We intend to provide an open set of specifications, tools and services for generating verifiable badges that users can take with them wherever they go and use however they like.
The latest open standard we released can be found in the latest assertion specification. The assertion includes the open standard, the metadata specification, we defined.
For more information, check out http://openbadges.org
Requirements:
- Webserver capable of serving requests to the general internet.
- Ability to make a POST request from your server backend and read a JSON response.
- Email addresses of the users you wish to issue badges.
- Badge image must be in PNG format.
Usage example:
- Generate an assertion (see below) for the user recieving the badge.
- Store that assertion at a public-but-secret URL and serve it with
content-type: application/json
-
The assertion contains private information about a user, so you want a non-predictable URL scheme to prevent automated scraping.
-
This URL should be stable - any badge issued from it relies on its existence for verification.
-
Both of these problems will be solved in the near-term future by supporting signed assertions, so you'll only need to expose a URL containing your public key.
- Make a POST request to the open badge creator with the assertion URL. If
validation passes, you will receive an HTTP 200 with
content-type: image/png
, the body being a yourbadge.image
with the assertion URL baked into it. - Send/give the image to the user (for example, email it).
We have an easy to use API built for Issuers to easily push badges into Users Backpacks, giving the User the ability to approve the push through a lightboxed modal. The API is written in Javascript, and is includable in your project with just a few lines of JS. Full documentation is available here.
Please see the documentation on Assertions to learn how to format your assertions, and [see the documentation on Badge Baking][baking-latest] to learn more about how to use the baking API and what kind of responses to expect in case of error.
The easy way*
Use Vagrant. vagrant up
in the project root will spin
up a fully provisioned VM (it'll take about two or three minutes, longer if
you don't have a lucid32
box), vagrant ssh
to get into the VM, then
start-server
will start up the server at
http://localhost:8888. The server will also watch for
changes, so you don't have to manually reload it.
Install Vagrant and VirtualBox. Vagrant will try to install in C:\vagrant, which is a protected location in Windows. Instead, tell it to install in C:\Program Files (if you're on 32 bit windows) or C:\Program Files (x86) (if you're on 64 bit windows) instead - don't worry, this will actually create a C:\Program Files\vagrant folder.
Ensure that your PATH variable contains the VirtualBox binaries folder: Go to your Control Panel -> System -> Advanced system settings -> "Environment Variables" button -> in the system variable section scroll down to the PATH variable, hit "edit" and add C:\Program Files\VirtualBox. After making sure this is the case, in the openbadges repo, run:
C:\...\Openbadges>"c:\Program Files (x86)\Vagrant\bin\vagrant.bat" up
This will do all the VM building. You may be prompted by the windows firewall and UAC to allow VirtualBox network and disk access: you'll have to allow this, otherwise things won't work. When the VM creation is done, let vagrant discover that it can't actually do SSH on windows:
C:\...\Openbadges>"c:\Program Files (x86)\Vagrant\bin\vagrant.bat" ssh
This will generate an output similar to the following:
`vagrant ssh` isn't available on the Windows platform. You are still able
to SSH into the virtual machine if you get a Windows SSH client (such as
PuTTY). The authentication information is shown below:
Host: 127.0.0.1
Port: 2229
Username: vagrant
Private key: C:/Users/You/.vagrant.d/insecure_private_key
If you don't already have it installed, get PuTTY or another SSH client and log into 127.0.0.1:2229 with the username "vagrant" and password "vagrant", using the SSH key that vagrant generated in the indicated directory (Connection -> SSH -> Auth -> "browse" button). Putty will expect a .ppk file, so just tell it to show all files (.) and select the insecure_private_key file. Connect, and you should get into the Vagrant VM just fine.
Once connected to the VM, you can now start the server using:
vagrant@lucid32:~$ start-server
and then back in windows you can fire up your favourite browser and connect to the OpenBadges server you now have running on localhost:
http://localhost:8888
-
Setup a MySQL database. Create a database and a user with full privileges on that db. For example:
CREATE DATABASE openbadges; GRANT ALL PRIVILEGES ON openbadges.* TO badgemaker@localhost IDENTIFIED BY 'secret'; CREATE DATABASE test_openbadges; GRANT ALL PRIVILEGES ON test_openbadges.* to badgemaker@localhost IDENTIFIED BY 'secret';
-
Copy the
openbadges/lib/environments/local-dist.js
toopenbadges/lib/environments/local.js
and edit the configuration to match your local development environment. The MySQL database credentials should match step #1. For example:database: { driver: 'mysql', host: '127.0.0.1', user: 'badgemaker', password: 'secret', database: 'openbadges' },
-
Install external tools:
- PhantomJS: We use PhantomJS for running unit tests. On a debian based Linux system you can run
sudo apt-get install phantomjs
to install and runphantomjs --version
to check it is installed. For other systems you can try downloading and installing it or building it from source.
-
Install local dependencies:
npm install
-
Install submodules:
git submodule update --init
-
Run the test suite:
npm test
-
Start your server:
npm start
No matter which way you choose, you should join the Open Badges Google Group. If you have any problems setting up the environment, feel free to post a message to the list.
I like to be able to use http://openbadges.local for accessing the
project. Assuming you used vagrant, you can change the hostname in local.js
and do sudo echo "33.33.33.11 openbadges.local" >> /etc/hosts
to make it
happen. If you're on OS X, you can also use
Gas Mask for temporary hosts file switching
rather than having to manually edit /etc/hosts
If you need to modify the database schema, you'll want to create a migration. You can do this as follows:
-
Come up with an alphanumeric name for your migration, e.g.
add-issuer-column
. -
Run
./bin/db-migrate create add-issuer-column
. This will create a new JS file preixed with a timestamp in themigrations
directory. Something like the following should be displayed:[INFO] Created migration at migrations/20130213205310-add-issuer-column.js
-
Edit the new JS file as per the node-db-migrate instructions.
-
Try out your migration using
./bin/db-migrate up
. -
Try rolling back your migration using
./bin/db-migrate down
.
And finally, note that during development, npm start
automatically runs
./bin/db-migrate up
for you. For production use, you'll need to manually
run this command yourself whenever you deploy changes that involve a
schema change.
If you want to write tests for your migration, check out
test/migration.test.js
for inspiration.
The codebase behaves slightly differently when run in an environment where
environment variable NODE_ENV=production
. These differences include:
- less verbose logging
- using precompiled templates for client-side rendering
- run
bin/template-precompile
to generate
- run
- "Test Site" banner will not show in the UI
- https://github.com/lmorchard/django-badger -- Issuing app for Django
- https://github.com/fedora-infra/tahrir -- A Pyramid (Python) app for issuing your own Open Badges.
- https://github.com/openmichigan/open_badges -- Drupal module for managing/issuing badges
- https://github.com/PRX/badges_engine -- Rails Engine for issuing.