August 31, 2026
A journey from WAP, Symbian and Windows Mobile to Android, SaaS and AI — and the lessons I learned along the way.
In the early 2000s, the mobile landscape was dominated by WAP, the Wireless Application Protocol. It was a time of experimentation with constrained resources—limited processing power, minimal memory, and painfully slow data connections. Developers built applications using markup languages that felt primitive compared to today's standards, crafting experiences that worked within severe technical boundaries.
Symbian and Windows Mobile represented the pinnacle of mobile operating systems then. These platforms required deep technical knowledge and patience. Every byte mattered, and optimization wasn't just good practice—it was survival. The development process was methodical, with long iteration cycles and careful consideration of every design decision.
The introduction of the iPhone in 2007 and Android shortly after fundamentally changed everything. Suddenly, developers had access to powerful processors, abundant memory, and rich APIs. The mobile web evolved from WAP to full HTML5 support, enabling experiences previously impossible on desktop.
Application stores created new distribution models, while touch interfaces demanded entirely new interaction paradigms. Development shifted toward rapid iteration, with continuous deployment becoming standard practice. The cloud emerged as a critical infrastructure component, enabling scalable backends for mobile applications.
Cross-platform frameworks like React Native and Flutter addressed the challenge of maintaining multiple codebases. Developers could now reach both iOS and Android audiences with shared logic, though platform-specific nuances still required careful attention.
The rise of Software-as-a-Service transformed how applications were built and deployed. Cloud-native architectures, microservices, and containerization became standard. Developers gained access to sophisticated AI capabilities through APIs, enabling features like natural language processing, computer vision, and predictive analytics.
Serverless computing abstracted away infrastructure concerns, allowing teams to focus purely on application logic. The pace of innovation accelerated dramatically, with new frameworks, libraries, and services emerging regularly.
Looking back, the journey from WAP to AI reveals consistent patterns. Technical constraints drive innovation, user expectations continuously rise, and successful adaptation requires embracing change rather than resisting it. The tools have evolved, but the fundamental challenge remains: building software that solves real problems for real people.
Today's developers benefit from decades of accumulated knowledge, robust ecosystems, and powerful abstractions that handle complexity behind the scenes. Yet the core principles endure—understanding user needs, writing clean code, and maintaining security throughout the development lifecycle.
Further reading: https://medium.com/@ikilic
In July 2026 a 28.9-million-parameter LLM ran fully on-device on an $8 ESP32-S3 at almost 10 tokens per second. Here's how the trick works, who built
Aug 31, 2026
