Data Security

These three terms sound similar but mean different things. Mixing them up is the most common way to lose marks in this topic, so learn each precisely.

• Data security — protecting data from unauthorised access and from loss or damage. (How do we keep it safe?)
• Data privacy — controlling who is allowed to see or use the data. (Who is permitted?)
• Data integrity — keeping the data accurate, complete and consistent, so it is not changed incorrectly or corrupted. (Is the data still correct?)

A quick test: "Only your doctor may view your record" is privacy; "a backup protects the record from being lost" is security; "a check stops an age of 999 being stored" is integrity.

Security is not only about the data — the computer system itself must be protected too. If the system is compromised, the data on it is automatically at risk.

Threats to the data: it could be stolen (copied by a hacker), deleted or corrupted (by malware or accident), or read by the wrong people (no access control).

Threats to the system: the hardware and software could be damaged or disabled by malware, taken over by a hacker, or have its settings changed so it no longer works correctly.

Because the two are linked, good security uses layers: physical protection (locked rooms, locked devices), software protection (anti-malware, firewall, encryption) and human protection (training, strong passwords). No single measure is enough on its own — this is called defence in depth.

Most systems give each person a user account so the computer knows who is using it. Authentication is the process of confirming that the person really is who they claim to be. Common techniques:

• Passwords — a strong password is long, mixes upper case, lower case, digits and symbols, and is not a dictionary word or personal detail. A larger number of possible combinations makes it harder to brute-force, and avoiding real words defeats a dictionary attack.
• Two-step / two-factor verification — after the password, a second factor is needed (e.g. a one-time code sent to a phone). Even a stolen password is then not enough.
• Biometrics — authentication by a unique physical feature (fingerprint, face, iris, voice). It cannot be forgotten or easily shared, but the scanner hardware costs more, can suffer a false rejection of a genuine user, and biometric data cannot be changed if it is ever stolen.
• Digital signatures — used to confirm identity for messages/documents (covered in §4).

A digital signature uses public-key cryptography to prove a message's sender (authentication) and that it has not been changed (integrity).

How it works:

1. The sender runs the message through a hashing algorithm to make a fixed-length digest (a fingerprint of the message).
2. The sender encrypts the digest with their own PRIVATE key — this encrypted digest is the digital signature, sent with the message.
3. The receiver decrypts the signature with the sender's PUBLIC key. It only works if the sender's private key created it, so this confirms the sender's identity.
4. The receiver re-hashes the received message to make their own digest.
5. If the two digests match, the message is unchanged (integrity); if they differ, it has been altered.

The key direction is what students get wrong: sign with the private key, verify with the public key. Only the sender owns the private key, which is exactly why a successful decryption proves identity.

Firewall — hardware or software placed between a network and the internet. It monitors all incoming and outgoing traffic and blocks anything that does not meet its rules (e.g. blocked ports or IP addresses). This stops hackers gaining unauthorised access and can stop suspicious software sending data out. (A firewall does NOT detect viruses — that is the job of anti-malware.)

Anti-virus / anti-spyware software — scans files and the computer for known malware and removes or quarantines anything it finds. It must be kept up to date so it can recognise the newest threats, and run regular scans.

Encryption — scrambles data using an algorithm and a key so it becomes ciphertext, meaningless to anyone without the key. The original data is recovered by decrypting with the correct key. Encryption protects confidentiality: if data is intercepted it cannot be understood — but it does not stop the data being intercepted or copied in the first place.

Connecting to a network or the internet exposes a system to several threats:

• Malware — malicious software. Viruses attach to files and replicate, corrupting or deleting data. Spyware secretly collects information (e.g. logging keystrokes) and sends it to an attacker.
• Hackers — people who gain, or try to gain, unauthorised access to a system or its data.
• Phishing — a fake email or message pretending to be from a genuine organisation, luring the user into clicking a link to a fake website and entering personal details. The user must take the bait.
• Pharming — malicious code that redirects the user to a fake website even when they type the correct address. No click is needed, so it is harder to spot.

Methods to restrict the risks (match the measure to the threat):

Threat	Main defence(s)
Viruses / spyware	Up-to-date anti-virus / anti-spyware, don't open unknown attachments
Hackers	Firewall, strong authentication, encryption of stored/sent data
Phishing	User training to spot fake emails, never click suspicious links, check the real URL
Pharming	Keep software/anti-malware updated, check the site is genuine (e.g. valid certificate / HTTPS)

Protecting the data itself relies most on encryption (intercepted data is unreadable) and access rights (each user can only read/edit the data their role allows, limiting damage from a compromised or insider account).