Select Two (Multi-Select) Questions

Define a class named Calculator

Why it fails: Pseudocode is language-agnostic and typically avoids class definitions.

Execute SQL query to retrieve numbers from database

Why it fails: Pseudocode focuses on logic, not database queries.

Full control over underlying hardware

Why it fails: Full control over underlying hardware is an on-premises or IaaS characteristic; DBaaS abstracts the host, patching and replication away from the customer. It is tempting because DBaaS does expose some instance sizing and parameter options, but that is configuration within a managed platform, not hardware control, so it is not an advantage.

Requires dedicated IT staff for database administration

Why it fails: Dedicated IT staff for database administration is a cost of on-premises operation, not an advantage of DBaaS; the provider handles patching, backups and scaling. It is tempting because DBaaS still needs some staff for schema, query tuning and access control, but that is shared responsibility, not dedicated infrastructure administration.

Higher initial capital expenditure

Why it fails: Higher initial capital expenditure describes on-premises databases, where servers and licences are bought upfront; DBaaS shifts spend to operational subscription costs. It is tempting because large cloud commitments can look capital-like, but the billing model remains consumption-based operating expenditure, so it is the opposite of a DBaaS advantage.

Brand name

Why it fails: Brand name identifies the vendor but does not itself determine licence rights, restrictions or cost model. It is tempting because vendor reputation can sway purchasing decisions, yet the question asks for licence-selection factors, and brand is a procurement preference rather than a term within the licence agreement.

Font style

Why it fails: Font style governs document rendering, not licensing terms; it cannot be evaluated when selecting a licence. It is tempting because typography is a genuine software configuration choice, and font licensing does exist separately, but that concerns typeface distribution rights, not the licence-selection factors this question targets.

Color of interface

Why it fails: Interface colour is a cosmetic design attribute with no bearing on licence terms such as permitted users, redistribution or warranty. It is tempting because colour accessibility can influence software procurement, but that is a usability consideration, not a licensing factor, so it cannot answer this question.

ROM

Why it fails: ROM retains its contents when power is removed, so it is non-volatile and cannot be a volatile memory example. It is tempting because ROM is memory and is often listed alongside RAM, but ROM is the correct answer when a question asks for non-volatile memory examples.

BIOS firmware

Why it fails: BIOS firmware is stored in non-volatile flash or ROM, retaining contents without power, so it is not volatile memory. It is tempting because BIOS runs at startup and interacts with RAM, but firmware is the correct answer when the question asks for non-volatile storage examples.

SSD

Why it fails: An SSD retains data without power, making it non-volatile storage, so it cannot be an example of volatile memory. It is tempting because SSDs are fast and sit close to RAM in the memory hierarchy, but they are correct answers when the question asks for non-volatile storage examples.

Firewall

Why it fails: A firewall is a network security control that filters traffic, not malicious software. It is tempting because firewalls and malware both feature in security discussions, but a firewall is the correct answer when the question asks for a defensive perimeter device.

Encryption

Why it fails: Encryption is a protective technique that renders data unreadable, not malicious code. It appears plausible because ransomware uses encryption, yet encryption itself is a legitimate control and would be the right answer when asked how to protect data at rest or in transit.

Phishing

Why it fails: Phishing is a social-engineering attack that tricks users into revealing credentials, not malicious code executing on a host. It is tempting because phishing often delivers malware as an attachment or link, so it appears in the same threat landscape, but it would be the correct classification for a question asking about attack vectors rather than malware types.

Adware

Why it fails: Adware is malware that displays unwanted advertisements; it manipulates software, not people, so it involves no human persuasion. It is tempting because adware often arrives via deceptive downloads, and it would be the correct choice for a question asking about types of malware or unwanted software.

Spoofing

Why it fails: Spoofing falsifies an address or identity at the protocol level, such as IP, ARP or email headers; it does not rely on manipulating a person's trust. It is tempting because spoofed emails often carry phishing lures, and it would be correct for a question about network-layer attacks or impersonation techniques.

String

Why it fails: A string is a composite sequence of characters, not a primitive scalar in most languages. It is tempting because strings are ubiquitous and often written with literal syntax, but the question targets single-value types such as integer, boolean and float.

Array

Why it fails: An array is a composite aggregate holding multiple elements, not a primitive; primitives store a single value directly. It is tempting because arrays appear in nearly every language, but the question asks for built-in scalar types, not containers.

Network topology

Why it fails: Network topology may affect backup speed but is not a critical factor for selecting a backup method.

