Contributing

tl;dr

🙋 How to help:

npm run build will reprocess the files and output warnings
Resolve warnings - show me
Remove generic names - show me
Add brand:wikidata and brand:wikipedia tags - show me
Add missing brands - show me
Edit Wikidata - show me

Tip: You can browse the index at https://nsi.guide/ to see which brands are missing Wikidata links, or have incomplete Wikipedia pages.

🚫 Don't edit the files in `dist/` - they are generated:

dist/names_all.json - all the frequent names and tags collected from OpenStreetMap
dist/names_discard.json - subset of names_all we are discarding
dist/names_keep.json - subset of names_all we are keeping
dist/wikidata.json - cached brand data retrieved from Wikidata

✅ Do edit the files in `config/` and `brands/`:

config/*
- config/filters.json- Regular expressions used to filter names_all into names_keep / names_discard
- config/match_groups.json- Groups of tag pairs that are considered equal when matching
brands/* - Config files for each kind of branded business, organized by OpenStreetMap tag
- brands/amenity/*.json
- brands/leisure/*.json
- brands/shop/*.json
- brands/tourism/*.json

Background

🗺️ About OpenStreetMap

OpenStreetMap is a free, editable map of the whole world that is being built by volunteers. Features on the map are defined using tags. Each tag is a key=value pair of text strings.

For example, a McDonald's restaurant might have these tags:

  "amenity": "fast_food"
  "cuisine": "burger"
  "name": "McDonald's"
  ... and more tags to record its address, opening hours, and so on.

💡 About the name-suggestion-index

The goal of this project is to define the most correct tags to assign to each common brand name. This helps people contribute to OpenStreetMap, because they can pick "McDonald's" from a list and not need to worry about the tags being added.

🗃️ About the brand files

The brands/* folder contains many files, which together define the most correct OpenStreetMap names and tags.

These files are created by a several step process:

Process the OpenStreetMap "planet" data to extract common names -> dist/names_all.json
Filter all the names into -> dist/names_keep.json and dist/names_discard.json
Merge the names we are keeping into -> brands/**/*.json files for us to decide what to do with them

The files are organized by OpenStreetMap tag:

brands/* - Config files for each kind of branded business, organized by OpenStreetMap tag
- brands/amenity/*.json
- brands/leisure/*.json
- brands/shop/*.json
- brands/tourism/*.json

Each brand entry looks like this (comments added for clarity):

In brands/amenity/fast_food.json:

  "amenity/fast_food|McDonald's": {         // Identifier like "key/value|name"
    "tags": {                               // "tags" - OpenStreetMap tags that every McDonald's should have
      "amenity": "fast_food",               //   The OpenStreetMap tag for a "fast food" restaurant
      "brand": "McDonald's",                //   `brand` - Brand name in the local language (English)
      "brand:wikidata": "Q38076",           //   `brand:wikidata` - Universal Wikidata identifier
      "brand:wikipedia": "en:McDonald's",   //   `brand:wikipedia` - Reference to English Wikipedia
      "cuisine": "burger",                  //   `cuisine` - What kind of fast food is served here
      "name": "McDonald's"                  //   `name` - Display name, also in the local language (English)
    }
  },

There may also be entries for McDonald's in other languages!

  "amenity/fast_food|マクドナルド": {         // Identifier like "key/value|name"
    "countryCodes": ["jp"],                 // Optional `countryCodes` - array of countries where this entry is used
    "tags": {
      "amenity": "fast_food",
      "brand": "マクドナルド",                // `brand` - Brand name in the local language (Japanese)
      "brand:en": "McDonald's",             // `brand:en` - For non-English brands, tag the English version too
      "brand:ja": "マクドナルド",             // `brand:ja` - Add at least one `brand:xx` tag that matches `brand`
      "brand:wikidata": "Q38076",           // `brand:wikidata` - Same Universal wikidata identifier
      "brand:wikipedia": "ja:マクドナルド",   // `brand:wikipedia` - Reference to Japanese Wikipedia
      "cuisine": "burger",
      "name": "マクドナルド",                 // `name` - Display name, also in the local language (Japanese)
      "name:en": "McDonald's"               // `name:en` - For non-English names, tag the English version too
      "name:ja": "マクドナルド",              // `name:ja` - Add at least one `name:xx` tag that matches `name`
    }
  },

Optional properties

`matchNames`/`matchTags`

Sometimes a single brand will have different tags in OpenStreetMap.

For example, we prefer "Bed Bath & Beyond" to be tagged as shop/houseware. However we also need to recognize:

less-preferred name spellings like "Bed Bath and Beyond"
less-preferred tag pairs like shop/department_store

Rather than adding every possible alternative to the index, we can use matchNames and matchTags properties to ignore less-preferred alternatives.

