One thing that stood out to me in the reading and lectures from week 4, 5 and 6 of the Data Information and Technology Applications module was the Semantic Web and its benefits for the cataloguing and accessing of relevant but potentially disparate pieces of information on the Web. I was intrigued by its practical applications, but could it ever become a practical reality?
The idea of the Semantic Web was proposed by Tim Berners-Lee, James Hendler and Ora Lassila with the goal that information on the Web would be encoded so that its meaning and its relationships to other bits of information could be “understood” by machines, and in doing so “bring structure to the meaningful content of web pages” (Berners-Lee, Lassila, Hendler, 2001) .
This “understanding” and structure is achieved through the use of Linked Data (Berners-Lee, 2006), put simply, a set of rules and best practices for encoding information on the Web, so that what it is and how it relates to other bits of information is defined and interlinked using a standardised, machine and human-readable metadata schema. In this case, Resource Description Framework (RDF) the specifics of which I won’t go into, but there is a good summary of it here (Shotton, 2013).
If used appropriately, it means that a program could connect related bits of information from across the Web, making it more easily indexable, searchable and accessible rather than siloed into disparate information pits. Google does this already using Semantic Web technologies to generate useful conceptual summaries next to search results, but Tim Berners-Lee takes its application further in his TED talk where the use of Linked Data would allow people to more effectively access data and information on Alzhemier’s research, ask more complex questions of a search engine and retrieve better search results (TED, 2009).
Many libraries are also putting stock into the Semantic Web and recognising it as an opportunity to update metadata formats to ones that are more interoperable with the Web, to convert increasingly outdated MARC records into a Linked Data format, so that library resources can be more easily indexed and accessed online (Hallo, Luján-Mora, Trujillo, 2014).
Beyond being indexable by search engines, and the benefit of being linked to other related sources of information this would also allow library records to be accessed with more granularity, as well-defined bits of information in their own right, rather than just a link to a catalogue (OCLCVideo, 2012).
So far, the benefits of the Semantic Web seem clear, a universal cataloging system that is machine readable where the connections between concepts are informed by descriptive, explicit data relationships so people can get the information they need from as many linked, authoritative sources as possible. But this version of the Web is yet to be realised and remains for many “A utopian dream” (Target, 2018), and this comes down to the difficulty of getting people to adopt one standard of metadata annotation, or any standard at all. This issue persists, not just on the Web, but in libraries as well, as Park and Kipp noted “Libraries worldwide are not using the same elements or even the same ontologies, schemas and data models to describe the same materials using the same general concepts” (Park, Kipp, 2019).
Furthermore, there is the issue of being able to trust the metadata used in Linked Data to be accurate. For example, Google now disregards Meta tags and Meta descriptions in their search algorithm because people would stuff them with irrelevant words and phrases. As well as this, although Linked Data is designed to be “understood” by machines, it would still be written by humans and therefore the data and connections being made are susceptible to bias.
Therefore, although the Semantic Web would be hugely beneficial not just for finding information but for also as a new way for cataloging library information on the Web, I think it is unlikely that it will be used as robustly as originally envisioned, due to human bias or apathy, inconsistencies with metadata and categorisation and as Floridi argues, current technological inefficiencies (Floridi, 2014).
Berners-Lee, T. (2006) Linked Data. Available at: https://www.w3.org/DesignIssues/LinkedData.html [Accessed: 28 Oct 2020]
Berners-Lee, T., Hendler, J. and Lassila, O. (2001) ‘The Semantic Web’, Scientific American, vol. 284, no. 5, 2001, pp. 34–43.
Floridi, Luciano (2014) The 4th revolution: how the infosphere is reshaping human reality. Oxford University Press
Hallo, M., Lujan Mora, S. and Trujillo, J. (2014) Transforming library catalogs into Linked Data. Available at https://rua.ua.es/dspace/bitstream/10045/50586/1/transforming-library-catalogs-linked-data.pdf [Accessed: 28 Oct 2020]
OCLCVideo (2012) Linked Data for Libraries, 9 August. Available at: https://www.youtube.com/watch?v=fWfEYcnk8Z8 [Accessed: 28 Oct 2020]
Park, H. and Kipp, M (2019) ‘Library Linked Data Models: Library Data in the Semantic Web’, Cataloging & Classification Quarterly, Vol. 57, Issue 5, pp. 261-277
Shotton, D. (2013) ‘Libraries and linked data #1: What are linked data?’, Semantic Publishing, [blog] 1 March. Available at: https://semanticpublishing.wordpress.com/2013/03/01/lld1-what-are-linked-data/ [Accessed: 28 Oct 2020]
Target, S. (2018) ‘Whatever Happened to the Semantic Web?’, Two-Bit History, [blog] 27 May. Available at https://twobithistory.org/2018/05/27/semantic-web.html [Accessed: 28 Oct 2020]
TED (2009) Tim Berners-Lee: The next Web of open, linked data. 13 March. Available at: https://www.youtube.com/watch?v=OM6XIICm_qo [Accessed: 28 Oct 2020]