just thought of an xkcd-style graph joke but idk what the easiest way to draw it is. it would be a log-log contour plot where the x-axis is “number of likes” and the y-axis is “number of RTs” and then you draw regions for good tweets and bad tweets based on popularity and ratio
Conversation
Replying to
you could also plot the location of individual tweets. it could be real interesting actually. anyone wanna make it? ?
1
9
i don't even know where to go on the internet to mock this up real quick
5
6
nice, thanks to the commenters who suggested excalidraw. this is a quick and dirty mockup of what i want the axes to look like at least
1
12
Replying to
best way is to draw it free hand, otherwise i would draw it free hand using a graph as a background, otherwise id use the graphics environment in mathematica
1
Replying to
This is not that hard to do with a bit of python. If you have a list of tweets you want to plot, I'd happily make a loglog plot out of them.
2
Replying to
Hmmm not sure if this would technically qualify as a contour plot in the absence of a 3rd dimension of data. Still a good idea tho!
1
Unless the z component was dedicated to "goodness" of tweets. Which is probably what you mean on second thought 😅