3 ms·

This could've been an informative, info-dense post but the choice of parameters used to display the data and the colors for the maps and the data make it unusab

This could've been an informative, info-dense post but the choice of parameters used to display the data and the colors for the maps and the data make it unusable. At least there is a link to the USDA data so that anyone can recreate a map and set display parms to something with higher interpretability.

doodlebugging 2y ago
I made this comment late at night when my eyes might've been tired. Since someone obviously didn't agree with it I have revisited the link post to see whether I was just tired. After reviewing the link following a full night's sleep I will add a bit more information for anyone caring to create graphics like the ones used in this post. Plan your foreground and background color schemes so that the information and data that you are trying to convey to your readers is apparent at a glance. Use contrasting colors for boundaries so that graphics can be easily interpreted on most monitors. When plotting data points use a contrasting color that stands out well from the background color at the scale you are using for the graphics and data points should be plotted at a resolution that allows the viewer to easily spot them to determine their distribution across the graphic area. My initial conclusions about the data display maps in this post are validated on re-examination. The US map overlay is drawn using a thin grey line on a charcoal grey background. The state-line boundaries appear to be double thickness relative to the outer boundary and as a result they stand out better when the user clicks one of the squares for enlarging. However, neither the state-line or the outer boundary lines stand out well against the charcoal background even after clicking one of the map displays. The state profiles are filled with the map background color instead of a neutral color that would allow the plotted crop data to stand out. As a result of that and the choice of low contrast colors for the plotted points the crops that are not widely planted (garlic for example) produce maps that make it difficult to see the true extent of production of that item. For garlic, most of it appears to be grown in central California but there are a couple of pixels that appear to be live in eastern Washington. The choice of color is not good for this display since garlic is displayed in grey on a charcoal background with state outlines in a light grey similar to the garlic data points. This is just one example where using a different fill inside state boundaries would improve the viewer's ability to interpret or make sense of the data plotted. The size of the plotted points makes it appear that the actual locations of each of the farms where a particular crop is grown are being plotted. That is fine as long as the viewer has an opportunity to visually locate each of the pixels. The choice of colors for the plotted points, the intensity of the color on the background, the size of the points, the choice of state infill colors, etc all contribute to make this plotted data harder to visualize than it should've been. In my 40 year career using map and cross-section views plotting bazillions of bits of data from all over the world I can tell you all straight up that if someone turned this in to me as a finished product ready for client use I would kick it back to them and have them change plot data display parameters so that there is zero ambiguity about the data being displayed. Never use a display scheme that leaves the audience asking question about what they are supposed to be seeing or worse, asking for improvements because the chosen parms make it difficult to follow the story you are trying to tell with your data. And after all that is said, the last map in your post is not clickable, making it more difficult to interpret than the others.

doodlebugging 2y ago

I made this comment late at night when my eyes might've been tired. Since someone obviously didn't agree with it I have revisited the link post to see whether I was just tired. After reviewing the link following a full night's sleep I will add a bit more information for anyone caring to create graphics like the ones used in this post. Plan your foreground and background color schemes so that the information and data that you are trying to convey to your readers is apparent at a glance. Use contrasting colors for boundaries so that graphics can be easily interpreted on most monitors. When plotting data points use a contrasting color that stands out well from the background color at the scale you are using for the graphics and data points should be plotted at a resolution that allows the viewer to easily spot them to determine their distribution across the graphic area. My initial conclusions about the data display maps in this post are validated on re-examination. The US map overlay is drawn using a thin grey line on a charcoal grey background. The state-line boundaries appear to be double thickness relative to the outer boundary and as a result they stand out better when the user clicks one of the squares for enlarging. However, neither the state-line or the outer boundary lines stand out well against the charcoal background even after clicking one of the map displays. The state profiles are filled with the map background color instead of a neutral color that would allow the plotted crop data to stand out. As a result of that and the choice of low contrast colors for the plotted points the crops that are not widely planted (garlic for example) produce maps that make it difficult to see the true extent of production of that item. For garlic, most of it appears to be grown in central California but there are a couple of pixels that appear to be live in eastern Washington. The choice of color is not good for this display since garlic is displayed in grey on a charcoal background with state outlines in a light grey similar to the garlic data points. This is just one example where using a different fill inside state boundaries would improve the viewer's ability to interpret or make sense of the data plotted. The size of the plotted points makes it appear that the actual locations of each of the farms where a particular crop is grown are being plotted. That is fine as long as the viewer has an opportunity to visually locate each of the pixels. The choice of colors for the plotted points, the intensity of the color on the background, the size of the points, the choice of state infill colors, etc all contribute to make this plotted data harder to visualize than it should've been. In my 40 year career using map and cross-section views plotting bazillions of bits of data from all over the world I can tell you all straight up that if someone turned this in to me as a finished product ready for client use I would kick it back to them and have them change plot data display parameters so that there is zero ambiguity about the data being displayed. Never use a display scheme that leaves the audience asking question about what they are supposed to be seeing or worse, asking for improvements because the chosen parms make it difficult to follow the story you are trying to tell with your data. And after all that is said, the last map in your post is not clickable, making it more difficult to interpret than the others.