  "shop/houseware|Bed Bath & Beyond": {
    "matchNames": ["bed bath and beyond"],     // also match these alternate spellings
    "matchTags": ["shop/department_store"],    // also match these alternate taggings
    "tags": {
      "brand": "Bed Bath & Beyond",
      "brand:wikidata": "Q813782",
      "brand:wikipedia": "en:Bed Bath & Beyond",
      "name": "Bed Bath & Beyond",
      "shop": "houseware"
    }
  },

The matching code also has some useful automatic behaviors:

Names are always matched case insensitive, with spaces and punctuation removed. You do not need to add matchNames properties for simple name variations.
Some tags are assigned to match groups (defined in config/match_groups.json). You don't need add matchTags: ["shop/doityourself"] to every "shop/hardware" and vice versa. Tags in a match group will match any other tags in the same match group.

`nomatch`

Sometimes there are multiple different entries that use the same name.

For example, "Sonic" can be either a fast food restaurant or a fuel station.

We want to allow both kinds of "Sonic" to exist in the index, and we don't want to be warned that they are potentially duplicate, so we can add a nomatch property to each entry to suppress the "duplicate name" warning.

  "amenity/fuel|Sonic": {
    "nomatch": ["amenity/fast_food|Sonic"],
    ...
  },
  "amenity/fast_food|Sonic": {
    "nomatch": ["amenity/fuel|Sonic"],
    ...
  },

Identical names, multiple brands

Sometimes multiple brands with the same name will operate in geographically distinct locations. You can modify the key to include a tilde ~ after the name to tell the difference between two otherwise identical brands. The text after the tilde can contain anything.

When using a tilde ~ name:

You should add "nomatch": properties to each name so they do not generate duplicate name warnings.

  "shop/supermarket|Price Chopper~(Kansas City)": {
    "countryCodes": ["us"],
    "nomatch": [
      "shop/supermarket|Price Chopper~(New York)"
    ],
    "tags": {
      "brand": "Price Chopper",
      "brand:wikidata": "Q7242572",
      "brand:wikipedia": "en:Price Chopper (supermarket)",
      "name": "Price Chopper",
      "shop": "supermarket"
    }
  },
  "shop/supermarket|Price Chopper~(New York)": {
    "countryCodes": ["us"],
    "nomatch": [
      "shop/supermarket|Price Chopper~(Kansas City)"
    ],
    "tags": {
      "brand": "Price Chopper",
      "brand:wikidata": "Q7242574",
      "brand:wikipedia": "en:Price Chopper Supermarkets",
      "name": "Price Chopper",
      "shop": "supermarket"
    }
  },

What you can do

🏗️ Building the project

To rebuild the index, run:

npm run build

This will output a lot of warnings, which you can help fix!

🤔 Resolve warnings

Warnings mean that you need to edit files under brands/*. The warning output gives a clue about how to fix or suppress the warning. If you aren't sure, just ask on GitHub!

Duplicate names

Warning - Potential duplicate brand names:
To resolve these, remove the worse entry and add "matchNames"/"matchTags" properties on the better entry.
To suppress this warning for entries that really are different, add a "nomatch" property on both entries.
  "shop/supermarket|Carrefour" -> duplicates? -> "amenity/fuel|Carrefour"
  "shop/supermarket|VinMart" -> duplicates? -> "shop/department_store|VinMart"

What it means: These names are commonly tagged differently in OpenStreetMap. This might be ok, but it might be a mistake.

For "VinMart" we really prefer for it to be tagged as a supermarket. It's a single brand frequently mistagged.

Add "matchTags": ["shop/department_store"] to the (preferred) "shop/supermarket|VinMart" entry
Delete the (not preferred) entry for "shop/department_store|VinMart"

For "Carrefour" we know that can be both a supermarket and a fuel station. It's two different things.

Add "nomatch": ["shop/supermarket|Carrefour"] to the "amenity/fuel|Carrefour" entry
Add "nomatch": ["amenity/fuel|Carrefour"] to the "shop/supermarket|Carrefour" entry

Existing tagging (you can compare counts in dist/names_keep.json), information at the relevant Wikipedia page or the company's website, and OpenStreetMap Wiki tag documentation all help in deciding whether to match or nomatch these duplicates.

If the situation is unclear, one may contact the local community and ask for help.

🔪 Remove generic names

Some of the common names in the index might not actually be brand names. We want to remove these generic words from the index, so they are not suggested to mappers.

