编程系统中数据的表示方法存在根本性的不一致性问题[1]。在Rust中，同一数据需要在代码、内存表示和打印输出中处理三种不同的记号方式[1]。JavaScript中的数据表示同样不统一：HashMap键在console.log中显示为字符串，但在Map对象中保持原类型[1]。相比之下，Clojure的EDN（Extensible Data Notation）设计使代码表示与打印表示完全一致，体现了统一记号的优势[1]。
为了改善这一状况，设计者们探索了多种解决方案[1]。分层设计方法（如Chrome HTML检查器所示）可以在保留文本表示的同时提供结构化GUI[1]。在标识符的选择上，ID相比指针具有语言独立性、可移植性和确定性的优点[1]。Unison编程语言通过采用全局唯一ID加本地名称的方式（name#id）平衡了可读性和稳定性[1]。此外，在数据模型中使用树形结构和ID来模拟图结构可以避免循环引用问题[1]。
Unison对"代码即数据"这一理念的处理最为深入，是唯一严肃对待此原则的编程语言，将代码存储为内容寻址数据库中的数据结构[1]。这种方法为统一数据和代码的表示提供了新的思路。
A 2022 discussion examines the fundamental challenge of data representation consistency across programming systems [1]. The core issue centers on how data appears in three distinct contexts—source code, memory representation, and debugging output—often requiring developers to navigate multiple notational systems simultaneously [1].
Rust exemplifies this fragmentation, where identical data must be handled through three separate notation approaches depending on context [1]. By contrast, Clojure's Extensible Data Notation (EDN) achieves complete alignment between code representation and printed output, providing a significant structural advantage [1]. JavaScript demonstrates similar inconsistencies: HashMap keys display as strings in console.log despite maintaining their original types within Map objects [1].
The analysis explores practical solutions to improve data legibility without sacrificing underlying structure [1]. Hierarchical design patterns, such as Chrome's HTML inspector, maintain textual representation while offering structured graphical interfaces [1]. The distinction between pointers and identifiers proves conceptually important—identifiers offer language independence, portability, and determinism but require additional contextual information for interpretation [1]. Unison addresses this trade-off through a hybrid approach combining globally unique identifiers with local names in the format name#id, balancing readability against stability [1].
Tree structures with identifier references can model graph relationships while avoiding circular reference complications [1]. The discussion also positions code itself as data, noting that Unison stands as the primary language treating "code as data" seriously by storing code as data structures within a content-addressed database [1].