Skip to content


Switch branches/tags

Name already in use

A tag already exists with the provided branch name. Many Git commands accept both tag and branch names, so creating this branch may cause unexpected behavior. Are you sure you want to create this branch?

Latest commit


Git stats


Failed to load latest commit information.
Latest commit message
Commit time


A bench in a park, birds fly up above. In the background is a tree. - an open data repository for memorial benches.


All contributions are welcome. Before making a pull request, please:

  1. Raise a new issue describing the problem and how you intend to fix it.
  2. Submit a Pull Request referencing the Issue.

Open Data API

You can get all the data, or partial data, from the API. Data is returned in geoJSON format.


  • All Bench Data
    • That last / is required.
  • Specific Bench
  • Geographic Area (Haversine)
    • latitude and longitude in WGS 84.
    • radius in Kilometres.
    • results maximum number of benches returned. By defaults, 20 results are returned.
  • Tags
    • Returns all the benches with a specific tag.
  • Inscriptions
    • By default, the inscriptions are truncated to 128 characters.
    • To get the full inscriptions, append &truncated=false
  • Formats
    • By default, the JSON starts with var benches =
    • To get pure JSON, append &format=raw
  • Media
    • By default, the API doesn't return media.
    • To get media, append &media=true


  • All available folksonomy tags
    • That last / is required.
    • Returned in a format suitable for Select2.


  • All User Data
    • That last / is required.
  • Specific User
  • Formats
    • By default, the JSON starts with var users =
    • To get pure JSON, append &format=raw

Alexa Skill

There is an Alexa Skill which allows you to interact with the site via your voice. This functionality is provided by an API.

  • How many benches have been uploaded
  • The latest bench
  • Details of a random bench
  • Formats
    • By default, the JSON starts with var alexa =
    • To get pure JSON, append &format=raw

Running Locally

This is a simple PHP and MySQL website. No need for node, complicated deploys, or spinning up containerised virtual machines in the cloud.


  • PHP 7 or greater.
  • MySQL 5.5 or greater with innodb.
  • ImageMagick 6.9.4-10 or greater.

External APIs

You will need to sign up to some external API providers:

Add them to config.php.example - rename that to config.php

Database Structure

In the /database/ folder you'll find a sample database. All text fields are utf8mb4_unicode_ci because we live in the future now.

Hopefully, the tables are self explanatory:


  • benchID
  • latitude
  • longitude
  • address text representation generated by reverse geocoding. For example "10 Downing Street, London SW1A 2AA, United Kingdom"
  • inscription the text written on the bench
  • description placeholder. Might be used for comments about the bench.
  • present if a bench has been physically removed, this can be set to false.
  • published set to FALSE if the bench has been deleted
  • added datetime of when the bench was uploaded to the site
  • userID foreign key


Originally we were going to force people to sign in with Twitter / Facebook / GitHub. But that discourages use - so users are now pseudo-anonymous. Hence this weird structure!

  • userID
  • provider could be Twitter, GitHub, Facebook etc.
  • providerID user ID number on the provider's service. Anonymous users stores their IP address.
  • name their display name. Anonymous users stores the time they added a bench.


We store the original image - smaller images are rendered dynamically.

Media storage can be complicated. Storing thousands of images in a single directory can cause problems on some systems. To get around this, we calculate the SHA1 hash of each image. The image is stored in a subdirectory based on the hash. For example, if the hash is 1A2B3C, the file will be stored in /photos/1/A/1A2B3C.jpg

  • mediaID
  • benchID
  • userID
  • sha1 A hash of the file.
  • importURL If the image was imported from an external source - like Flickr.
  • licence The default is CC BY-SA 4.0, imported images may be different.
  • media_type We allow different types of photo - in the future, we might have other types of media.
  • width The image's width in pixels.
  • height The image's width in pixels.
  • datetime The date and time the image was created - based on EXIF metadata.
  • make The make of camera which took the photo - based on EXIF metadata.
  • model The model of camera which took the photo - based on EXIF metadata.

Media Types

At the moment, we only accept photos - of the inscription, the bench, the view from the bench, a panorama, and a VR photosphere.

  • shortName Internal ID.
  • longName Displayed to the user.
  • displayOrder When rendering a form in HTML, this determines the order they are presented in.


  • shortName Internal ID.
  • longName Displayed to the user.
  • url For more information.


Benches can be given multiple "tags". For example "cat" if the bench commemorates a feline, or "beach" if the bench is at the seaside.

Tagging uses the Toxi structure.

  • tags contains:

    • tagID a unique ID
    • tagText the displayed text
  • tag_map contains:

    • mapID a unique ID
    • tagID the ID of a tag
    • benchID the ID of a bench

Tags are hard-coded in the database and can't be added or edited by regular users.

Open Source Licenses

Everything we do builds on someone else's hard work.

And thanks to the many contributors who have improved this codebase.