3 ms·

I want to like the article, but the lack of units on the axes are killing me! With the automation one... what's that mean? 300 what? 300 robots? And similarly w

I want to like the article, but the lack of units on the axes are killing me! With the automation one... what's that mean? 300 what? 300 robots? And similarly with jobs declining and such. Some of them describe what's going on in the title, but a lot are sort of handwavy. I'm sure there's real data under it, but it's not presented very well.

alejohausner 9y ago
300, relative to an initial value of 100. This is common practice.
- djokkataja 9y ago
  An initial value of 100 what? And over what time period? It's common practice to not label graphs correctly or at all, but without labels, graphs don't offer any representation of a model of reality (so the author is free to claim the graph indicates whatever he wants it to indicate). ... So if you click the link below the graph for the source data, the data that was graphed was the number of robots per ten thousand workers in the manufacturing industry in 2016. The report also indicated that the US had the second highest number of robots per 10000 workers in the automotive industry in 2016 after South Korea. Given that the article's main point was that the computer industry distorts the overall growth numbers for US manufacturing, it may be worth pointing out that different industries within the overall manufacturing industry may find different levels of automation appropriate at our current level of technology. Without a clearer picture of the relative representation of different manufacturing industries in each of these countries and what's considered state-of-the-art in terms of automation for each, I don't think the data for this graph can be used to unambiguously support the author's conclusion. Link to the source report: https://ifr.org/downloads/press/Executive_Summary_WR_2017_Industrial_Robots.pdf https://ifr.org/downloads/press/Executive_Summary_WR_2017_In...
  - burfog 9y ago
    100% We start at 100% of the starting value, by definition. We then go from there. So if there were 256 robots, and that increased to 384 robots, the graph would start at 100 and end at 150.
lopmotr 9y ago
For some reason, Jounalists, no matter how techincal the thing they're writing about, uniformly forget the basic graph-drawing rules their math and science teachers made them follow when they were 13. It might have seemed like pointless pedantry to their child selves, but then they get it wrong and their graph becomes an empty emotional gimmick instead of actually informative.
- Gigablah 9y ago
  To be fair, that might not necessarily be the fault of the journalist. Could be some hapless intern told to conjure up graphics within a few hours.

alejohausner 9y ago

300, relative to an initial value of 100. This is common practice.

djokkataja 9y ago
An initial value of 100 what? And over what time period? It's common practice to not label graphs correctly or at all, but without labels, graphs don't offer any representation of a model of reality (so the author is free to claim the graph indicates whatever he wants it to indicate). ... So if you click the link below the graph for the source data, the data that was graphed was the number of robots per ten thousand workers in the manufacturing industry in 2016. The report also indicated that the US had the second highest number of robots per 10000 workers in the automotive industry in 2016 after South Korea. Given that the article's main point was that the computer industry distorts the overall growth numbers for US manufacturing, it may be worth pointing out that different industries within the overall manufacturing industry may find different levels of automation appropriate at our current level of technology. Without a clearer picture of the relative representation of different manufacturing industries in each of these countries and what's considered state-of-the-art in terms of automation for each, I don't think the data for this graph can be used to unambiguously support the author's conclusion. Link to the source report: https://ifr.org/downloads/press/Executive_Summary_WR_2017_Industrial_Robots.pdf https://ifr.org/downloads/press/Executive_Summary_WR_2017_In...
- burfog 9y ago
  100% We start at 100% of the starting value, by definition. We then go from there. So if there were 256 robots, and that increased to 384 robots, the graph would start at 100 and end at 150.

lopmotr 9y ago

For some reason, Jounalists, no matter how techincal the thing they're writing about, uniformly forget the basic graph-drawing rules their math and science teachers made them follow when they were 13. It might have seemed like pointless pedantry to their child selves, but then they get it wrong and their graph becomes an empty emotional gimmick instead of actually informative.

Gigablah 9y ago
To be fair, that might not necessarily be the fault of the journalist. Could be some hapless intern told to conjure up graphics within a few hours.