06 January 2013

Thought for Food: the Bullipedia Project

Ferran Adria, the Catalan master chef who closed down his disruptive El Bulli restaurant in 2012, is embarking on a fascinating new culinary project that goes to the heart of Europe’s innovation agenda. After flirting with the molecular, Adria now appears to be going all-digital.

He is building a collaborative, interactive website called La BulliPedia that will offer an online wiki-style database containing every bit of (European?) gastronomic knowledge since the 17th century. His ultimate purpose is to provide a “creative archive” of resources rather than just recipes. He has backing from Telefonica Digital and the site is due to go live in 2015.

Here are three good reasons why this experiment will be worth watching:
  1. First, it will almost inevitably mean building a knowledge base about cooking food products that will link intelligently to other digital cooking resources: no wiki is an island. This in turn means building some kind of semantics of raw foodstuffs, chemical structures, menu ingredients, cooking methods and more to optimise searches over the database. The result will presumably be a gigantic gastronomy graph that could set a new standard for this domain.
  2. Second, it will need some clever translation functionality to make it optimally accessible to everyone. However, foodie language is exceptionally linked to terroir, locality, and cultural specifics. This raises the question of when to use translation, transcreation or borrowing as the appropriate tool, and how to automate this as much as possible. A fully language-neutral (i.e. what you speak is what you get) approach would offer a wonderful test case for a best-practice localisation/translation strategy.
  3. Third will BulliPedia be able to mesh with the recent announcement from IBM about the outcomes of its cognitive computing project on taste?
The system analyses foods in terms of how chemical compounds interact with each other, the number of atoms in each compound, and the bonding structure and shapes of compounds. Coupled with psychophysical data and models on which chemicals produce perceptions of pleasantness, familiarity and enjoyment, the end result is a unique recipe, using combinations of ingredients that are scientifically flavourful.
In a world of increasingly personalised healthcare, artificial cooking intelligence could in due course draw on the BulliPedia database of dishes to map specific culinary compositions to individual physical conditions. This would eventually help associate the recuperative virtues of good food more closely with the creative imaginations of great cooks.

¡Bon profit!

No comments:

Post a Comment