6. Browsing and information access

Observation without participation.

What gets recorded when you read

Reading online leaves a trail even if you never log in or post a word. Every website you visit receives:

• your IP address, which reveals your network and usually a rough location;
• a browser fingerprint: screen size, installed fonts, time zone, the way your browser renders features. Together these can be distinctive enough to recognise repeat visits without cookies.

The network itself is a second layer. Your internet service provider can see the domains you connect to and, depending on how the site is set up, sometimes the full web address.

In the UK, providers are legally required to keep certain records and may be asked to hand them over under lawful authority. That does not mean your browsing is actively watched, but logs can exist and can be accessed.

Myth: "private browsing" stops websites or networks tracking you. Reality: it only stops your own device storing history and cookies after the window closes. The site still sees your traffic; your network still routes it.

Seeing without being seen

To read without an obvious link to your everyday identity, separate browsing from your usual accounts. Do not sign in to services while you research, and avoid a browser session that is already logged in elsewhere.

Classic example: reading about a health condition. Search while logged in to a large account and the topic can shape your recommendations or adverts later. A separate browser profile avoids that cross-linking.

Browser profiles are a practical tool. Keep one for routine use and another for sensitive reading, each with its own cookies and saved data.

This is less about secrecy and more about avoiding accidental correlations. It will not hide you from the website, but it stops different parts of your digital life merging.

Search engines and discovery

Search engines are not neutral pipes. They log queries, tie them to identifiers and tailor results.

For everyday use the convenience is often worth it. For sensitive research it can be a problem: a search about a legal dispute or a workplace issue could later become part of a profile attached to your account or device.

Ways to reduce it:

• use engines that do not keep long-term query logs;
• or use a mainstream engine signed out, in a fresh browser profile;
• vary your wording and leave overly specific personal details out of queries.

None of this erases the fact that your network made a request, but it lowers the chance a query becomes part of a durable personal profile.

Encrypted connections and what they do not hide

Most websites now use HTTPS, which encrypts traffic between your browser and the site. Your provider and others on the network cannot read the page content.

But HTTPS does not hide which site you visited. The domain is still visible in the connection data, and that alone can reveal your interests at a glance.

Example: the articles you read on a news site are protected, but your provider can still see the news domain. If the domain itself is sensitive, encryption does not remove that risk. With standard web browsing you can reduce this, but not fully eliminate it.

DNS and the map of where you go

Before connecting, your browser uses the Domain Name System (DNS) to turn a name into an IP address. DNS requests can be logged by your provider or by whichever resolver you use.

Switching to a privacy-focused resolver can cut logging outside your network, but it concentrates trust in that resolver. You are choosing who sees your map of destinations.

The risk is quiet rather than dramatic: over time, DNS logs build a useful picture of what you read.

The fix is encrypted DNS (often called DoH or DoT) from a resolver whose policies you accept. It reduces casual observation on local networks such as public Wi-Fi, but it does not make browsing anonymous. Trade-off: better protection on the path, more trust in the resolver.

Public Wi‑Fi and shared networks

Cafes, hotels and trains are handy places to browse, but shared networks come with predictable risks:

• even with HTTPS, a malicious hotspot can try to intercept traffic;
• a poorly configured site can expose content;
• people can simply read over your shoulder or in a screen reflection, a very common failure in public spaces.

The mitigations are ordinary:

• prefer HTTPS-only sites;
• avoid logging in to sensitive services on unknown networks;
• use a VPN if you already trust one.

A VPN does not make you invisible. It moves the observation point from the local network to the VPN provider, and you are giving that provider a view of your traffic instead.

Content blockers and the limits of consent

Many websites load third-party scripts for advertising, analytics and social media widgets, and those third parties can track you across different sites.

Content blockers stop known tracking scripts loading, which limits cross-site identification. Bonus: pages are often faster and less noisy.

The catch: blocking is imperfect and sometimes breaks things, such as video playback or comment sections. So it is contextual. You might allow tracking on a newspaper site to use a specific tool, while blocking it almost everywhere else.

There is no universal setting that delivers both full functionality and strong privacy.

Cookies, local storage, and persistence

Cookies are small pieces of data websites store in your browser. They make sessions possible, and they also allow long-term tracking.

Modern sites also use local storage and other browser features that can survive clearing cookies. That is not a reason never to allow cookies; it is a reason to understand that they create continuity over time.

A practical habit: clear site data regularly for sites you do not want to remember you, or use a browser that removes site data automatically on close.

That cuts long-term tracking, but it logs you out and wipes preferences. Acceptable in some contexts; needless annoyance in others.

Reading without interaction

Sometimes it helps to read with no interaction at all. A journalist checking a press release might fetch it through a text-only reader to reduce exposure to trackers and scripts.

RSS readers and text-only browsers are not new, but they still work: they retrieve the content and skip most of the tracking code.

Downside: some content will not load, especially on script-heavy sites. Simple fix: use these tools by default for routine reading, and switch to a full browser only when a site needs it.

Remote viewing and the difference between content and access

Some services fetch a page through their servers and then show it to you, which changes who appears to have accessed the site. Handy when you do not want your own IP address in the site's logs.

Familiar examples: a third-party text-only "reader" view, or a search engine cache. You still see the content; the site sees the intermediary.

The risk: you now trust the intermediary with what you read and the fact that you read it. If the content is sensitive, that trust decision matters.

It also does not stop your own network seeing that you contacted the intermediary. This is a partial protection, not a cloak.

Social platforms and logged-in browsing

Social platforms are among the most aggressive at linking browsing to personal profiles. Signed in to a major platform in one tab? Embedded buttons and scripts on other sites can still report your activity.

Myth: if you do not click "like", nothing is shared. Reality: the mere presence of the script is often enough.

The fix is separation. Use a different browser or a dedicated profile for social platforms, and do not use it for general browsing. Not paranoia, just a reasonable way to stop platforms building a fuller picture of your interests.

Local devices and the people around you

The quietest browsing risk is usually the device itself. Ordinary, everyday exposures:

• a shared computer remembers previous searches, visible to a family member or colleague;
• you hand someone your phone to check the weather and autocomplete shows recently visited sites.

Mundane but effective mitigations:

• separate user accounts on shared machines;
• autocomplete off for sensitive browsing;
• a habit of closing tabs when you are done.

Nothing dramatic, but it matches the way browsing habits actually leak.

When anonymity is not the goal

Not every situation calls for hiding. Sometimes a stable identity helps: it builds trust, supports professional work, or unlocks services that need verification.

The real question is whether your reading should be linked to that identity, not whether you must conceal yourself at all times.

In the UK, people often juggle public and private roles online: a professional profile for work, a separate space for personal reading. Keeping them distinct is self-preservation rather than secrecy. It reduces accidental mixing without claiming impossible levels of privacy.