📚 A-Level AQA Computer Science: Clarifying Concepts | A-Level AQA 计算机：概念辨析
In A-Level AQA Computer Science, students often encounter pairs of concepts that sound similar yet have distinct meanings. Distinguishing them accurately is not only vital for exams but also builds a solid foundation for advanced study. This article selects ten common areas of confusion, explaining each pair clearly and concisely.
在 A-Level AQA 计算机科学中，学生经常会遇到听起来相似但含义不同的概念对。准确区分它们不仅对考试至关重要，还能为深入学习打下坚实基础。本文精选了十个常见的混淆点，逐一进行清晰简洁的解释。
1. Abstract Data Type (ADT) vs Data Structure | 抽象数据类型与数据结构
An abstract data type is a conceptual model that defines what operations can be performed on data, without specifying how these operations are implemented. It describes the logical behaviour from the user’s perspective, such as ‘push’ and ‘pop’ for a stack.
抽象数据类型是一种概念模型，它定义了可以对数据执行哪些操作，但不规定这些操作如何实现。它从用户角度描述逻辑行为，例如栈的“压入”和“弹出”。
A data structure, on the other hand, is the concrete implementation of an ADT in a programming language. It determines how data is organised in memory, for instance using an array or linked list to build a stack.
另一方面，数据结构是 ADT 在编程语言中的具体实现。它决定了数据在内存中如何组织，例如使用数组或链表来构建一个栈。
So a stack is an ADT, while an array-based stack is a data structure. The ADT focuses on what, and the data structure focuses on how.
因此，栈是一个 ADT，而基于数组的栈是一种数据结构。ADT 关注“做什么”，数据结构关注“怎么做”。
2. Stack vs Queue | 栈与队列
Both are linear ADTs, but their access patterns differ. A stack follows Last In, First Out (LIFO) – the last element added is the first one removed. Typical operations are push and pop.
两者都是线性 ADT，但访问模式不同。栈遵循后进先出 (LIFO) 原则——最后添加的元素最先被删除。典型操作是压入和弹出。
A queue operates on First In, First Out (FIFO). Elements are added at the rear (enqueue) and removed from the front (dequeue), exactly like a waiting line.
队列遵循先进先出 (FIFO) 原则。元素在队尾添加（入队），从队首删除（出队），就像排队一样。
Choosing between them depends on the problem: function call management uses a stack, while print spooling often uses a queue.
在问题中选择使用哪个取决于场景：函数调用管理使用栈，而打印假脱机通常使用队列。
3. Compilation vs Interpretation | 编译与解释
A compiler translates high-level source code into machine code all at once, producing an executable file that can be run independently. The translation process is slower upfront, but execution is fast and does not require the compiler to be present.
编译器一次性将高级源代码翻译成机器码，生成一个可以独立运行的可执行文件。前期翻译过程较慢，但执行速度快，且不需要编译器在场。
An interpreter translates and executes code line by line. No separate executable is generated; the interpreter must be present each time the program runs. This makes debugging easier but execution slower.
解释器逐行翻译并执行代码。不会生成单独的可执行文件；每次运行程序时都必须有解释器在场。这使得调试更容易，但执行速度较慢。
Hybrid approaches like bytecode compilation (e.g., Java) combine both: source is compiled into bytecode, which is then interpreted or JIT-compiled by a virtual machine.
字节码编译等混合方式（如 Java）将两者结合：源代码被编译成字节码，然后由虚拟机解释或即时编译。
4. Black-box Testing vs White-box Testing | 黑盒测试与白盒测试
Black-box testing examines software functionality without any knowledge of internal code structure. Tests are derived from specifications; the focus is on inputs and expected outputs. Techniques include equivalence partitioning and boundary value analysis.
黑盒测试在不了解内部代码结构的情况下检查软件功能。测试依据规格说明设计，关注输入和预期输出。技术包括等价划分和边界值分析。
White-box testing uses the internal logic to design test cases. The tester examines code paths, branches, and conditions to ensure every statement is executed at least once. Examples are statement coverage and decision coverage.
白盒测试利用内部逻辑设计测试用例。测试者检查代码路径、分支和条件，确保每条语句至少被执行一次。例如语句覆盖和判定覆盖。
Both are essential: black-box validates the system against requirements, white-box uncovers hidden logic errors.
两者都必不可少：黑盒验证系统是否符合需求，白盒发现隐藏的逻辑错误。
5. Program Counter (PC) vs Current Instruction Register (CIR) | 程序计数器与当前指令寄存器
The Program Counter holds the memory address of the next instruction to be fetched. After a fetch, the PC is incremented to point to the subsequent instruction, or altered on a jump.
程序计数器保存下一条要取指的指令的内存地址。取指后，PC 递增以指向下一条指令，或在跳转时被修改。
The Current Instruction Register holds the actual binary instruction that has just been fetched from memory. It waits in the CIR while the control unit decodes and executes it.
当前指令寄存器保存刚从内存中取出的实际二进制指令。它在 CIR 中等待控制单元解码和执行。
Thus the PC is a pointer to an address; the CIR stores the instruction itself. Both are key components in the fetch-decode-execute cycle.
