A data-visualization of Meetup info using D3

Overview

This is an extension of the 'egovis' slide from Mike Dewar's talk on D3. We use the Meetup API and Mike Bostock's D3 library.

Approach

Your Meetup API key goes in data/apiKey.txt and is excluded from versioning via .gitignore.

To Do Plan

• add other D3 layout options and allow for switching them
• ajax calls to the Meetup API: more events, paginate members and display while loading, etc.
  • local store of retrieved records (light-weight local DB)

Implementation

Wrap the Meetup API

I assume there is a Python library out there already that handles calls to the Meetup API. I'm rolling my own here just to get some chops. This is the plan:

• navigate the Meetup API graph:
  • select an attendee and see their other attended meetups
  • select another event and see attendees (positive RSVPs)
  • etc.

The file code\meetup.py is a module whose functions should map directly to the Meetup API endpoints. So far, I have getEventAttendees(event_id) and getMembers(group_id) working. The basic use case is import code.meetup as m then members = m.getMembers('1909691').

From Mike's github repo:

a talk aimed to try and get data scientists over that first hump of d3's learing curve and into the beautiful vista beyond...

The code for the slide mechanics was written by Mike Bostock in d3! The data is from meetup.com's public API.

In order to actually run through talk, clone this repository and run python -m SimpleHTTPServer 8000. Then point your browser at http://localhost:8000.