any material related to a topic or issue (here: SSI): facts, claims, data, descriptions, arguments, interpretations, quotations, statistics, images, media etc.
\n→ misinformation, disinformation, maliinformation
\nspecific unit of information that someone encounters in a given situation
→ appears in the form of a genre (e.g., an article, a post, a video, a report, a tweet, an interview)
→ in an information environment (e.g., a website, a social media platform, a conversation, a newspaper)
date when an information item was originally published, posted, or broadcast
→ indicates how current the content is and whether it may have been overtaken by newer findings, events, or developments
what is actually said, shown, claimed, or explained in an information item
\n→ fabricated content, manipulated content
\ncommunicative meaning that results from how an author or publisher selects, emphasises, frames, and structures the content → viewpoints / positions
\ncompact visual representation of an information item, typically combining title, snippet, domain/URL, and sometimes a thumbnail — used in feeds, search results, and when links are shared on social media or in messaging apps
\nthe brief descriptive text component within a preview — typically a sentence or two excerpted or generated from the content of the underlying information item
\ninformation item or source that is cited, quoted, linked, or referred to within another information item
→ makes it possible to trace information back towards its origin
→ following references is a key strategy for evaluating reliability
origin of an information item — used in two senses:
\nas actor:
\nas item:
\n{{creator}}person or organisation that creates the original content of an information item
→ may be identified, pseudonymous, or anonymous
The degree of mediation does show is how far the information item is from the original material — and therefore how many steps of selection, interpretation, and potential distortion lie in between.
\nThe degree of mediation does not determine reliability or trustworthiness.
→ A primary source can be biased, incomplete, or wrong (e.g., an eyewitness account, a misleading dataset).
→ A secondary source can be more reliable than the primary material it reports on (e.g., a fact-checked article correcting a politician's claims).
| Degree of Mediation | \nEpistemic Proximity / Relation to Original Information Item | \nExamples | \n
|---|---|---|
Primary Source | \nDirectly presents, documents, analyses, or interprets original information: raw data, original events, or firsthand experience | \nWHO research report, eyewitness account, original dataset, parliamentary speech | \n
Secondary Source | \nReferences and reports on, analyses, interprets, comments on, or reframes one or more primary information items | \nNewspaper article about the WHO report, textbook chapter summarising research findings, social media post quoting from a scientific study and commenting on the quote | \n
Tertiary Source | \nReferences, compiles, or summarises multiple primary and/or secondary information items, typically without presenting new original information | \nTikTok video reacting to a tweet that commented on a scientific study, Wikipedia entry, encyclopedia overview | \n
| Source Type What kind of source is this? | \nAuthors / Creators Who is typically behind it? | \nPrimary Interest What is their main purpose? | \nCommunicative Intentions How do they typically communicate? | \nPre-Publication Review How carefully is the information checked before publication? | \nImplications for Reliability and Trustworthiness How does this affect reliability and trustworthiness? | \nGenres What formats do information items in this source type typically take? | \n
|---|---|---|---|---|---|---|
| Academic / Educational | \n- Universities - Research institutions - Schools - Scholarly publishers | \n- Knowledge production - Education | \n- Informing - Explaining - Documenting evidence | \nOften formally reviewed, though the level of review varies. | \nOften stronger in evidence use and documentation, but findings may be preliminary, narrowly scoped, or superseded by newer research. | \n- Journal article - Textbook - Lecture recording - Encyclopedia entry | \n
| Commercial / Promotional | \n- Companies - Brands - Industry associations - Sponsored creators | \n- Profit - Market position - Reputation | \n- Promoting - Selling - Building brand trust | \nUsually internally reviewed for brand, legal, or marketing purposes. | \nProfessionally produced and shaped by commercial interests; marketing and information are often blurred. | \n- Sponsored post - Brand blog - Product microsite - Corporate press release | \n