因此，PC 是指向地址的指针；CIR 存储指令本身。两者都是取指-解码-执行周期中的关键组件。
6. TCP vs UDP | TCP 与 UDP
Transmission Control Protocol provides reliable, connection-oriented communication. It establishes a session, numbers packets, acknowledges receipt, and retransmits lost data. Ideal for web browsing and email.
传输控制协议提供可靠的、面向连接的通信。它建立会话，对数据包编号，确认接收，并重传丢失的数据。适用于网页浏览和电子邮件。
User Datagram Protocol is connectionless and sends datagrams without error checking or sequencing. It is faster but less reliable, making it suitable for live streaming or online gaming where occasional loss is tolerable.
用户数据报协议是无连接的，发送数据报时不进行错误检查或排序。它更快但可靠性较低，适用于偶尔丢失可以容忍的直播或在线游戏。
In summary, TCP emphasises accuracy, while UDP prioritises speed.
总之，TCP 强调准确性，而 UDP 优先考虑速度。
7. Recursion vs Iteration | 递归与迭代
Recursion occurs when a function calls itself to solve smaller instances of the same problem. It relies on a base case to stop. Recursive solutions are often elegant for problems like tree traversal, but may consume more stack memory.
递归是指函数调用自身来解决同一问题的更小实例。它依赖基线条件来停止。对于树遍历等问题，递归方案通常很优雅，但可能消耗更多栈内存。
Iteration uses loops (for, while) to repeat a block of code. It generally has less overhead and uses a single stack frame, so it is more memory-efficient. Most recursive algorithms can be rewritten iteratively.
迭代使用循环（for、while）重复执行代码块。它通常开销较小，使用单个栈帧，因此内存效率更高。大多数递归算法都可以改写成迭代形式。
Choosing between them often depends on readability and the natural structure of the problem. For deeply nested problems, recursion may be much clearer even if slightly slower.
在两者之间选择通常取决于可读性和问题的自然结构。对于深度嵌套的问题，递归即使稍慢也可能清晰得多。
8. Linked List vs Array | 链表与数组
An array is a static, contiguous block of memory that stores elements of the same type. Access is fast via index (O(1)), but insertion or deletion can be slow because elements must be shifted (O(n)). Size is fixed at creation in some languages.
数组是一块静态、连续的内存块，存储相同类型的元素。通过索引访问速度很快 (O(1))，但插入或删除可能很慢，因为需要移动元素 (O(n))。某些语言在创建时大小固定。
A linked list consists of nodes, each containing data and a pointer to the next node. Memory is allocated dynamically. Insertion and deletion at a known position are O(1), but accessing an element by index requires traversal from the head (O(n)).
链表由节点组成，每个节点包含数据和一个指向下一个节点的指针。内存是动态分配的。在已知位置插入和删除为 O(1)，但通过索引访问元素需要从头遍历 (O(n))。
Use arrays when frequent random access is needed; use linked lists when frequent insertions/deletions occur and memory size is unpredictable.
需要频繁随机访问时使用数组；频繁插入/删除且内存大小不可预测时使用链表。
9. Hashing vs Encryption | 哈希与加密
Hashing transforms input data of any size into a fixed-size string of characters, called a hash digest. It is a one-way process – given a hash, the original data cannot feasibly be recovered. Used for verifying data integrity and storing passwords.
哈希将任意大小的输入数据转换为固定大小的字符序列，称为哈希摘要。这是一个单向过程——给定哈希值，原始数据几乎无法恢复。用于验证数据完整性和存储密码。
Encryption, conversely, is a two-way process. It scrambles data using an algorithm and key, producing ciphertext. Using the correct key, the original plaintext can be decrypted. Encryption ensures confidentiality.
相反，加密是双向过程。它使用算法和密钥扰乱数据，生成密文。使用正确的密钥可以将原始明文解密。加密确保机密性。
In short, hashing is for integrity and verification; encryption is for protecting secrecy and enabling reversible transformation.
简而言之，哈希用于完整性和验证；加密用于保护机密并实现可逆转换。
10. Lossy Compression vs Lossless Compression | 有损压缩与无损压缩
Lossless compression reduces file size without any loss of data. The original file can be perfectly reconstructed. Techniques like run-length encoding and Huffman coding are used for text and executable files.
无损压缩在不丢失任何数据的情况下减小文件大小。原始文件可以完美重建。游程编码和霍夫曼编码等技术用于文本和可执行文件。
Lossy compression permanently removes some data, especially details that are less noticeable to human perception. It achieves much higher compression ratios, used for images (JPEG), audio (MP3), and video. The original data cannot be perfectly recovered.
有损压缩会永久性地移除部分数据，尤其是人类感知不那么敏感的细节。它能实现高得多的压缩比，用于图像 (JPEG)、音频 (MP3) 和视频。原始数据无法完美恢复。
The choice depends on whether exact reproduction is mandatory. Lossless for crucial documents, lossy for media where slight quality drop is acceptable.
选择取决于精确再现是否必要。重要文档使用无损，媒体文件若轻微质量下降可接受则使用有损。
Published by TutorHao | AQA Computer Science Revision Series | aleveler.com
更多咨询请联系16621398022（同微信）
或截图后用「扫一扫 → 相册」选择本图