The Impact of ICT on Information Intermediaries and Their Functions

Walk into any modern library or information centre and you will notice that the work of an information intermediary looks almost nothing like it did three decades ago. The card catalogue has given way to databases, the reference desk now competes with search engines, and a single query can pull resources from servers located thousands of kilometres away. Behind all of this sits one driving force: Information and Communication Technology (ICT). ICT has reshaped how intermediaries find, organise, and deliver information, and in many cases it has handed parts of their traditional job to software. This post examines how ICT has transformed the role and functions of information intermediaries, and what new tools now sit alongside them in the information chain.

Who are information intermediaries and why ICT matters 🔗

An information intermediary is anyone or anything that stands between an information source and the person who needs it. Librarians, documentation officers, reference staff, and information consultants are the classic examples. Their job is to bridge the gap between a vast, scattered universe of recorded knowledge and a specific user with a specific need.

For most of the twentieth century this bridging was a manual craft. The intermediary memorised collections, maintained physical indexes, and personally guided users to the right shelf or volume. ICT changed the foundation of that craft. The benefits of ICT in information services are often summarised as the “4 Es” – economy, ease, expansion, and efficiency. Each of these directly affects what an intermediary can offer, how quickly, and at what cost.

ICT’s role in information management 🔗

The most visible change ICT brought is in how information is retrieved and disseminated. Conventional services such as the catalogue, reference work, bibliographic services, current awareness services, document delivery, and inter-library loan can now all be delivered electronically. This shift has changed information services in their format, content, and method of production and delivery, making the whole process faster and more convenient for the end user.

From custodian to facilitator 🔗

Perhaps the single biggest effect of ICT is a change in the intermediary’s identity. The rise of the internet as the largest repository of information shifted the role of LIS professionals from intermediaries to facilitators. In the older model, the intermediary held the information and parcelled it out. In the new model, the resources are already accessible to the user; the intermediary’s value lies in helping the user navigate, evaluate, and make sense of an overwhelming volume of choices.

This is not a minor relabelling. A facilitator teaches users to fish rather than handing them a single fish. Information literacy training, guidance on evaluating sources, and help with search strategy have become central functions, precisely because the raw material is now abundant rather than scarce.

The shift from physical to virtual 🔗

ICT also dissolved the walls of the information centre. Resources are no longer confined to a building but are integrated into local, regional, national, and international networks. A user in a small town can access journals subscribed to by a consortium, and an intermediary can deliver a document electronically within minutes. Round-the-clock and remote access mean the intermediary’s reach is no longer limited by opening hours or geography.

The rise of automated systems 🔗

As information volumes exploded, software began to take over functions that once required a human intermediary. These automated systems can be thought of as new-age information intermediaries. They do not replace the professional entirely, but they handle the heavy lifting of filtering and matching at a scale no person could manage.

Information filtering software 🔗

Filtering software addresses a basic problem of the digital age: most of what is available is irrelevant to any given user. Information sources on the internet carry high noise, meaning most pages are irrelevant to a particular user’s interests. Filtering systems work to reduce this noise. They typically compare a stream of incoming information against a stored profile of what the user wants, passing through only the items that match. This is the same logic that powers a current awareness service, but automated and continuous.

Expert systems 🔗

An expert system is software that captures the knowledge and decision rules of a human specialist and applies them to solve problems in a defined domain. In an information setting, an expert system can mimic some of the reasoning a reference librarian uses – asking clarifying questions, narrowing down a subject, and pointing the user towards the most suitable resources. The strength of an expert system is consistency: it applies the same expert logic to every query, at any time, without fatigue.

Intelligent agents 🔗

Intelligent agents go a step further than simple filters. An intelligent agent is a software-based system with properties such as autonomy, social ability, reactivity, and pro-activeness. In plain terms, an agent can act on its own behalf, communicate with other systems, respond to its environment, and even anticipate needs. Where a traditional search engine waits passively for a query and returns whatever its database holds, an agent can adapt its strategy to the individual user.

This adaptability matters because searching for useful information is difficult precisely because of information overload across heterogeneous sources. Agents tackle this by building and refining a model of the user. By understanding context and user intent, intelligent agents can deliver more relevant and personalised results than simple keyword matching, filtering out irrelevant noise and saving the end user considerable time.

The internet as an information hub 🔗

The internet is the environment in which all these tools operate, and two ICT developments in particular have become essential to the intermediary’s work: search engines and subject gateways.

