Skip to content
niles edited this page Jun 8, 2011 · 34 revisions

What is Clean Format?

  1. A standard data schema to be used with APIs and import / export tools to ensure all field names and data types / format are uniform.

Objects Defined:

  • restaurant
  • restaurant lists
  • menu
  • menu item
  • ingredient (raw material)
  • ingredient source (farms, suppliers)
  • ingredient packaged (products, grocery items)

Interconnect Types:

  • Free for all (open API calls)
  • Private (private API calls)
  • Paid (metered API calls)
  • Dump (full database / dataset definition)

2- In the future: A centralized key system for tracking unique ids for each record between various systems. For the time being, http://www.freebase.com/ will be used as the central key value store of choice.

Why is this being created?

Clean Format is the Food Data Rosetta Stone

This data format was created to support rapid application development with high data integrity. It was built to make hackathon projects easier to develop because of easy access to data and easy data mashups. Ultimately, once ratified by key industry players, it will be used as a database interconnect standard.

  • Problem 1 - CONNECTIVITY: Database schema varies greatly between different databases, some data must be kept private.
  • Problem 2 - COMPLETENESS: Duplicate data exists, lots of data is missing, and its hard to keep it up to date
  • Problem 3 - TRUST: Data is often not accurate, or is miscategorized, or is simply out of date

What can I use Clean Format for?

This is a free to use (and free to modify for your own use) format for sharing data between computer systems.

It will also be used as the basis for other initiatives which will take a more complete approach to addressing the problems above. In other words, just formatting the data won't solve all of our problems..

  • Import data dumps from another database to yours to "seed" a dataset

  • Subscribe to updates to keep your database up to date (and your modifications in tact)

  • Send updates from your database to another database

  • Ensure that the object in your database matches another (restaurant, menu, menu item, ingredient, etc)

What is the project Timeline?

Initial Draft Specification (and example hack) - Tech Crunch Disrupt & BarCampNYC6 - Late May 2011

Lessons Learned Briefing - June

First Full Application(s) - June / early July

Enhancements & Additions - August

Standard RC 1 - End of summer

Comment Period for Standard RC 1 - September

Standard v 1.0 - Fall 2011

The initial specifications will be based in part on existing industry standards (draft RDFa specs, schema.org, Open Graph, Open Menu) and in part on existing database schema of initial supporters of the format. Once a few applications have been created and we have a chance to address any performance, compatibility, legal or business concerns, we will then approach a targeted list of industry heavyweights to ratify and utilize the standard.

Who behind this project?

We're inviting broad industry participation in the creation of this standard interconnect format. If you have a personal interest in food data, or if your company has food data to share, feel free to get involved.

Add your name to this wiki list:

The initial concept was developed by Clean Plates http://www.cleanplates.com. Use the "message" link on this page for more details http://github.com/niles or niles at cl..pl.. dot com.