Display resolution

Why it fails: Display resolution is unrelated to backup strategy.

Operating system version

Why it fails: OS version may affect compatibility but is not a primary concern for backup selection.

Reinstalling the operating system

Why it fails: Reinstalling the operating system destroys the environment rather than isolating the faulty logic, so it cannot reveal where output diverges. It is tempting as a generic remedy for corrupted system state, and would be justified when the OS itself is unstable or infected, not when code produces wrong results.

Changing the program requirements

Why it fails: Changing requirements alters what the program should do, not how you locate the defect, so it is not a debugging technique. It belongs to requirements gathering or scope negotiation, where stakeholders revise specifications before development begins.

Creating separate tables for each customer order without linking them

Why it fails: Separate unlinked order tables duplicate customer and product data and destroy referential integrity, which is the opposite of normalisation. Normalisation links tables through primary and foreign keys. This approach would only suit archiving isolated historical data where no cross-order querying is required.

Storing all product details in a single column as a list

Why it fails: A list in one column violates first normal form, which requires atomic values with no repeating groups; queries cannot filter or join on individual attributes. It is tempting because denormalised storage suits document databases or read-heavy caches, but the stem specifies a normalised relational design.

Storing calculated totals in the same table to avoid joins

Why it fails: Storing derived totals breaches third normal form: non-key attributes must not depend on other non-key attributes, and the value becomes stale whenever source rows change. It is tempting to avoid join cost in reporting workloads, where materialised aggregates or indexed views are the appropriate mechanism instead.

Cellular compatibility

Why it fails: Cellular compatibility concerns mobile carrier bands, which are unrelated to the 2.4 GHz and 5 GHz unlicensed Wi-Fi spectrum. It is tempting because both involve radio frequencies, but the band choice depends on range, interference and channel width instead.

Number of devices

Why it fails: Band selection does not directly limit device count.

Writing passwords on sticky notes and attaching them to monitors

Why it fails: Attaching written passwords to monitors exposes credentials to anyone passing or photographing the workspace, defeating physical access controls. It tempts because users genuinely do this to remember complex passwords, and written records would be acceptable only if stored in a locked container with controlled access.

Sharing access badges with colleagues

Why it fails: Badge sharing destroys the audit trail linking individuals to entry events, so accountability for access is lost. It is tempting because lending a badge feels helpful when a colleague forgets theirs, and visitor-management processes do issue temporary credentials — but those are logged and time-bound, unlike passing over a personal badge.

Leaving doors propped open for convenience

Why it fails: Propping doors defeats the access-control boundary, letting anyone enter without authentication. It is tempting because holding a door for a laden colleague seems courteous, and legitimate scenarios exist — scheduled deliveries or moves — but those use staffed, logged entry points rather than disabling the latch.

SQLite

Why it fails: SQLite is an embedded, file-based library, not a cloud service, so it cannot deliver the managed high availability or automatic scaling the e-commerce application demands. It is tempting because it is a genuine relational database, and it would be the right choice for a local, single-user application or mobile app needing zero configuration.

Microsoft SQL Server Express

Why it fails: SQL Server Express is an on-premises edition limited to 10 GB per database, with no automatic scaling or high availability. It is tempting because it is free and relational, and would be correct for a small local application with modest data and no resilience requirement.

MySQL Community Edition

Why it fails: MySQL Community Edition is self-hosted software; the administrator must provision servers, clustering and scaling themselves, so it provides no managed cloud high availability or automatic scaling. It is tempting because MySQL is a relational database, and it would be correct for an on-premises deployment where the team controls the infrastructure.

Using the same password for all accounts

Why it fails: Reusing one password across accounts lets a single credential leak unlock everything, and ransomware operators harvest credentials to spread laterally. It is tempting as a memorability shortcut, but password reuse is correct only in scenarios demonstrating poor hygiene, not protection.

Disabling the firewall

Why it fails: Disabling the firewall removes filtering of inbound and outbound traffic, letting ransomware payloads and command-and-control connections through unchecked. It is tempting when troubleshooting connectivity, but disabling a firewall would only be defensible temporarily during diagnosis, never as a protective control.

Opening email attachments from unknown senders

Why it fails: Opening attachments from unknown senders executes untrusted payloads, the primary ransomware delivery vector. It is tempting because the attachment may appear to be a legitimate invoice or CV, but opening such files is correct only when demonstrating how infection occurs, not how to prevent it.