Search engines 🔗

Search engines are the most familiar information retrieval tool on the internet. They crawl, index, and rank billions of pages, allowing anyone to retrieve documents in seconds. For an intermediary, a search engine is both a powerful ally and a source of difficulty. It provides instant access to a colossal pool of material, but it also returns large quantities of unevaluated, mixed-quality results. The traditional retrieval model of search engines operating on huge databases struggles because a user often cannot accurately express what they want, and the engine does not tailor its strategy to different users. This gap is exactly where the trained intermediary still adds value, by crafting precise queries and judging which results deserve trust.

Subject gateways 🔗

Subject gateways were developed to solve the quality problem that general search engines create. Subject gateways are online services and sites that provide searchable and browsable catalogues of internet-based resources, typically focused on a related set of academic subject areas. Rather than indexing the entire web, a gateway covers a defined field and includes only resources that have passed quality checks.

What sets gateways apart is the human expertise built into them. Subject gateways are typically databases of detailed metadata records that describe internet resources and offer hyperlinks to them, allowing users to search by keyword or browse under subject headings. Each entry is selected against explicit criteria, classified using schemes such as DDC, and described using standards such as Dublin Core. In effect, a subject gateway is the work of an intermediary frozen into a reusable tool – a subject specialist and an information specialist combine their judgement so that every later user benefits.

India has its own example in this space. The LISgateway, launched in April 2004 as a project of the University Grants Commission, was the country’s first gateway in the field of library and information science, created to provide a trusted source of selected, high-quality internet information.

Collaborative filtering and personalisation 🔗

The final piece of the picture is personalisation. A good intermediary always tried to tailor recommendations to the individual user. ICT now does this automatically, often through a technique called collaborative filtering.

How collaborative filtering works 🔗

Collaborative filtering is the process of filtering or evaluating items through the opinions of other people, bringing together the preferences of large interconnected communities to filter substantial quantities of data. The core assumption is simple: users who agreed in the past are likely to agree in the future. If two readers have rated the same set of articles highly, then an item liked by one is a strong candidate to recommend to the other.

This technique is everywhere in daily digital life. Real-world examples include Google News recommendations, music recommendations on Last.fm, and suggested follows on social platforms, showing how domain-independent the method is across books, music, movies, and more. In an information service, the same approach can surface a journal article or database that similar users have found valuable, even if the current user never thought to ask for it.

Benefits and limits for the intermediary 🔗

Collaborative filtering helps intermediaries customise information at a scale that manual recommendation never could. Because it only needs user behaviour data, it does not require domain knowledge and can be applied easily across many sectors. As more interaction data accumulates, the recommendations improve.

The technique is not flawless, though, and understanding its limits is part of an intermediary’s modern skill set. A new user who enters the system has no historical data, so the algorithm cannot offer personalised recommendations – a difficulty known as the cold start problem. Data sparsity, where most items have very few ratings, also weakens the results. This is where the human intermediary remains valuable, stepping in to guide users whom the algorithm cannot yet serve well, and combining automated suggestions with professional judgement.

The intermediary’s evolving skill set 🔗

Pulling these threads together, ICT has not made the information intermediary obsolete. It has redefined the role. The intermediary is now expected to manage digital collections, operate and evaluate automated systems, guide users through search engines and gateways, and interpret the output of recommendation tools. ICT has shifted the focus of professionals from managing physical collections to digital information management, demanding new technical and soft skills alike.

The constant across all this change is judgement. Software can filter, match, and recommend, but it cannot fully replace the contextual understanding, ethical care, and critical evaluation a trained professional brings. The most effective information services today are those where human intermediaries and automated systems work as partners rather than rivals.

What do you think? If automated systems can filter and recommend information so effectively, what unique value should a human information intermediary focus on developing? And as recommendation tools shape what we see, how should an intermediary guard against users only ever encountering information that confirms what they already believe?

References

1. https://www.academia.edu/26930224/Role_and_Importance_of_ICT_for_Library_and_Information_Services
2. https://granthaalayahpublication.org/journals/granthaalayah/article/view/IJRG16_C09_114
3. https://rsisinternational.org/journals/ijriss/Digital-Library/volume-5-issue-9/198-212.pdf
4. https://www.ijlsit.org/html-article/23423
5. https://scialert.net/fulltext/?doi=itj.2007.117.122
6. https://link.springer.com/chapter/10.1007/978-3-642-01665-3_16
7. https://smythos.com/developers/agent-development/intelligent-agents-in-information-retrieval/
8. https://www.sciencedirect.com/science/article/abs/pii/S1057231705000676
9. https://files.eric.ed.gov/fulltext/ED441486.pdf
10. https://link.springer.com/chapter/10.1007/978-3-540-72079-9_9
11. https://www.sciencedirect.com/topics/computer-science/collaborative-filtering
12. https://builtin.com/data-science/collaborative-filtering-recommender-system
13. https://www.lisedunetwork.com/impact-of-ict-in-library-and-society/