-
Notifications
You must be signed in to change notification settings - Fork 1
/
london-geomob-presentation.html
459 lines (369 loc) · 12.9 KB
/
london-geomob-presentation.html
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
246
247
248
249
250
251
252
253
254
255
256
257
258
259
260
261
262
263
264
265
266
267
268
269
270
271
272
273
274
275
276
277
278
279
280
281
282
283
284
285
286
287
288
289
290
291
292
293
294
295
296
297
298
299
300
301
302
303
304
305
306
307
308
309
310
311
312
313
314
315
316
317
318
319
320
321
322
323
324
325
326
327
328
329
330
331
332
333
334
335
336
337
338
339
340
341
342
343
344
345
346
347
348
349
350
351
352
353
354
355
356
357
358
359
360
361
362
363
364
365
366
367
368
369
370
371
372
373
374
375
376
377
378
379
380
381
382
383
384
385
386
387
388
389
390
391
392
393
394
395
396
397
398
399
400
401
402
403
404
405
406
407
408
409
410
411
412
413
414
415
416
417
418
419
420
421
422
423
424
425
426
427
428
429
430
431
432
433
434
435
436
437
438
439
440
441
442
443
444
445
446
447
448
449
450
451
452
453
454
455
456
457
458
459
<!DOCTYPE html>
<html>
<head>
<title>Title</title>
<meta charset="utf-8"/>
<link rel="stylesheet" href="res/london-geomob-presentation/style.css"></link>
<script src="res/js/remark.js"></script>
<script src="res/js/d3/d3.v4.min.js"></script>
<script src="res/js/topojson.v1.min.js"></script>
<script src="res/js/sql/sql.js"></script>
<script src="res/js/libs.js"></script>
<script src="res/london-geomob-presentation/js/hex/geometry.js"></script>
<script src="res/london-geomob-presentation/js/hex/grid.js"></script>
<script src="res/london-geomob-presentation/js/databaseinteractions.js"></script>
</head>
<body>
<textarea id="source" data-proofer-ignore >
class: center, middle
# Coding Cartography
## Effectively Communicating Geospatial Data with Javascript
### alexander-brett.co.uk
???
Hi guys, my name's Alex, and I'm really happy to be here this evening.
---
# Motivation
???
The other day I put a graphic up on my website.
Ed saw it and thought that...
--
- We care about geospatial data
???
...since we're all here because we care about geospatial data, you might
enjoy some of my thoughts about what I made, how, and why.
--
- Maps are easy
???
It turns out that drawing maps in the browser is pretty straightforward
with help from appropriately-chosen libraries...
--
- Good visualisations are also easy
???
...and that with a some thought we can produce other visualisations
that more effectively communicate our geospatial data.
---
[The New York Times election coverage](http://elections.nytimes.com/2010/results/senate)
???
I'm talking about things like this
---
[The Guardian election coverage](http://www.theguardian.com/politics/ng-interactive/2015/may/07/live-uk-election-results-in-full)
---
# Content
- We're going to draw pictures.
???
The idea is that I will demonstrate some different approaches using
using javascript.
A warning though:
Ed sent us all instructions not to show screens full of code...
--
<br/>
- We're going to look at code.
???
... which I'm going to ignore.
---
# A quick note to talk about frameworks
--
- I like frameworks for getting stuff done quickly
--
- d3 and topojson are developed by Mike Bostock
--
- they rock
--
- sql.js also rocks
???
I'll get back to why this is important later
---
# Chloropleth
???
We'll start out with the most straightforward example, a Chloropleth,
which is simply a map with regions coloured in according
to some metric.
I also think the name Chloropleth is wonderful.
--
(With thanks to Mike Bostock - https://bl.ocks.org/mbostock/4060606)
???
There is a detailed tutorial for this sort of thing at this link.
---
#Get geography
???
Before we can think about putting data in place, we need some basic geography.
--
[http://data.london.gov.uk/dataset/statistical-gis-boundary-files-london](http://data.london.gov.uk/dataset/statistical-gis-boundary-files-london)
<!--
TRANSFORMING THIS TO A TOPOJSON FILE
The boundaries are provided as an ESRI shapefile, which is only of limited usefulness
when working with javascript.
