agents completing sophisticated activities on behalf of their human counterparts; and
3. a method for federating particular knowledge bases and databases to perform anticipated tasks for humans and their agents."

Quote (reformatted/corresponding numbers added):

"The Semantic Web is the outgrowth of many diverse desires and influences, all aimed at making better use of the Web as it stands.

1. The anxiety over the apparent disorder of this new world of digital documents – how one makes sense of new genres, new technologies, and new uses and modes of publishing and organizing materials – is one such influence.
2. from the field of Artificial Intelligence, with its maturing sense of the kinds of computation that can take place given formal representations – what kinds of problems are tractable to the methods that have been developed over the past 30 years.
3. utopian desire to offload the burden of information overload and the complexity of everyday life onto the computer, using the vast resources that have accumulated on the Web as a backdrop to help us in our everyday activities and to address the most normal of problems."

Another set of 3 perspectives:

1. "perceived need to bring order to the loosely connected networks of digital documents that made up the Web"
2. "globally distributed knowledge base" (which can be used by machine agents)
3. "infrastructure for the coordinated sharing of data and knowledge"

Paper analyzes feasibility of and expectations from these perspectives. Places them in two dimensions: what (particular to universal) and who (human to computer). Current web use is particular and human. The three perspectives map to:

1. universal, human
2. universal, machine
3. particular, human

Authors assess

1. taming web: less of a semantic web focus due to success of text indexing and retrieval (ie Google web search).
2. ai: must learn from experience of Cyc and KIF. Explosion of URIs for same concept "not a problem that can be solved by using a uniform knowledge editor, the Semantic Web equivalent to FrontPage" will require heuristics.
3. federated data: much semweb activity is in this cagegory, requires some shared knowledge, social processes cannot be ignored.

Formal knowledge representation considered harmful?

• overhead includes: learning, making decisions about representation
• more effort to express knowledge formally
• (but connecting eixsting formally represented knowledge, eg from databases, may be easier)
• hard to express tacit knowledge formally
• change inevitable, evolving formal schemas costly
• (formal representation of products and services easier than abstract materials)
• knowledge representation requires context, eg people use different names for things in different contexts; within narrow fields agreement may be possible and useful

Practical issues

SemWeb as metadata

• community: "Metadata is not simply a description of the information contained in a work or web page; the choice of a metadata scheme also signifies community membership. Every aspect of metadata – from how it is obtained and verified to the expectations of how it will be used by humans or computer systems – stems from the practices of a particular community"
• cost: "Will any community actually have the financial wherewithal to conform to a jumble of multiple standards?...while we might think that semantic links are possible and desirable, they are certain to be costly to code and verify."
• authority and trust: If metadata relies on community, but SemWeb is universal, trust will be difficult to establish.

SemWeb as markup: HTML ubiquitous, but use dominated by visual design, not structure. How will SemWeb transcend this? "Semantic Web mark-up will be good enough to solve the immediate problem or produce the desired behavior in a limited range of high payoff situations."

SemWeb as its own "killer app": criticizes realism of use case in original SemWeb SciAm article.

Quote:

"Do pragmatic considerations always work against the Semantic Web? No, but scenarios of the complexity of (agent)-like approach to interacting with people and things in the world seem unlikely. Similarly, the “taming the Web” approach violates what we know about the costs and problems with metadata creation. On the other hand, cost-benefit tradeoffs can work in favor of specially-created Semantic Web metadata directed at weaving together sensible well-structured domain-specific information resources; close attention to user/customer needs will drive these federations if they are to be successful."

Existing SemWeb applications: briefly describes scholarly annotation and ecommerce SemWeb efforts, noting how they face challenges described earlier.

Conculsion re 3 perspectives:

1. "more in the realm of Google-like approaches that do not rely on such an abrupt shift in the practices and economies surrounding the Web"
2. "seems out of reach for both theoretic and pragmatic reasons"
3. "What we are left with then is the Semantic Web’s potential in the realm of the particular, especially its ability to weave together specific resources well motivated" (by business or institutional constituencies)

SemWeb future must remember basic theoretical and practical questions:

• "Knowledge stability (How well are the domain and the practices surrounding it understood? How much incremental formalization and restructuring do we expect?)
• Competing conceptual approaches (Is the knowledge intrinsic or extrinsic? Can intrinsic structure be recognized through heuristic approaches, thus avoiding declared representations?)
• Cost/benefit (Who will do the knowledge representation, and to what end? What are competing interests, e.g. other metadata standards?)
• Negotiation among information resource stakeholders (What is the role of negotiation, facilitation, or intervention in representing the knowledge within a socio-technical framework? Are there identified and accepted approaches that work in the domain?"

Theoretical and Practical Relevance

Seems apt 11 years later.