For example, "Универмаг" is just a Russian word for "Department store":

  "shop/department_store|Универмаг": {
    "tags": {
      "brand": "Универмаг",
      "name": "Универмаг",
      "shop": "department_store"
    }
  },

To remove this generic name:

Delete the item from the appropriate file, in this case brands/shop/department_store.json
Edit config/filters.json. Add a regular expression matching the generic name in either the discardKeys or discardNames list.
Run npm run build - if the filter is working, the name will not be put back into brands/shop/department_store.json
git diff - to make sure that the entries you wanted to discard are gone (and no others are affected)
If all looks ok, submit a pull request with your changes.

🕵️‍♀️ Add wiki tags

Adding brand:wikipedia and brand:wikidata tags is a very useful task that anybody can help with.

Example #1 - Worldwide / English brands...

Find an entry in a brand file that is missing these tags:

In brands/amenity/fast_food.json:

  "amenity/fast_food|Chipotle": {
    "matchNames": ["chipotle mexican grill"],
    "tags": {
      "amenity": "fast_food",
      "brand": "Chipotle",
      "cuisine": "mexican",
      "name": "Chipotle"
    }
  },

Google for that brand - if you are lucky, you might find the Wikipedia page right away.

From the Wikipedia page URL, you can identify the brand:wikipedia value.

OpenStreetMap expects this tag to be formatted like "en:Chipotle Mexican Grill".

Copy the page name from the URL.
Add the language prefix - "en:" for the English Wikipedia.
Replace the underscores '_' with spaces.

On the brand's Wikipedia page, you can also find its "Wikidata item" link. This appears under the "tools" menu in the sidebar.

👉 protip: @maxerickson has created a user script to make copying these values even easier - see #1881

On the brand's Wikidata page, you can identify the brand:wikidata value. It is a code starting with 'Q' and several numbers.

Update the brand file, in this case brands/amenity/fast_food.json:

We can add the "brand:wikipedia" and "brand:wikidata" tags.

  "amenity/fast_food|Chipotle": {
    "matchNames": ["chipotle mexican grill"],
    "tags": {
      "amenity": "fast_food",
      "brand:wikidata": "Q465751",                            // added
      "brand:wikipedia": "en:Chipotle Mexican Grill",         // added
      "brand": "Chipotle",
      "cuisine": "mexican",
      "name": "Chipotle"
    }
  },

(comments added for clarity)

Rebuild and submit a pull request.

Run npm run build
If it does not fail with an error, you can submit a pull request with your changes (warnings are OK).

Example #2 - Regional / non-English brands...

This example uses a brand "かっぱ寿司". I don't know what that is, so I will do some research.

Find an entry in a brand file that is missing these tags:

In brands/amenity/fast_food.json:

  "amenity/fast_food|かっぱ寿司": {
    "tags": {
      "amenity": "fast_food",
      "brand": "かっぱ寿司",
      "name": "かっぱ寿司"
    }
  },

Google for that brand - if you are lucky, you might find the Wikipedia page right away.

Tip: You might want to narrow you search by Googling with a site: filter: "かっぱ寿司 site:ja.wikipedia.org"

From these results, we can know that the brand is "Kappazushi", owned by a Japanese company called "Kappa Create". We can also find the Wikipedia page.

As with English brands, you can identify the brand:wikipedia value from the URL. Because this is a Japanese brand, we will link to the Japanese Wikipedia page.

OpenStreetMap expects this tag to be formatted like "ja:かっぱ寿司".

Copy the page name from the URL.
Add the language prefix "ja:".
Replace the underscores '_' with spaces.

Although I can not read Japanese, I can identify the "Wikidata item" link because it always appears in the sidebar and mouseover will show the Wikidata 'Q' code in the URL.

On the brand's Wikidata page, you can identify the brand:wikidata value. It is a code starting with 'Q' and several numbers.

Note: The Wikidata page looks a bit sparse - you can edit this too if you want to help!

Update the brand file, in this case brands/amenity/fast_food.json:

We can add:

"brand:en" and "name:en" tags to contain the English name "Kappazushi"
"name:ja" and "brand:ja" tags to contain the local name "かっぱ寿司"
"brand:wikipedia" and "brand:wikidata" tags
"cuisine": "sushi" OpenStreetMap tag
"countryCodes" property, to indicate that this brand should only be used in Japan.

  "amenity/fast_food|かっぱ寿司": {
    "countryCodes": ["jp"],                // added
    "tags": {
      "amenity": "fast_food",
      "brand": "かっぱ寿司",
      "brand:en": "Kappazushi",            // added
      "brand:ja": "かっぱ寿司",            // added
      "brand:wikipedia": "ja:かっぱ寿司",     // added
      "brand:wikidata": "Q11263916",        // added
      "cuisine": "sushi",                   // added
      "name": "かっぱ寿司",
      "name:en": "Kappazushi",              // added
      "name:ja": "かっぱ寿司"               // added
    }
  },

(comments added for clarity)

Rebuild and submit a pull request.

