FB2 was designed with a clear philosophical difference from EPUB: whereas EPUB packages multiple HTML files, CSS, and images into a ZIP container, FB2 is a single, self-contained XML document. It encodes a book’s structure—chapters, epigraphs, poems, footnotes—using a custom XML schema that prioritizes semantic tagging over visual presentation. This simplicity makes FB2 easy to generate and read in theory, but in practice, real-world FB2 files are riddled with non-compliant extensions, inline HTML fragments, and encoding issues. Enter fb2lib . As a dedicated parser library, its core function is to transform raw FB2 XML into a traversable, in-memory object model—whether in C, C++, or bindings to other languages. Unlike generic XML parsers (e.g., libxml2 or Xerces), fb2lib does more than just validate well-formedness. It understands FB2’s semantics: where an <epigraph> can appear, how a <poem> should be nested, and how to resolve internal links between <section> elements.
Yet the existence of fb2lib raises broader questions about software sustainability. Unlike EPUB, which is maintained by the W3C, FB2 has no official steering committee. The specification has drifted, with multiple forks (FB2 2.0, FB2 2.1, FB3) that are partially incompatible. fb2lib implementations often become forks themselves—maintained by lone volunteers for open-source readers like FBReader, CoolReader, or Calibre’s FB2 plugin. When a maintainer loses interest, the library stagnates. Newer tools then fall back to converting FB2 to EPUB using outdated regex-based scripts, which routinely strip poetry indentation or mangle tables. This fragility exposes a hidden cost of niche formats: their continued readability depends on the ongoing labor of a handful of developers who understand both the XML stack and the literary conventions embedded in the schema.
In the ecosystem of digital publishing, EPUB has become the de facto standard for e-books, supported by virtually every major platform from Apple Books to Kindle. Yet the digital world is not monolithic. In Russian-speaking and Eastern European online libraries, the FictionBook (FB2) format has maintained a loyal following for nearly two decades. At the heart of software that reads, converts, or validates FB2 files lies a small, unglamorous, but essential piece of code: fb2lib . While often overlooked in broader software engineering discussions, fb2lib serves as a powerful case study in the challenges of maintaining a domain-specific XML parsing library, the importance of format fidelity, and the quiet labor required to keep niche digital cultures alive.