A world in upheaval
August 26, 2013 1:31 PM Subscribe
A map of every protest everywhere since 1979 (some caveats are noted in the accompanying article).
Hmmmm. According to this, protests weren't discovered in Australasia until the late 1990s.
posted by yoink at 1:35 PM on August 26, 2013 [1 favorite]
posted by yoink at 1:35 PM on August 26, 2013 [1 favorite]
What's that spot in central USA that keeps flashing?
As the article explains, any protest without a city mentioned is put on the map in the centre of the country it takes place in, in this case Kansas.
posted by MartinWisse at 1:37 PM on August 26, 2013
As the article explains, any protest without a city mentioned is put on the map in the centre of the country it takes place in, in this case Kansas.
posted by MartinWisse at 1:37 PM on August 26, 2013
The map doth protest too much, methinks.
posted by Kabanos at 1:40 PM on August 26, 2013 [2 favorites]
posted by Kabanos at 1:40 PM on August 26, 2013 [2 favorites]
The visualization makes it look like the world is getting angrier and angrier, and that might be part of it. However how much is due to increased reporting of protests over time and how much is an increase also due to the internet/social media making organization easier?
posted by charred husk at 1:42 PM on August 26, 2013 [3 favorites]
posted by charred husk at 1:42 PM on August 26, 2013 [3 favorites]
The GDELT dataset, of which this map is based on, has the majority of its entries from the last decade, IIRC.
posted by MisantropicPainforest at 1:43 PM on August 26, 2013
posted by MisantropicPainforest at 1:43 PM on August 26, 2013
However how much is due to increased reporting of protests over time
I'm going to guess "most of it." In my experience, it was a lot easier to get people into the streets in the late 60s-80s in Western countries than in the 90s-now.
posted by yoink at 2:05 PM on August 26, 2013
I'm going to guess "most of it." In my experience, it was a lot easier to get people into the streets in the late 60s-80s in Western countries than in the 90s-now.
posted by yoink at 2:05 PM on August 26, 2013
I'm listening to "Autumn Leaves" by the Bill Evans trio as I look at this, and the dots are flashing in the same tempo. I don't know if that's relevant.
posted by randomkeystrike at 2:21 PM on August 26, 2013
posted by randomkeystrike at 2:21 PM on August 26, 2013
From the article: There are some strange things that happen when the data are mapped, as well. A cursory glance at the map would suggest that Kansas is the most restive state in the union, but really the frequent protests popping up somewhere near Wichita are every media mention of a protest in the United States that doesn't specify a city (the same goes for that flickering dot north of Mongolia in Middle-of-Nowhere, Russia).
Ugh, this is so irritating. You can't have it both ways. You can either go by the specific location or by the country. Combining both levels of data on the same map is incredibly misleading cartography. One way this could have been avoided is by aggregating all data up to the country level and having the whole country flash, perhaps a brighter color depending on the number of protests that month.
/geogrump
posted by troika at 2:57 PM on August 26, 2013 [5 favorites]
Ugh, this is so irritating. You can't have it both ways. You can either go by the specific location or by the country. Combining both levels of data on the same map is incredibly misleading cartography. One way this could have been avoided is by aggregating all data up to the country level and having the whole country flash, perhaps a brighter color depending on the number of protests that month.
/geogrump
posted by troika at 2:57 PM on August 26, 2013 [5 favorites]
So apparently Wichita is the default for a protest with no mention of a city. Which then leads one to ask what kind of protests are they including and what kinds are they not? It would seem to be more sensible to only include protests that have a city mentioned because most likely the protest was a significant one. As opposed to 15 people who showed up for an hour to make noise.
The map obviously is telling us that protests are being reported with more regularity in the past few years. But without a criteria of what constitutes a protest and where it is actually happening then the data doesn't mean much.
posted by Rashomon at 3:10 PM on August 26, 2013
The map obviously is telling us that protests are being reported with more regularity in the past few years. But without a criteria of what constitutes a protest and where it is actually happening then the data doesn't mean much.
posted by Rashomon at 3:10 PM on August 26, 2013
Which then leads one to ask what kind of protests are they including and what kinds are they not?