| Individual / Personal | \n- Private persons - Unaffiliated creators - Personal networks | \n- Self-expression - Sharing - Opinion - Experience - … | \n- Expressing - Sharing - Commenting - Entertaining | \nUsually not formally reviewed before publication. | \nCan offer firsthand perspectives, but reliability, expertise, and accountability vary widely. | \n- Personal blog - Forum comment - Private message - Personal video | \n
| Journalistic | \n- News organisations - Editorial teams - Independent journalists | \n- Public reporting - Audience reach | \n- Reporting - Interpreting - Investigating - Commenting | \nOften editorially reviewed, though standards vary. | \nCan provide verified and contextualised reporting, but quality varies by outlet, genre, speed, and commercial pressure. | \n- News article - Feature - Opinion piece - Editorial - Investigative podcast | \n
| Philanthropic / Service-oriented | \n- Humanitarian organisations - Charitable foundations - Service-oriented NGOs | \n- Public benefit - Aid - Welfare | \n- Informing - Supporting - Coordinating aid - Raising awareness | \nOften internally reviewed, though standards vary. | \nOften mission-driven and grounded in practical experience, but scope, independence, and evidence use vary. | \n- Aid programme page - Foundation project report - Charity newsletter | \n
| Political / Advocacy | \n- Parties - Campaigns - Advocacy groups - Activists - Lobbying groups | \n- Advocacy - Persuasion - Mobilisation | \n- Persuading - Pressuring - Agenda-setting - Mobilising | \nUsually strategically reviewed to support a position or campaign. | \nImportant for public debate, but often shaped by advocacy, persuasion, or institutional interests. | \n- Party programme - Campaign ad - Manifesto - Lobbying paper | \n
| Professional / Membership-based | \n- Professional associations - Trade unions - Chambers - Learned societies | \n- Member interests - Professional standards | \n- Representing - Standard-setting - Coordinating - Informing members | \nOften internally reviewed for accuracy and consistency. | \nOften grounded in domain expertise, but perspective may reflect member interests rather than broader public interest. | \n- Professional guideline - Member newsletter - Position paper - Sector report | \n
| Public / Official | \n- Governments - Ministries - Public authorities - International organisations | \n- Public mandate - Regulation - Policy | \n- Announcing - Regulating - Legitimising - Informing the public | \nUsually formally reviewed and approved before publication. | \nOften authoritative about official positions and decisions, but may be selective, strategic, or self-presentational. | \n- Policy document - Official statement - FAQ page - Ministry report | \n
Genres are not exclusive to a single source category. A newsletter, for example, can be commercial, philanthropic, or professional. The genre alone is not a sufficient indicator of reliability or trustworthiness.
Modern knowledge is so specialised that no individual can verify all claims independently. We have to rely on experts and on credible sources. But this reliance is only as good as our ability to identify who genuinely is an expert and which sources have been produced under conditions that support reliability.
\nThree complementary angles support this judgement:
\nA formal pre-publication process by which information items are selected for publication, framed, modified, or omitted — and through which their content is quality-checked against the publisher's standards before reaching the public.
\nEditorial review thus combines two functions:
\nRigour varies — from informal one-person editing to formal multi-stage review. In traditional publication contexts, these roles were typically performed by separate entities — a journalist (content-creator) and a newspaper editorial team (content-editor). Peer review in academic publishing is a specialised form.
\nEditorial review is not neutral.
A formal process and a rigorous quality check do not preclude bias. Editorial decisions reflect the motives, interests, and editorial line of the publishing organisation — its mission, ownership, funding, and target audience. The same piece can be carefully checked and framed in ways that serve the publisher's agenda. Quality and bias are not mutually exclusive. This is Editorial Gatekeeping in action. An analogous function on digital platforms is performed by Algorithmic Gatekeeping.
In digital environments, the boundaries shift. The editorial role is often bypassed — anyone with an account can post or share information items without editorial review or institutional backing. At the same time, the operator role becomes more prominent: algorithmic curation, sponsored content, and recommendation systems shape visibility in ways that traditional distribution networks did not.
\nThe reliability of an information item depends on its content-creator and the editorial review it received — not on whoever shared it.
\nThe practice of verifying factual claims in information items.