Run npm run build
If it does not fail with an error, you can submit a pull request with your changes (warnings are OK).

🏪 Add missing brands

If it exists, we want to know about it!

Some brands aren't mapped enough (50+ times) to automatically be added to the index so this is a valuable way to get ahead of incorrect tagging.

Before adding a new brand, the minimum information you should know is the correct tagging required for instances of the brand (name, brand and what it is - e.g. shop=food). Ideally you also have brand:wikidata and brand:wikipedia tags for the brand and any other appropriate tags - e.g. cuisine.
Add your new entry anywhere into the appropriate file under brands/* (the files will be sorted alphabetically later) and using the "tags" key add all appropriate OSM tags. Refer to here if you're not familiar with the syntax.
If the brand only has locations in a known set of countries add the "countryCodes": [] key to your new entry. This takes an array of ISO 3166-1 alpha-2 country codes in lowercase (e.g. ["de", "at", "nl"]).
If instances of this brand are commonly mistagged add the "matchNames": [] key to list these. Again, refer to here for syntax.
Run npm run build and resolve any duplicate name warnings.

Using Overpass Turbo

Sometimes you might want to know the locations where a brand name exists in OpenStreetMap. Overpass Turbo can show them on a map:

Go to https://overpass-turbo.eu/
Enter your query like this, replacing the name and other OpenStreetMap tags. Because we don't specify a bounding box, this will perform a global query.

nwr["name"="かっぱ寿司"]["amenity"="fast_food"];
out center;

Tip: The browsable index at https://nsi.guide/ can open Overpass Turbo with the query already set up for you.

Click run to view the results.

As expected, the "かっぱ寿司" (Kappazushi) locations are all concentrated in Japan.

📝 Edit Wikidata

Editing brand pages on Wikidata is something that anybody can do. It helps not just our project, but anybody who uses this data for other purposes too! You can read more about contributing to Wikidata here.

Add Wikidata entries for brands that don't yet have them.
Improve the labels and descriptions on the Wikidata entries.
Translate the labels and descriptions to more languages.
Add social media accounts under the "Identifiers" section. If a brand has a Facebook, Instagram, or Twitter account, we can fetch its logo automatically.

Tip: The browsable index at https://nsi.guide/ can show you where the Wikidata information is missing or incomplete.

Adding properties to Wikidata

Social media accounts may be used to automatically fetch logos, what is used by iD

Adding references to Wikidata

Entries without matching Wikipedia article must have some references by independent sources. For our entries usually the easiest one to add is something in form of "this shop brand had N shops on some specific date".

Creating Wikidata entries

For minor brands there may be no Wikipedia article and it may be impossible to create one. In such cases one may still go to Wikidata and select "Create a new item" in menu. For such entries it is mandatory to add some external identifier or references (see section above with animation showing how it can be done).

Provide feedback

Saved searches

Use saved searches to filter your results more quickly

CONTRIBUTING.md

CONTRIBUTING.md

Contributing

tl;dr

🙋 How to help:

🚫 Don't edit the files in `dist/` - they are generated:

✅ Do edit the files in `config/` and `brands/`:

Background

🗺️ About OpenStreetMap

💡 About the name-suggestion-index

🗃️ About the brand files

Optional properties

`matchNames`/`matchTags`

`nomatch`

Identical names, multiple brands

What you can do

🏗️ Building the project

🤔 Resolve warnings

Duplicate names

🔪 Remove generic names

🕵️‍♀️ Add wiki tags

Example #1 - Worldwide / English brands...

Example #2 - Regional / non-English brands...

🏪 Add missing brands

Using Overpass Turbo

📝 Edit Wikidata

Adding properties to Wikidata

Adding references to Wikidata

Creating Wikidata entries

Files

CONTRIBUTING.md

Latest commit

History

CONTRIBUTING.md

File metadata and controls

Contributing

tl;dr

🙋 How to help:

🚫 Don't edit the files in dist/ - they are generated:

✅ Do edit the files in config/ and brands/:

Background

🗺️ About OpenStreetMap

💡 About the name-suggestion-index

🗃️ About the brand files

Optional properties

matchNames/matchTags

nomatch

Identical names, multiple brands

What you can do

🏗️ Building the project

🤔 Resolve warnings

Duplicate names

🔪 Remove generic names

🕵️‍♀️ Add wiki tags

Example #1 - Worldwide / English brands...

Example #2 - Regional / non-English brands...

🏪 Add missing brands

Using Overpass Turbo

📝 Edit Wikidata

Adding properties to Wikidata

Adding references to Wikidata

Creating Wikidata entries

🚫 Don't edit the files in `dist/` - they are generated:

✅ Do edit the files in `config/` and `brands/`:

`matchNames`/`matchTags`

`nomatch`