Clearly they are including protests about lineman-related issues. At least, at a county level.
posted by yoink at 3:21 PM on August 26, 2013
Clearly they are including protests about lineman-related issues. At least, at a county level.
posted by yoink at 3:21 PM on August 26, 2013
The map obviously is telling us that protests are being reported with more regularity in the past few years.
But protests were widely reported in Australia during the 1980s and 1990s. And there were lots of them. We had around 1,500 industrial actions a year in 1987 compared to 150 a year in 2007. Why show the earlier data if it's so hopelessly skewed by reporting bias that it shows the direct opposite of the truth?
This is kind of an object lesson in misleading data visualisation.
posted by dontjumplarry at 3:28 PM on August 26, 2013 [2 favorites]
But protests were widely reported in Australia during the 1980s and 1990s. And there were lots of them. We had around 1,500 industrial actions a year in 1987 compared to 150 a year in 2007. Why show the earlier data if it's so hopelessly skewed by reporting bias that it shows the direct opposite of the truth?
This is kind of an object lesson in misleading data visualisation.
posted by dontjumplarry at 3:28 PM on August 26, 2013 [2 favorites]
Every pro-protest friend of mine seems to have shared this on Facebook with some kind of comment like OMGZ REVOLUTION IS NIGH but has failed to note the important qualifier from the article:
Another issue is that the results are only as good as the data. While the scale of GDELT's database is impressive, it's influenced by its source: international news reporting. Kalev Leetaru, the Yahoo! fellow at Georgetown University working on the GDELT project, told FP by email that the apparent uptick in protests around the world starting in the mid-1990s may be misleading. "In some other work we are doing right now, preliminary results suggest that as a percentage of all events captured in GDELT, protests have not become more common overall," he explained. "So, the majority of that increase in protest events over time stems from the increase in available digital media," especially news.
posted by Noisy Pink Bubbles at 3:49 PM on August 26, 2013
Another issue is that the results are only as good as the data. While the scale of GDELT's database is impressive, it's influenced by its source: international news reporting. Kalev Leetaru, the Yahoo! fellow at Georgetown University working on the GDELT project, told FP by email that the apparent uptick in protests around the world starting in the mid-1990s may be misleading. "In some other work we are doing right now, preliminary results suggest that as a percentage of all events captured in GDELT, protests have not become more common overall," he explained. "So, the majority of that increase in protest events over time stems from the increase in available digital media," especially news.
posted by Noisy Pink Bubbles at 3:49 PM on August 26, 2013
Got tired of waiting for all the CRAP on that page to load. Obvious that maps are the last thing on their minds.
posted by Twang at 6:25 PM on August 26, 2013
posted by Twang at 6:25 PM on August 26, 2013
There's some serious lack of San Francisco activity on that map.
posted by aspo at 9:15 PM on August 26, 2013
posted by aspo at 9:15 PM on August 26, 2013
Odd that a Penn State researcher would omit PSU's Gentle Thursday, the annual Spring pot protestival that ran through 1981, including auxiliary gatherings at PSU's sister campus at University of Cologne, Germany. At least, I couldn't get a blip in central Pennsylvania during the relevant years.
posted by Ardiril at 11:28 PM on August 26, 2013
posted by Ardiril at 11:28 PM on August 26, 2013
It's an interesting idea, but I feel like the dataset is small enough that it would be worth modifying it by hand to actually make it accurate. Also, to have it start in the 60s so we could see the explosion of protests from '64 on.
posted by empath at 11:49 PM on August 26, 2013
posted by empath at 11:49 PM on August 26, 2013
It's an interesting idea, but I feel like the dataset is small enough that it would be worth modifying it by hand to actually make it accurate.
The GDELT datset is the largest ever of its kind.
posted by MisantropicPainforest at 7:45 AM on August 27, 2013
The GDELT datset is the largest ever of its kind.
posted by MisantropicPainforest at 7:45 AM on August 27, 2013
« Older What That Pet is Really Thinking | Video Game 3,000 Newer »
This thread has been archived and is closed to new comments
posted by Foci for Analysis at 1:33 PM on August 26, 2013