\nFact-checking exists in two distinct contexts:
\nFunding: donations (major donors include Luminate, Brost-Stiftung, Open Society Foundations, Google, Deutsche Telekom); a commercial subsidiary receives payment from Facebook for fact-checking content
\nMost major fact-checking organisations receive funding or partnership payments from large digital platforms (Meta/Facebook in particular). This relationship is openly disclosed but warrants critical reflection.
EDMO Monthly Fact-Checking Briefs: https://edmo.eu/resources/fact-checking-publications/fact-checking-briefs/
Lateral reading is a strategy of evaluating sources like a fact-checker.
\nThe initial focus is on questioning the source (who is behind the content), not the content itself. Instead of reading vertically — staying on a website and analysing it from top to bottom — you leave the site quickly, search for the source's name, and check what other trustworthy sites say about it. If the source seems untrustworthy, don't waste your time on it. Find a better source.
\nThe same lateral approach also works for verifying digital platforms, specific claims, quotes, or statistics within the content: leave the page and search for independent confirmation.
\nDon't try to judge a source by reading it. Leave it. Use the rest of the web to find out what it is.
Identifiability
\nQualification and Subject Expertise
\nInstitutional Backing, Funding, and Bias
\nPublication Track Record
\nRecency and Continuity
\nIndicators require interpretation. Some signals are easily misread.
→ Fame is not credibility. Public visibility, media presence, or large following do not equal expertise on a specific topic.
→ A high citation count is not in itself credibility. Frequently cited work may also be frequently challenged or contested — the context of citations matters. → Institutional affiliation is not in itself credibility. Institutions can carry their own agendas or biases and should be examined alongside the author.
→ A sudden shift to topics outside the author's previous expertise is not automatically a red flag, but warrants closer examination.
Indicators require interpretation. Some signals are easily misread.
→ Non-profit status alone is not credibility. Many advocacy and front organisations operate as non-profits.
→ Professional design and a polished website alone are not credibility. Web design has outpaced our methods for evaluation.
→ A long operating history alone is not reliability — ownership, funding, and practices can change over time.
→ Institutional affiliation is not in itself credibility. Institutions can carry their own agendas or biases and should be examined alongside the organisation.
\nphysical or digital contexts in which information items are available and encountered directly
| Information Environments | \nDefinition | \nWhat do I usually encounter there first? | \n
|---|---|---|
| Analogue Information Environments | \nnon-digital environments in which people encounter information items directly | \ninformation items in printed, transmitted, spatial, live, or spoken form | \n
| Digital Information Channels & Platforms | \ndigital environments in which people encounter information items directly | \ninformation items in posted, uploaded, or streamed form | \n
The boundary between analogue and digital environments is increasingly fluid: live broadcasts may be streamed simultaneously, print publications often appear in digital form, and live events are frequently broadcast or streamed online.
| Type | \nDefinition | \nTypical Examples (encounter on purpose) | \nTypical Examples (encounter in passing) | \nHow do I encounter information here? | \n
|---|---|---|---|---|
| Print / Publications | \nPrinted materials — such as newspapers, magazines, books, brochures, flyers, posters, that are physically distributed, displayed, or made available for reading. | \n- newspaper subscription - magazine purchase - academic textbook - library book | \n- campaign flyer in a letterbox - brochure in a waiting room - free newspaper on a train - billboard - poster at a bus stop | \n- by purchasing, subscribing, or picking up a copy - by reading, browsing, or leafing through - sometimes by noticing printed material placed in a public space | \n
| Broadcast & Public Address | \nAudio or audiovisual information transmitted to a wide audience through radio, television, screens, or public address systems either as scheduled programming or as continuous / situational announcements. | \n- TV news at home - radio programme - live sports broadcast | \n- TV screen in a waiting room - Music radio in a shop - loudspeaker announcement at a train station | \n- by tuning in to a channel or station - by switching between channels - sometimes by being in a space where broadcast or address content is playing | \n
| Installations, Monuments & Exhibits | \nSpatial, static, and typically permanent or semi-permanent objects, structures, or displays placed in public or institutional spaces to inform, commemorate, or express. | \n- museum exhibition - memorial visit - gallery installation - information pavilion | \n- monument in a public park - commemorative plaque - temzorary public art installation | \n- by visiting a space that contains them - sometimes by passing by or being physically present where they are placed | \n
