My co-founder and I have spent over a decade building mixed reality projects and have been growing more and more frustrated with the process. From the number of tools needed, to client sign-off, to the complex esoteric tech stacks. And especially how slow the iteration loop is when dealing with interactivity and UX.
Two years ago we decided to stop whining and fix the fundamental issues. Ordinary Objects [0] was built with our core needs for AR prototyping: a very tight iteration loop between editor and real device, real-time interactivity while editing and clear and concise flow management + mapping. There are many things that layered on to make all of that possible: making it multi-user from the ground up, handling assets without a fuss, and building up a new interaction language.
From a technical standpoint we wanted to be native on the all of the platforms that we support, and do that as quickly as possible. Two years ago the best tool to achieve that was Unity, and I believe that is still the case today. Everything else is inside our custom C# Redux implementation. Our multi-user needs are very different from games, and it helped a lot to learn from Figma's technical notes to implement our pseudo eventual consistency setup. Its been super nice to be multi-user from the get go, we've been able to explore much more functionality this way.
Once the core churn eases up a bit more we will be open sourcing this particular C# Redux setup. As it has nothing to do with any engine code.
The website has some quick examples of how the design tool works [0]. But if you want to view a more complete prototype here is something Gregor, my co-founder, put together recently [1].
Over the past year we've been testing with closed groups, and have been excited by what everyone is making. Now we are ready to open it up for all of you to try! Give it a spin and let me know what you think! And happy to answer any questions here :)
[1]: https://www.linkedin.com/posts/onlinegregor_mixedreality-spo...