Tor: What It Is, How It Works, and Its Limits

Tor makes it harder to link your connection to the sites you visit. Understanding how it routes traffic helps you assess its privacy benefits and limits.

By Daniel Sardá · Published on

7 min read1,441 words

In this article · 7 sections

Tor is a network that routes connections through several servers to make it harder to link a person to the sites they visit. Its value becomes clearer when you ask a specific question: Who can see which part of a connection?

Three names are often confused. The Tor network is the infrastructure of relays, volunteer-operated servers that forward traffic. Tor Browser uses that network and includes measures to reduce tracking. The Tor Project is the nonprofit organization that maintains and develops the software. The network and the browser are related, but they are not the same thing, as the Tor Project's introduction explains.

Tor can help people read, search for information, or communicate with greater privacy. Understanding its limits is part of using that protection wisely.

The path of a connection

Imagine opening Tor Browser to read a newspaper on its conventional website. Instead of connecting directly to the newspaper, the browser builds a circuit that typically passes through three relays: an entry, a middle, and an exit relay.

Traffic travels with multiple layers of encryption. Each relay can remove its own layer and learn the next step without receiving all the information about the route. That is where the onion metaphor comes from: the layers are removed one after another. The official overview of how Tor works describes the mechanism in more detail.

The key is how information is divided:

  • The entry relay knows your IP address and the next relay, but does not directly receive the address of the newspaper you want to visit.
  • The middle relay knows the neighboring relays between which it forwards traffic.
  • The exit relay connects to the newspaper and knows that destination, while the site sees the exit IP address instead of your original IP address.

This explains why a single relay does not have the complete route. It also shows why different observers must be considered separately. With a typical direct connection to Tor, your internet provider can recognize that you are using the network. That fact alone does not reveal which pages you visit through it.

Dividing information does not prevent every attack. An adversary able to observe both ends can compare traffic timing and patterns and try to link them. The Tor Project acknowledges this limit in its explanation of attacks on onion routing.

Key idea: Tor makes it harder to link a connection's origin and destination by distributing information among relays. That separation does not guarantee anonymity against every observer.

What HTTPS protects along the way

Return to the newspaper example. Tor's layers protect the path through the network up to the exit. A different technology, HTTPS, protects the content between your browser and a conventional website.

With HTTPS, the exit relay forwards encrypted content to the site and cannot read it simply by acting as the exit. Without that additional encryption, it can observe the content transmitted. Tor and HTTPS have different, complementary roles, as the browser's safe browsing guidance explains.

The newspaper does receive the information you send it. If you sign in to an account bearing your name, it can recognize you as the account holder even though it still sees a Tor exit IP. Hiding your original address does not erase the identity you disclose to the destination yourself.

This distinction clears up a common confusion: a connection can hide your IP from a site while still allowing that site to know who you are.

When Tor can be useful

Tor can help someone read the news without as easily exposing the link between their connection and their reading. It can also help people access information despite certain forms of censorship or support sensitive communications. The project's documentation describes these uses, though their effectiveness depends on the environment and the observer.

From the perspective of individual liberty, having space to read and communicate without constant surveillance has value. It can reduce barriers to expression and association: a person may need to explore ideas before deciding whether to defend them publicly. This is an editorial judgment about the importance of privacy; no tool can guarantee the exercise of those rights on its own.

Fear of being watched can also lead to self-censorship, when someone avoids expressing or exploring certain ideas because of the consequences they anticipate.

The same infrastructure can be misused. Recognizing its legitimate uses does not justify every action taken through it, just as recognizing abuse does not fully explain what it is for.

Tor can also be blocked. Some sites restrict connections from its exits, and routing through several relays can add delays. Documentation on speed and sites that block Tor exits describes these tradeoffs. Access and performance should not be expected to be uniform.

Privacy, anonymity, and security are different goals

To assess Tor, we can use three working definitions. Privacy means limiting the exposure of information about ourselves. Anonymity aims to make it harder to link an activity to our identity. Security encompasses the protection of systems and data against unauthorized access, alteration, and other harm.

The three goals can support one another, but they are not interchangeable. Tor Browser includes defenses to reduce tracking, including measures against identification based on browser characteristics. That does not make every site trustworthy or every download safe.

Nor does it automatically protect every connection on a device. Using Tor Browser does not cause another application open at the same time to send its traffic through Tor. Protection depends on which application uses the network and how the user behaves, according to the project's safe-use guidance.

Key idea: Before assessing a privacy tool, identify what information you want to protect and from whom. Hiding your IP from a site and hiding your identity when you sign in are different problems.

VPNs and incognito mode: differences that matter

A VPN and Tor both change the path of a connection, but they distribute trust differently. A VPN introduces a provider trusted to carry the traffic. Tor distributes the route among several relays so that no single relay has all the information. That difference does not mean one option is always better: the choice depends on the problem you are trying to solve.

Combining the two technologies does not automatically improve protection either. The Tor Project advises against generally recommending a VPN with Tor to beginners, because the combination requires an understanding of its implications. There is no need to add tools before first understanding what protection each one provides.

Incognito mode in an ordinary browser mainly affects traces left in the browser itself, such as the session's history. It does not by itself hide your IP address from the sites you visit. The official comparison with Tor Browser helps distinguish local privacy from observation by third parties.

Onion services take a different path

Tor also provides access to services with addresses ending in .onion. Unlike the conventional newspaper site in the example, these services are inside the Tor network: the connection does not need to reach the internet through an exit relay.

The system encrypts communication between the client and the service and allows the operator to hide the server's IP address. These are properties of onion services, not guarantees about the honesty of their operators. Knowing that you are connected to a particular address does not establish that its content or offers are trustworthy.

Tor is therefore not synonymous with the “dark web.” It allows access to conventional sites as well as onion services; the latter are only one part of what it can do.

Choose according to the problem you want to solve

If your goal is to make it harder for a site to learn your original IP address or to reduce the possibility of linking your connection to particular reading, Tor offers a specific mechanism for doing so. If you need a site not to know who you are, you must also consider the information you give it. And if the risk includes an observer able to monitor both ends, the circuit alone cannot resolve every limitation.

A useful decision begins with that assessment. To move from understanding the mechanism to using it, consult the official Tor Browser documentation, which explains its features and usage guidance. The more precisely you define your need, the easier it is to see what Tor can address and what calls for other precautions.

About the author

Daniel Sardá is an SEO Specialist, a university-level technician in Foreign Trade from Universidad Simón Bolívar, and editor of Libertatis Venezuela. He writes on liberalism, political economy, institutions, propaganda and individual liberty from an independent, non-partisan perspective.

Read full profile