| Live Events & Public Action | \nTime-bound, live gatherings or actions involving people — organised to inform, persuade, express, commemorate, or mobilise. | \n- conference - lecture - theatre performance - campaign event - planned demonstration | \n- street protest - flashmob - information stand - protest camp - vigil - street performance | \n- by attending at a specific time and place - sometimes by being in the vicinity when an event or action takes place | \n
| Personal Conversations & Word of Mouth | \nDirect, face-to-face exchange of information between individuals — in private, social, or professional settings. | \n- asking a friend for advice - family dinner discussion - consulting a colleague - parent–teacher conversation | \n- overhearing a conversation - casual remark at a social gathering - word of mouth in a community | \n- by talking, listening, and asking in direct personal interaction | \n
| Type | \nDefinition | \nTypical Examples | \nWho can publish information here? | \nIs what I see here the same for everyone, algorithmically personalised for me, or based on my own configuration? | \nHow do people find and move through information here? | \n
|---|---|---|---|---|---|
| Websites / Blogs | \nAn individual's or organisation's own online space — such as an institutional website, a news site, a company page, or a personal blog — where content is published directly in the owner's name. | \n- who.int - bbc.co.uk - greenpeace.org- my-travelblog.eu | \nOwner-controlled - The owner - Authorised contributors | \nSame for everyone Content is structured by the site owner through menus, categories, and page layout | \n- By following menus - By clicking links - By browsing sections or categories - By using internal search, if available - By moving from page to page | \n
| Social Media | \nNetworked digital spaces where users and organisations post, share, and circulate content. | \n- Facebook - TikTok - X - Threads - Mastodon | \nOpen - Almost anyone with an account - Organisations - Public figures - Advertisers | \nAlgorithmically personalised for me Feeds, recommendations, trending content, and promoted posts are shaped by my behaviour and engagement | \n- By scrolling through feeds - By following accounts - By opening comments, replies, and threads - By clicking shares, reposts, and recommendations - By searching hashtags, keywords, or account names | \n
| Video / Audio Platforms | \nDigital spaces where many different users or organisations upload and share video or audio content. | \n- YouTube - Vimeo - SoundCloud | \nOpen - Almost anyone with an account or channel - Organisations - Media producers - Podcasters | \nAlgorithmically personalised for me Recommendations, autoplay queues, and suggested items are shaped by my viewing / listening history | \n- By searching for specific items - By moving through playlists or queues - By following recommendations - By opening channels, episodes, or series - By using subscriptions | \n
| Streaming / On-demand Services | \nCurated digital services that offer access to a catalogue of on-demand media. | \n- Netflix - Disney+ - BBC iPlayer - Arte - RaiPlay | \nCurated - The provider - Authorised producers - Licensed content partners | \nAlgorithmically personalised for me Recommendations, featured selections, and catalogue presentation are shaped by my watch history | \n- By browsing the catalogue - By selecting from featured content - By continuing series or programmes - By following recommendation rows - By searching titles, genres, or categories | \n
| Communication / Messaging Apps | \nPrivate or semi-private spaces — such as messaging or email apps — for direct exchange between individuals or groups. | \n- WhatsApp - Signal - Messenger - Gmail - Outlook | \nMember-controlled - Participants in the conversation - Members of the group - Mailing-list senders | \nBased on my own configuration What I see depends on my contacts, conversations, groups, and how I organise them | \n- By opening chats or threads - By reading chronological message streams - By following forwards, links, and attachments - By searching chat history, senders, or keywords | \n
| Discussion Forums / Community Spaces | \nInteractive spaces — such as forums, discussion boards, comment sections, or community groups — where users discuss, comment, ask questions, and respond to one another. | \n- Reddit - Discord - Stack Overflow - Quora | \nSemi-open - Members - Registered users - Moderators | \nSame for everyone / Based on my own Configuration What I see depends first on which communities, servers, forums, or threads I join or enter; within them, visibility may then be shaped by upvotes, moderation, and sorting. | \n- By browsing threads - By opening replies and subthreads - By following notifications - By moving across categories or communities - By searching topics, tags, or thread titles | \n
Some platforms fall into more than one type — for example, Telegram (messaging + public channels), Discord (messaging + community), Reddit (forum + social media), Spotify (streaming + podcast discovery).