```
"C:\Program Files\GDAL\ogr2ogr.exe"\
-f GeoJSON\
-t_srs EPSG:4326\
-lco COORDINATE_PRECISION=7\
wards.json\
London_Ward_CityMerged.shp
```
```
topojson\
-o wards_simplified.topojson\
--simplify-proportion 0.5\
--properties\
--id-property GSS_CODE\
-- wards.json
```
-->
???
Fortunately, there are some fantastic resources available for us in this regard.
All of my London data today comes from data.london.gov.uk - it's brilliant, easy-to-navigate,
and there are some really surprisingly detailed and thorough datasets here. I'll show you
a couple later.
--
???
For now we'll find the ward boundaries, then download and unzip them.
---
#Write Code
---
```html
<!DOCTYPE html>
<html>
<head>
<script src="d3.v4.min.js"></script>
<script src="topojson.v1.min.js"></script>
<link rel="stylesheet" href="style.css"/>
</head>
<body>
<svg id="Uncoloured-Chloropleth" width="500" height="500"></svg>
<script>
var projection = d3.geoMercator()
.scale(28000)
.center([-0.09, 51.48])
.translate([250,250]);
d3.json("wards_simplified.topojson", function(error, data){
d3.select("#Uncoloured-Chloropleth").append("g")
.selectAll("path")
.data(topojson.feature(data, data.objects.wards).features)
.enter().append("path")
.attr("d", d3.geoPath().projection(projection))
.attr("class", "wards-outline");
});
</script>
</body>
</html>
```
???
I won't force you to live through every statement in this - if you're interested
you should hit up my github - but what I mean to emphasise by this slide is that
what I'm about to do isn't vast or intimidating: this tiny snippet of code will
produce...
---
name: chloropleth-example-outline
<svg id="chloropleth-example-outline" width="600" height="600"></svg>
???
... this result - which I reckon is pretty good going
and a powerful endorsement of the power of d3 and topojson for achieving this
goal.
So far we have successfully drawn the outline of all the wards in London.
---
#Make it colourful
???
but it's not very interesting like that.
Let's colour each ward according to which borough it's in. It's a neat demo, the
data is included in the file we already downloaded, and it only takes...
--
```javascript
function boroughColourGenerator(){
var boroughMap = {}
, boroughIndex = 0
, boroughPalette = d3.scaleOrdinal(d3.schemeCategory20);
return function(datum){
if(boroughMap[datum.properties.BOROUGH] === undefined){
boroughMap[datum.properties.BOROUGH] = boroughIndex++%20;
}
return boroughPalette(boroughMap[datum.properties.BOROUGH]);
}
}
```
???
this much more code to do this:
---
name: chloropleth-example-coloured
<svg id="chloropleth-example-coloured" width="600" height="600"></svg>
---
#Interesting Demographics
--
[http://data.london.gov.uk/dataset/ward-profiles-and-atlas](http://data.london.gov.uk/dataset/ward-profiles-and-atlas)
--
---
name: chloropleth-example-transition
<svg id="chloropleth-example-transition" width="600" height="600"></svg>
???
Of course these things are better when they are animated
###tap
So here is london coloured by the proportion of social housing in each ward.
Thank you open data!
Now it turns out that I am not the world's biggest fan of this particular visualisation.
---
#Chloropleths
???
I think that they are
--
- Perceptually difficult
???
perceptually difficult, because I think that humans tend to weight things quite
a lot by size
which means that you easily end up overlooking one of those tiny
inner-city wards and disproportionately.
that matters particuarly when you're
looking at a variable like election results, because the tiny one in the centre
will influence the outcome just as much.
--
- Representationally inefficient
???
It's extremely difficult to represent more than one variable on a chloropleth,
and you may well want to correlate two things together!
---
#Hex Grid
???
The second idea, inspired by the Guardian coverage as well as a chat I had last
time I was here, was to make a grid of hexagons.
Making each ward the same size makes a lot of sense a lot of the time, because
wards are designed to have approximately equal populations. This dramatically
reduces the perceptual bias of low-density areas.
---
class:center
name: hexgrid-algorithm
<svg id="Hexgrid-algorithm" width="600" height="600"></svg>
???
I ended up with an algorithm which took the wards ordered by how close they
were to the centre, and placed them adjacent to a cell which was already there.
