Experiment #2 – Surway, a social experiments game

After a first experiment combining 3D mapping and artificial intelligence, we are proud to present our newest creation: surway.

The first time, we were on familiar ground, drawing on many years of experience in web mapping. This time, we wanted to step out of our comfort zone and challenge ourselves to tackle a segment of development largely unexplored internally: video game development. At first glance, this may seem radically different from what we do on a daily basis building websites and web applications and in some ways it can be, depending on the technology and rendering engine chosen. Unreal Engine, for instance, uses C++ as its programming language, which is far removed from the languages we work with day to day. In our case, and despite the desire to push our limits, we opted for a more pragmatic approach given the eight days we had available, covering both onboarding and the design and development work.

We are first and foremost a web agency, and we are deeply committed to building within the web platform. It would therefore have made no sense to ask you to download our game on Steam just to try it out. We opted for a combined approach using TypeScript as our programming language, React to handle the user interface, Phaser as the game engine, and Grid Engine to manage physics and movement. With technologies we already knew in part, delivering a first version suddenly felt within reach.

Even though there is no artificial intelligence component in this experiment, none of it would have been possible without agentic development. The learning curve for Phaser alone would have eaten up all five available development days. So after setting up the technical architecture of the game, we applied the same process consistently for every feature: drafting a development plan (ARD), executing that plan with a development agent, and having the human in charge of the technical side review the code. This approach has the great advantage of enabling delivery without needing to fully master the project's underlying technical ecosystem. The downside, however, is that you still need solid development knowledge to be able to assess the output produced and the concepts applied.

During the research phase, we explored several directions, all converging around the idea of leaving a mark in a world that is both virtual and social, the web as we know it. Now, imagine the result of leaving a public notice board at the disposal of internet users with no heavy moderation process in place... The idea was to design a game with different levels where the player, from the end of the first level onwards, becomes aware that the actions of other players remain inscribed in the game. just like their own. We wanted to play with the notion that the player is no longer playing just for themselves, but also with the trace they will leave behind. Each level was designed to challenge the player, while also observing their behaviour, knowing that future players will see them, even if they eventually get lost in the crowd (or not).

Another challenge, as with any self-respecting web project, was to deliver a mobile experience as good as the desktop one. Physical keyboards and clunky on-screen buttons are out of the question; we implemented a fairly forgiving movement system that combines point-and-click and drag-and-drop. This allows the player to move smoothly using a single finger, in portrait mode, which is far more comfortable in practice.

The interface fully embraces its aesthetic bias: it revisits the visual language of early digital interfaces, the 2-bit era, where every pixel mattered and minimalism wasn't a stylistic choice but a technical constraint. No superfluous animations, just monospaced typography and a restricted color palette. The visual becomes almost a protocol, a neutral space that makes human choices all the more salient.

Antistatique is recognised for its institutional and media work, areas in which we have deep expertise. The idea here was to deliver an experience in a space where we are not yet known, at the opposite end of the spectrum from our usual projects. But also to demonstrate that with the web platform, anything is possible, and that we are capable of taking on and delivering anything that falls within this ecosystem full of promise and potential, even in areas where we are not yet known.