| Barrier | \nExamples | \nPeople typically excluded | \n
|---|---|---|
| Financial barriers | \n- Paywalls - Subscription fees - Premium memberships | \n- Low-income households - Students without institutional access - Users in regions with weaker currencies | \n
| Language barriers | \n- Foreign language - Academic or technical vocabulary - Complex sentence structures | \n- Non-native speakers - Users with lower formal education - Younger users | \n
| Disability-related barriers | \n- No screen reader support - No captions or transcripts - Low contrast or small text | \n- Users with visual impairments - Users with hearing impairments - Users with cognitive impairments | \n
| Technical barriers | \n- High bandwidth requirements - Modern devices needed - Specific software or apps | \n- Users with older devices - Rural users with limited internet - Low-income households | \n
| Geographic barriers | \n- Region locks - Local platform restrictions | \n- Users in restricted regions - Travellers | \n
| Political / regulatory barriers | \n- Government censorship - Government bans on platforms or apps - Sanctions or embargoes | \n- Users in countries with restrictive media policies - Activists, journalists, dissidents | \n
| Account & identification barriers | \n- Mandatory account creation - Identity verification - Age verification | \n- Users without an email or phone - Users without official ID - Minors without parental permission | \n
| Privacy-related barriers | \n- Mandatory data sharing - Tracking requirements - Required app permissions | \n- Privacy-sensitive users- Users avoiding government surveillance- Users wary of corporate data collection | \n
These barriers are not mutually exclusive, they often overlap or reinforce each other — for example, the Great Firewall of China blocks specific platforms only within that country, acting as both a geographic and a political/regulatory barrier.
Systems through which information items are accessed indirectly — through previews / snippets, or through synthesised responses based on them.
\n| Information Access Systems | \nDefinition | \nWhere do I start? | \nWhat appears there? | \n
|---|---|---|---|
| Search Systems | \nSystems that locate and rank information items in response to a user query, returning previews rather than the items themselves. | \na search interface | \na ranked list of previews: - titles - snippets - URLs | \n
| Discovery Systems | \nSystems that surface and curate information items in feeds or directories, returning previews rather than the items themselves. | \na feed or browse interface | \ncurated previews: - headlines - cards - recommendations | \n
| Generative AI Systems | \nSystems that generate or synthesise responses based on information items rather than directing users to them. | \na prompt window | \na generated, synthesised response, sometimes with citations | \n
information items accessed indirectly through search
→ users enter a query in a search interface and receive a ranked list of previews pointing to the underlying items
| Type | \nDefinition | \nTypical Examples | \nWhat is my starting point and how do I look for information here? | \nWhat determines which information items are displayed and in what order? | \nWhat do I get back? | \nHow do I move from here to the underlying information item(s), if at all? | \n
|---|---|---|---|---|---|---|
Search Engines | \nSystems that index and rank information items across the web based on user queries. | \n- Google - Bing - DuckDuckGo - Ecosia | \n- search bar - by entering keywords or questions | \n- my query - ranking algorithms - relevance - recency - search engine optimisation - sometimes personalisation based on location or search history | \nPreviews of existing information items: - ranked list of results with snippets, titles, and URLs - AI-generated summaries | \n- by clicking a result link, which takes me directly to the information item in its original space | \n
Scholarly Databases | \nSystems that make scholarly information items searchable and accessible through search, filtering, and structured metadata. | \n- PubMed - JSTOR - Google Scholar - ERIC | \n- advanced search form with filters - by entering keywords - by filtering metadata - by combining search fields | \n- my query - metadata matching - citation count - relevance ranking - database-specific indexing | \nPreviews of existing information items: - bibliographic records - abstracts - sometimes links to or direct access to full texts | \n- by clicking through to the full text - sometimes directly within the same system - sometimes via external publisher - sometimes behind a paywall or institutional login | \n
