Comments on: Where was your bike stolen?

By: scum

scum — Fri, 04 Nov 2011 19:30:39 +0000

KTVU just now reported on this

By: Andrew Sarkarati

Andrew Sarkarati — Thu, 03 Nov 2011 20:05:22 +0000

no, i was doing an impression of you.

By: phil

phil — Thu, 03 Nov 2011 19:16:28 +0000

Err.. ok.

http://en.wikipedia.org/wiki/Treemap

By: EH

EH — Thu, 03 Nov 2011 18:47:38 +0000

yeah, that wikipedia definition doesn’t apply to what you’re trying to do.

By: Herr Doktor Professor Deth Vegetable

Herr Doktor Professor Deth Vegetable — Thu, 03 Nov 2011 18:36:15 +0000

Hey, don’t mention it Andy, It is just one of the many useful services I provide!

By: Andrew Sarkarati

Andrew Sarkarati — Thu, 03 Nov 2011 14:35:54 +0000

I MUST HAVE EVERYTHING EXPLAINED TO ME OR I CAN NOT COMPREHEND

By: Data vis expert

Data vis expert — Thu, 03 Nov 2011 06:00:41 +0000

Either way, you you’ll have to make adjustments to make it visually obvious there’s a difference. You have one large data set and the rest are smaller and don’t vary that much between them.

By: phil

phil — Thu, 03 Nov 2011 00:27:20 +0000

datavis expert,

Thanks for the tips! You’re right, the areas of the blocks should correlate to the number of posts. The blocks for Oakland and Santa Cruz are roughly the same size because they have relatively the same number of posts – 40 vs 46. The difference of 6 is pretty tiny compared to the total number of data points (633 from what I recall) and the number from San Francisco (224).

By: data visualization expert

data visualization expert — Thu, 03 Nov 2011 00:11:24 +0000

sorry, but these are a mess. it should look more like a map itself, and the size of each block should correlate (roughly) to the amount of posts for that region. the only obvious one is SF on ‘map by city’ but the blocks for oakland and santa cruz are roughly the same size (was that what you were trying to do?). Should be more obvious variation. it should be more clear that these are related to the numbers of posts. don’t hyphenate the city names. you refer to ‘other – text’ but aren’t these posts? if i’m asking these questions it’s because the chart isn’t clear. good idea though

By: Herr Doktor Professor Deth Vegetable

Herr Doktor Professor Deth Vegetable — Wed, 02 Nov 2011 22:51:09 +0000

Agreed.

By: Herr Doktor Professor Deth Vegetable

Herr Doktor Professor Deth Vegetable — Wed, 02 Nov 2011 22:49:57 +0000

The IMAGE clearly states that, however the post does NOT. So… you’re welcome.

By: rod

rod — Wed, 02 Nov 2011 22:37:31 +0000

as the image clearly states, it is a sampling of people who reported their bikes stolen on craigslist. thanks for repeating the obvious.

By: phil

phil — Wed, 02 Nov 2011 21:59:43 +0000

why, EH,

You guys are right – it’s not a map in the traditional sense. I should have labeled it a “Treemap”. I used “map” as in this definition: http://en.wikipedia.org/wiki/Map#Non_spatial_maps

Herr,
You’re absolutely correct – the treemap only includes people who bothered to post about their stolen bikes on Craigslist. I don’t know how well it correlates either but I don’t think we have any really good datasets at this point – not everybody compiles a police report nor submits to the stolen bike registry.

By: EH

EH — Wed, 02 Nov 2011 18:59:08 +0000

Their use of the word “map” can only be explained with a definition that renders the word meaningless.

By: Herr Doktor Professor Deth Vegetable

Herr Doktor Professor Deth Vegetable — Wed, 02 Nov 2011 17:52:51 +0000

It looks like that is just a sampling of where people who post to Craigslist had bikes stolen. I’m not sure how well that would correlate to the actual number of bicycles stolen, just the number of people who had their bikes stolen but decided to complain on craigslist about it?