Final Thesis: Optimization of Content Delivery in Web Applications

Abstract: This thesis improves content delivery for QDAcity, a localized React single-page application, by enabling incremental localization alongside ongoing feature development and systematic management of the translation
status. It introduces a selective build-time prerendering pipeline for a subset of routes, while keeping dynamic pages client-rendered, improving build performance without sacrificing user experience. Localization is refactored into route-scoped namespaces with on-demand message loading, caching, deduplication, and predictive prefetching,
so new languages can be shipped incrementally with minimal payload and fast navigation. The translation lifecycle is integrated via Weblate and CI automation, treating each namespace as stateful (missing, translated, approved) to govern publication, fallbacks, and user-facing disclosure for unreviewed translations. The system is evaluated against the baseline in terms of build efficiency, Core Web Vitals, translation bundle size, and correctness, providing a scalable workflow for multilingual web applications.

Keywords: Web App, Content Delivery, I18N, SSG

PDF: Master Thesis

Reference: Tobias Schmid. Optimization of Content Delivery in Web Applications. Master Thesis. Friedrich-Alexander-Universität Erlangen-Nürnberg: 2026.