information items accessed indirectly through curated feeds or directories → users browse a feed or directory and see previews pointing to the underlying items
\n| Type | \nDefinition | \nTypical Examples | \nWhat is my starting point and how do I look for information here? | \nWhat determines which information items are displayed and in what order? | \nWhat do I get back? | \nHow do I move from here to the underlying information item(s), if at all? | \n
|---|---|---|---|---|---|---|
| Feeds | \nSystems that present a continuous, personalised stream of recommended information items based on topics, interests, or user behaviour. | \n- Google News - Google Discover - Apple News - Upday - Microsoft Start | \n- personalised feed or browse interface - by scrolling through the feed - sometimes by selecting topics or categories - sometimes by entering keywords | \n- editorial curation - algorithmic curation (esp. trending detection) - algorithmic personalisation (esp. personalised ranking, recommendations) | \nPreviews of existing information items: - headlines - cards - snippets - thumbnails - recommendations from various sources | \n- by tapping or clicking a headline, card, or preview - this takes me to the information item in its original space | \n
| Directories | \nSystems that organise information items into browsable catalogues with categories, charts, and recommendations. | \n- Apple Podcasts - Spotify Podcasts - Pocket Casts | \n- browse interface with categories and charts - by browsing categories or top lists - by following recommendations - sometimes by entering keywords | \n- editorial curation - algorithmic curation (esp. trending detection - "popularity charts") - algorithmic personalisation (sometimes recommendations) | \nPreviews of existing information items: - listings with titles, descriptions, and episode lists - thumbnails - ratings - sometimes recommendations | \n- by tapping or clicking a listing - sometimes I can play the item directly within the app - the transition to the underlying host is not always visible | \n
Some platforms function as both Discovery Systems and Digital Channels — Apple Podcasts and Spotify Podcasts, for example, surface podcasts hosted elsewhere but also play the content directly within the app.
\ninformation items accessed indirectly through generative synthesis
→ users enter a prompt and receive a synthesised response, sometimes with references to underlying items
| Type | \nDefinition | \nTypical Examples | \nWhat is my starting point and how do I look for information here? | \nWhat determines which information items are displayed and in what order? | \nWhat do I get back? | \nHow do I move from here to the underlying information item(s), if at all? | \n
|---|---|---|---|---|---|---|
Training-data based AI | \nSystems that generate responses based on patterns learned during training, without real-time retrieval from external sources. | \n- ChatGPT - Claude - Gemini | \n- prompt window or chat interface - by entering prompts, questions, or follow-up instructions | \n- my prompt - the system's underlying model - training data with a knowledge cutoff | \nGenerated content: - synthesised response - based on patterns from training data - rarely with citations to specific sources | \n- the generated response is typically the endpoint - underlying sources are usually not retrievable | \n
RAG* AI | \nSystems that synthesise responses by retrieving relevant content from external sources at query time — from the open web or from a user-provided corpus. | \n- Perplexity - NotebookLM - Copilot - Bing Chat | \n- prompt window or search interface - by entering prompts or questions - sometimes after uploading user-provided sources | \n- my prompt - the retrieval mechanism (web search or document corpus) - the system's underlying model | \nGenerated content: - synthesised response based on retrieved sources - usually with citations to the underlying items | \n- through cited sources or links, when provided - to verify the synthesis against the original items | \n
Training-data based AI Assistants increasingly include retrieval modes (e.g., ChatGPT with web search, Claude with search). When activated, they function as RAG systems for that query — the boundary between the two types is mode-dependent rather than product-dependent.
\n*RAG (Retrieval-Augmented Generation): a method that combines AI generation with retrieval from a defined corpus of sources — the system synthesises responses based on the retrieved content rather than only on its training data.
","UPDATEDAT":"2026-05-27T06:11:11.985Z","ID":"e2005735-1c9b-4e19-b33b-001832b8657a","TITLE":"Information, Sources & Information Environments"}