This prevents too much straggling around the edges, and keeps the relationships
between wards more-or-less intact.
---
class:center
name: hexgrid-transition
<svg id="Hexgrid-transition" width="600" height="600"></svg>
???
the result is this - an aesthetically pleasing, but also quite accurate depiction.
Of course, we can also make this one pleasantly animated.
###tap
So here we have the proportion of social housing on a greyscale, inside the ward.
This is one way of showing two different data for each point - it works because
we know how much space we have to play with for each hex, wheras with the
chloropleth there wouldn't have been enough space in the smallest wards.
We can animate the inside and the outside separately
###tap
So here is the winning candidate of the mayoral election in each ward, compared
to the proportion of social housing in the ward.
Does it seem like all of the wards with higher social housing proportions voted
Labour?
This brings me to a very important point..
---
#Politics
--
- Data presentation is a political act
--
- ... whether you meant it to be or not
---
#Politics
- Colour matters
???
For a really obvious example of this, the colours you choose to represent data
matter.
If I use a scale of black to white, or red to green, am I implicitly applying a
normative valuation to the data?
Are we either consciously or unconciously using
black or red to represent 'bad' and white or green to represent 'good'?
And if so, have we examined our decision about what counts as good? For instance,
had you found yourself going around colouring in neighbourhoods with high BAME
populations red or black, you might find yourself in hot water - and not, I think,
entirely unfairly.
Also...
--
- Size matters
???
Size does matter, for similar reasons.
I'll show you why later.
--
- What you present matters
???
Lastly, the data you choose to present matters. If you show people a map which
has mayoral election results and social housing density, then it's entirely to
be expected that they draw an conclusion about your values based on the fact
that you showed them that data in that way.
---
##1: Be aware of your biases
???
What I want you to take away from this is that you should be prepared to learn
things about yourself as a result of undertaking this sort of endeavour. I thought
that I was enjoyably playing around with open data and javascript frameworks,
and I ended up producing a way to correlate house prices with crime rates, and
BAME population with mayoral election results...
--
##2: Be prepared to respond to criticism
???
... and this means that every so often someone will read something into what
you've done that you didn't intend at all, at which point you might need to
adjust a colour scheme, or add a disclaimer.
---
#Bubbles
---
class:center
name: bubbles-animated
<svg id="bubbles-animated" width="600" height="600"></svg>
???
The swiss army knife of bubble animations is the d3 force simulation. It
iteratively repositions each node according to constraints that you programme
in. In this case, we'll specify a collision constraint - which is to say we'll
just mandate that the circles cannot overlap.
###tap
This works well!
But essentially the whole point of this is that we are able to toggle between
different data axes to define the radius. So let's now switch the radius to be
defined by population instead of land area.
###tap
So that worked. Wonderful - and this demonstrates in a way that if the things
you care about are principally aligned along population lines, just treating
every ward as the same is pretty much ok.
I'm still a little dissatisfied with this though, because you can see, especially
around the centre, that things got a little bit confused.
---
class:center
name: bubbles-animated-better
<svg id="bubbles-animated-better" width="600" height="600"></svg>
???
In this example, as well as an anti-collision force, there is a force drawing
the node representing each ward towards its actual position
###tap
this means that
they tend to shuffle themselves towards the right place - although it's only
a weak force, because it's more important that they don't overlap and that they
are pleasingly laid out.
---
class:center, middle
name: bubbles-animated-comparison
<img src="res/london-geomob-presentation/img/before-centering.svg" width="350px"/>
<img src="res/london-geomob-presentation/img/after-centering.svg" width="350px"/>
???
It's not hugely obvious in the moment, but you can see here that it's the kind
of tweak that makes things stay a bit organised in the long run.
---
class: center
#Size matters
<img src="res/london-vis/prices_linear_scale.PNG" width="350px"/>
<img src="res/london-vis/prices_sqrt_scale.PNG" width="350px"/>
---
#The Result
---
class: center
<img src="res/london-geomob-presentation/img/my-website.png"/>
---
class: center, middle
#The End
###alexander-brett.co.uk/2016/06/01/Visualising-London.html
###github.com/alexander-brett/london-geomob-presentation
</textarea>
<script src="res/london-geomob-presentation/js/slideshowfunctions.js"></script>
</body>
</html>