Miles McCrocklin milroc

This is an implentation of the Bubble Cursor, which was originally introduced by Tovi Grossman and Ravin Balakrishnan at CHI 2005.

Source: American Community Survey, 2011 5-Year Estimate

This map was inspired by a similar map found on Wikipedia. I wasn’t wild about the diverging color scale, so I thought it would be a fun challenge to recreate.

Finding the shapefiles was easy; I used the U.S. National Atlas 1:1,000,000 scale dataset, conveniently packaged in my U.S. Atlas repository. I reprojected the shapefiles to the California Albers projection using ogr2ogr:

ogr2ogr \
	-f 'ESRI Shapefile' \
	-t_srs 'EPSG:3310' \

A minimum spanning tree of the canvas is generated using randomized depth-first traversal. Hue encodes Manhattan distance from the start. (This is not an optimal visual encoding, but it suffices and is pretty.)

Randomized depth-first traversal can also be used to generate mazes. See a maze generated with randomized depth-first traversal flooded with color, and compare color floods of spanning trees generated by Prim’s algorithm, Wilson’s algorithm and random traversal.

This is a fork of Interrupting Chained Transitions. It is showing the power of d3.dispatch for building reusable components.

d3.legend

d3.legend is a quick hack to add a legend to a d3 chart. Simply add a g and .call(d3.legend). Any elements that have a title set in the "data-legend" attribute will be included when d3.legend is called. Each title will appear only once (even when multiple items define the same data-legend) as the process uses a set based on a existing names, not an array of all items.

Color

By default the color in the legend will try to match the fill attribute or the stroke attribute of the relevant items. Color can be explicitly defined by attribute "data-legend-color"

Order

The order of items in the legend will be sorted using the top of the bounding box for each included item. The order can be explicitly defined by attribute "data-legend-pos"

	# Code for slides:

	# 1. Avoid code in your slides, if you have 10 slides and more
	# than 2 have code in it, you are making people do a diff review
	# while you babble on in front of them...

	# 2. Explain the hard concepts with anecdotes, or visuals.

	# 3. Code on slides should be taken with MUCH more care to comprehension
	# than code in your repository, imagine if you only had 3 seconds

	// micro example for the d3.js day 1 course taught by @vicapow
	// https://github.com/mbostock/d3/wiki/Selections#control
	// is part of the API to look at
	// most of this code is about readability
	// the performance should be congruent through all options
	// roughly

	var circles = d3.selectAll('.circles').data([3, 2, 1, 0]);

	// option one

	/*!
	* EventEmitter v4.2.7 - git.io/ee
	* Oliver Caldwell
	* MIT license
	* @preserve
	*/
	(function(){"use strict";function t(){}function r(t,n){for(var e=t.length;e--;)if(t[e].listener===n)return e;return-1}function n(e){return function(){return this[e].apply(this,arguments)}}var e=t.prototype,i=this,s=i.EventEmitter;e.getListeners=function(n){var r,e,t=this._getEvents();if(n instanceof RegExp){r={};for(e in t)t.hasOwnProperty(e)&&n.test(e)&&(r[e]=t[e])}else r=t[n]\|\|(t[n]=[]);return r},e.flattenListeners=function(t){var e,n=[];for(e=0;e<t.length;e+=1)n.push(t[e].listener);return n},e.getListenersAsObject=function(n){var e,t=this.getListeners(n);return t instanceof Array&&(e={},e[n]=t),e\|\|t},e.addListener=function(i,e){var t,n=this.getListenersAsObject(i),s="object"==typeof e;for(t in n)n.hasOwnProperty(t)&&-1===r(n[t],e)&&n[t].push(s?e:{listener:e,once:!1});return this},e.on=n("addListener"),e.addOnceListener=function(e,t){return this.addListener(e,{listener:t,once:!0})},e.once=n("addOnceListener"),e.defineE

	<!DOCTYPE html>
	<meta charset="utf-8">
	<html>
	<head>
	<!-- CDN IMPORTS -->
	<link href='http://fonts.googleapis.com/css?family=Inconsolata' rel='stylesheet' type='text/css'>
	<!-- LOCAL IMPORTS -->

	<!-- IN GIST -->
	<link rel="stylesheet/less" type="text/css" href="style.less">