Interesting read about what can happen if you use a technology – IP addresses in this case – for something completely different than it was meant to:
For the last decade, Taylor and her renters have been visited by all kinds of mysterious trouble. They’ve been accused of being identity thieves, spammers, scammers and fraudsters. They’ve gotten visited by FBI agents, federal marshals, IRS collectors, ambulances searching for suicidal veterans, and police officers searching for runaway children. They’ve found people scrounging around in their barn. The renters have been doxxed, their names and addresses posted on the internet by vigilantes. Once, someone left a broken toilet in the driveway as a strange, indefinite threat.
The title is a little bit misleading – it’s not really a mapping glitch, but I can see how using IP-based geocoding in the title would turn most readers away without even reading the first sentence of the article.
Andrej Verity tells the story of the Humanitarian eXchange Language. Even though it turned out very different from our initial prototype, I’m glad it is still going strong. I learned a lot working on the HXL prototype and had a great time with the guys at UN OCHA.
I wrote a little wrap-up of this year’s Google Summer of Code Mentor Summit over on the 52° North Blog.
This is the most funny thing you will read this weekend:
The plan: get a robot arm, have it pour cereal and milk into a bowl and feed it to me with a spoon. The only catch was that I had to learn how to program a robot arm and code a pretty complicated sequence to have it serve me breakfast. But I was up for the challenge, because the best way to avoid real problems is to deal with fake ones.
The US Census provides an incredible wealth of data but it’s not always easy to work with it. In the past, working with the tabular and spatial census data generally meant downloading a table from FactFinder and a shapefile from the boundary files site and joining the two, perhaps in a GIS system. These files could also be handled in R but getting the data, reading it into R and, in particular, merging tabular and spatial data can be a chore. Working with slots and all the different classes of data and functions can be challenging.
A recent interesting post on stackoverflow by Claire Salloum prompted me to revisit this issue in R and I’ve definitely found some valuable new packages for capturing and manipulating Census data in R.
Great post explaining how to wrangle and map census data in R.
Very handy tutorial for hexbinning directly in PostGIS.
Short version: In QGIS, use the Create grid layer function in the MMQGIS plugin to create the hexbins. Then import this layer into PostGIS and use the ST_Contains function to join your points spatially to the hexbins. Voilà.
Google Summer of Code 2015 is coming to an end today. I’ve been mentoring Deepak over the summer, who built an awesome prototype to bring the enviroCar data into the LOD cloud.
If you were always wondering how machine learning works, but didn’t dare to ask.
CartoDB’s Chris Whong wrote this super-handy data exporter for NYC PLUTO data, the city’s land use and geographic data set. Lets you select geographic and thematic subsets of the data and either download them in various formats, or directly load them into your CartoDB account.