Food Ecosystem

Restaurants

Beyond ideal groceries: what would an ideal food location look like? Research and concept only here, no new scoring engine and no directory app yet. Three parts: what already exists for restaurant discovery, how the values frame from ideal groceries extends to eating out, and a chef-friendly slow food model built on a single, pre-agreed menu.

What exists

Restaurant discovery services researched for open-source status, license, coverage and the gap versus a values-scored ecosystem vision. Two are open source and OSM-based (a real foundation to build on); the rest are proprietary directories or reservation platforms with no values criteria at all.

HappyCow

proprietary

Global directory and app for vegan, vegetarian and veg-friendly restaurants: over 240,000 listings in 185 countries with user reviews and photos.

Gap: Diet-filter only. No values scoring for ownership, regional sourcing, seasonality or provenance, and no open dataset behind the proprietary listings.

Replicate as OSS: not a fit, Its value is 25 years of user-submitted reviews and photo moderation, a community-building problem, not a values-scoring gap. Better to link out than rebuild.

TheFork

proprietary

Pan-European restaurant discovery and reservation platform (a Tripadvisor company): browse by cuisine, price and reviews, book a table directly.

Gap: Reservation convenience only. No values criteria (ownership, sourcing, seasonality, traceability), and restaurant data is not portable outside the platform.

Replicate as OSS: not a fit, Booking and payment integration with restaurant POS systems is a commercial-integration problem outside this project's scope, not worth rebuilding for a values-scoring layer.

Google Maps / Places

proprietary

De facto default for restaurant discovery worldwide: ratings, photos, hours and reviews via Google Maps and the Places API.

Gap: No values criteria at all. Places API terms explicitly forbid caching results into a competing or derivative places database.

Replicate as OSS: not a fit, Terms of service forbid derivative datasets built from the API. Any replication has to source location data independently, e.g. from OpenStreetMap, not from Google.

Slow Food Austria: Bündnis der Köche (Chefs' Alliance) and Genussführer

proprietary

Editorial, non-commercial guide (Genussführer) plus the Carinthia-led Slow Food Bündnis der Köche: curated restaurant and inn listings vetted for authentic craft, with named producers on the menu.

Gap: Curated and trustworthy but regional (Carinthia-led) and small. Not a structured open dataset: inclusion is editorial, not per-criterion scored.

Replicate as OSS: candidate, The named-producer-on-menu vetting practice is exactly the provenance signal the values frame wants. Worth encoding as structured, filterable fields rather than prose listings.

Organic Maps

open source (Apache-2.0)

Free, offline-first Android/iOS map app built on crowd-sourced OpenStreetMap data, browsable by amenity tag including amenity=restaurant. No ads, no tracking.

Gap: Shows what OSM knows (name, cuisine, opening hours) but nothing about ownership structure, sourcing or values criteria. OSM restaurant tagging is sparse outside dense cities.

Replicate as OSS: candidate, Best open-data foundation for a values-scored restaurant layer: Apache-2.0 code, ODbL map data, no proprietary lock-in. Needs a values-scoring overlay, not a new map.

OsmAnd

open source (GPL-3.0)

Offline navigation and map app on OpenStreetMap data with POI search including restaurants, on Android and iOS, widely used as an OSM-data reference client.

Gap: Same OSM-data ceiling as Organic Maps: no values scoring, sourcing or ownership data, and OSM restaurant coverage in Austria outside Vienna and Graz is uneven.

Replicate as OSS: not a fit, Organic Maps already covers the same OSM-data-plus-open-license niche with a simpler codebase. No reason to build against two OSS map bases when one anchor suffices.

Ideal food locations

Groceries already have a 14-criterion values frame in the ideal-groceries tool, scoring ownership structure, regional sourcing, seasonality, organic status, packaging, fair labor and more, folded together with a traced provenance chain. No new engine here, the same criteria simply mean something slightly different for a restaurant than for a product on a shelf:

  • Ownership: independent, owner-operated kitchen versus franchise or chain, the same axis the tool already scores for retailers.
  • Regional: ingredients sourced from the surrounding region rather than a centralized national supply chain, visible as named local producers on the menu.
  • Seasonal: the menu itself changes with what is in season, rather than an identical printed menu running year round.
  • Organic: certified organic ingredients where available, same certification marks the grocery tool already recognizes.
  • Traceability: can the kitchen name the farm or producer behind a dish, the restaurant equivalent of a product's provenance chain, with the same honest gap where a stage is unknown.

A restaurant version of the tool would score a menu, or a kitchen's sourcing practice, the way the current tool scores a product. That scoring engine is not built here, this section only establishes that the same 14 criteria transfer.

Slow food, one menu

The chef-friendly model: a single, pre-agreed menu per day or per week, agreed with guests in advance through subscription or RSVP rather than an on-the-night a la carte choice. This is not a new invention, it is the pattern behind several existing formats:

  • Table d'hôte (French, "the host's table"): a fixed, multi-course menu at a fixed price with a short choice per course, the historic ancestor of the format, running in inns and guesthouses since the seventeenth century (Wikipedia, 2026, [C]).
  • Omakase (Japanese, "I leave it to you"): the chef selects every dish from the freshest available ingredients, often on a set menu that changes frequently, which the Michelin Guide notes keeps prices down and reflects the chef's own philosophy for how the meal should unfold (Michelin Guide, 2026, [B]).
  • Supper clubs: participatory, subscription or RSVP-based dinners where guests commit before they know the exact menu, a format press coverage describes as a breakout dining trend for 2026 (TableMesh, 2026, [B]).

Why it is nice for chefs:

  • No menu sprawl: one dish list to prep and plate well, not a card of thirty options held in reserve.
  • Planned purchasing: tomorrow's covers are known today (via RSVP or subscription), so the shopping list is exact, not a forecast.
  • Near-zero waste: nothing bought "just in case" for an a la carte item that might not sell.
  • Real mise en place: prep matches one menu instead of hedging across many, freeing time for craft rather than inventory management.

How it pairs with values scoring and provenance: a single pre-agreed menu is also the easiest case to score. One dish list, one set of named producers, one seasonal snapshot, so the same ownership, regional, seasonal, organic and traceability criteria above apply cleanly to a whole sitting rather than to hundreds of possible a la carte combinations.

Existing exemplars in Austria: the Carinthia-led Slow Food Bündnis der Köche (Chefs' Alliance) asks its member chefs to name the producers on their menus, verified members include Die Forelle (chef Hannes Müller), Restaurant Moritz (chef Roman Pichler) and der daberer.das biohotel (chef Florian Bucar), among others (Slow Food Foundation for Biodiversity, 2026, [B]; Slow Food International, 2026, [B]). This is a values-alignment signal, not necessarily a one-menu format on its own, most Bündnis members still run an a la carte card alongside seasonal specials.

Scope note

This page is research and concept only. A restaurant directory, a working values-scoring engine for eating out, and a booking flow for pre-agreed menus are all out of scope for this prototype and belong to a later wave. What exists today is the landscape survey above and the two concept specs; nothing here has been built as a working tool.