In Georgia Tech's Fall 2016 semester I took a course where our final was to practice ethnographic research on a craft of our choice, then create a digital design intervention for that craft based on our research. To that end, I researched craft beer brewing and I designed a digital application for smartphones that helps beginner brewers design recipes. The app would use existing recipes from services that expert users design reciepes on and suggest ingredients based on the taste of the user and what ingredients expert brewers tend to use together.
This was my first experience doing design research before design practice, and my first experience creating a digital app that wasn't a game.
I ended up taking a course ran by a local brewery in Atlanta designed for beginners. On a cold winter day, I brewed a vanilla porter outside in a class of two. The other student and I were new brewers, and he and I both thought that the recipe crafting was the most interesting part. This was despite the fact that it was the most confusing too.
Personally, I found the math of balancing the various ingredients mathematically the hard problem. When experts design receipes, there's a wealth of knowledge they draw on when making decisions. New brewers don't have that. That knowledge barrier was what I wanted to try to solve for, and after some thought I realized that this is a problem that data might be able to solve.
I attempted to think through how features like Amazon's "Frequently Bought Together" leverage large datasets to filter good projects and use that in my project. It might not be foolproof, but if a user wants to make a Porter, and the majority of Porters in a given database who have ingredient A in them also have ingredient B, it's fair to assume they go well together. The same sort of logic could apply to the ratio of ingredients to use, though in that case formulas tend to dominate the discussion.
I didn't have as much time as I would have liked to actually build the prototype and I couldn't build a technical protoype because I'm not an expert in brewing. Despite those two facts, I like the fundemental idea underpinning this design and I'd be curious what an expert brewer might say about it.