Eliminates need for antivirus software

Why it fails: A VPN secures data in transit through encryption; it does not inspect, detect or remove malicious files on the endpoint, so antivirus remains necessary. It is tempting because encrypted tunnels feel comprehensive, but a VPN would be the correct answer to a question about protecting data confidentiality over untrusted networks, not malware defence.

Reduces hardware costs

Why it fails: A VPN is software or a service running on existing endpoints and routers; it does not replace physical equipment, so hardware spending is unaffected. It is tempting because cloud-delivered VPNs avoid some appliance purchases, but a VPN would be the correct answer to a question about secure connectivity, not cost reduction.

Increases network bandwidth

Why it fails: A VPN encapsulates and encrypts traffic, adding overhead that typically reduces throughput rather than increasing available bandwidth, which is capped by the underlying link. It is tempting because tunnelling can compress some payloads, but a VPN would be the correct answer to a question about confidentiality or secure remote access, not bandwidth gains.

PDF

Why it fails: PDF is a page-description document format for text and print layout, not a raster or vector image format, so it does not belong in a list of image formats. It is tempting because PDFs can embed images, and it would be correct when the requirement is portable documents.

TXT

Why it fails: TXT is a plain-text container, carrying no pixel data or image encoding, so it cannot store a picture. It is tempting because TXT is a widely recognised, universally readable file format, and it would be the right choice when the requirement is to capture unformatted text such as logs, notes or configuration output.

MP3

Why it fails: MP3 is a lossy audio compression format carrying sound data, not image pixels or vector graphics, so it cannot be a common image format. It is tempting because it is a widely recognised media container, and it would be correct when the requirement is compressed audio playback.

Lower latency than on-premises in all cases

Why it fails: Latency depends on network distance between client and database region, so a cloud database can be slower than a co-located on-premises server. It is tempting because cloud providers offer many regions and edge locations, which reduce latency only when the database is placed near the users.

Full control over the operating system

Why it fails: DBaaS abstracts the underlying operating system, so the provider patches and manages it; customers gain no OS-level control, unlike self-managed databases on IaaS virtual machines. It is tempting because full OS control is a real advantage of running a database on your own infrastructure.

Two different passwords

Why it fails: Two passwords both sit in the knowledge factor, so combining them satisfies only one authentication category and never constitutes MFA. This is tempting because it uses two separate credentials, but MFA requires distinct factors such as knowledge plus possession, for example a password with an authenticator app code.

A username and a password

Why it fails: A username identifies the account rather than authenticating it, and a password alone is single-factor knowledge authentication. This pairing is tempting because it is the most familiar login pattern, but MFA demands two different factor types, such as a password plus a hardware token or biometric.

A password and a security question

Why it fails: A security question is recalled knowledge, the same factor category as the password, so pairing them yields single-factor authentication. It is tempting because two prompts appear, but MFA requires different categories, for example a password combined with a smart card or fingerprint.

Apple App Store License

Why it fails: The Apple App Store License governs distribution of apps through Apple's proprietary storefront, imposing Apple's terms rather than granting the freedoms (use, modification, redistribution) that define open-source licences. It is tempting because it is a genuine software licence, and would be the correct answer if the question asked for a proprietary or end-user distribution agreement.

Microsoft EULA

Why it fails: The Microsoft EULA is a proprietary end-user licence agreement that reserves all rights to Microsoft and prohibits modification or redistribution, so it cannot be an open-source licence. It is tempting because it is a genuine, familiar software licence, and would be the correct choice if the question asked for a proprietary commercial licence.

Adobe EULA

Why it fails: The Adobe EULA is a proprietary end-user licence agreement restricting copying, modification and redistribution, so it grants none of the freedoms an open-source licence provides. It is tempting because it is a real, widely encountered software licence, and would be correct if the question asked for a commercial proprietary licence instead.

Table is too large

Why it fails: A large table slows retrieval but does not itself produce an error; the statement remains valid. It is tempting because size correlates with poor performance, and it would be correct when capacity or query-timeout tuning is the concern rather than a malformed or unresolvable SELECT.

Index is missing

Why it fails: A missing index causes slow scans, not an error; the query still returns correct rows. It is tempting because indexing is a common tuning topic, and it would be correct when optimising a valid query's performance rather than explaining why execution fails.

Network is slow

Why it fails: Slow network affects query latency, not the validity of the SQL statement, so it produces no execution error. It is tempting because performance degradation feels like failure, and it would be correct when diagnosing timeouts on a remote database connection rather than syntax or schema faults.