slightly exaggerated PyPlot example in the manual

[stevengj]

The PyPlot [example in the manual](https://juliaplots.github.io/pipeline/ seems a lot more complicated than necessary. The simplest analogue of the using Plots; plot(...) example in the manual would be

using PyPlot
figure(figsize=(4,3), dpi=100)
plot(x, y[:,1], alpha = 0.5, "steelblue", linewidth = 4)
scatter(x, y[:,2], alpha = 0.5, marker = "+", s = 100, c="orangered")
plt[:hist](y[:,3], orientation = "horizontal", alpha = 0.5,
                   normed = true, bins=30, color="green",
                   linewidth = 0)
grid()
title("My title")
legend(["y1","y2"])

[stevengj]

Also, a lot of the complexity here is simply that Matplotlib has different defaults than Plots. If you are willing to accept more of Matplotlib's defaults (since you were willing to accept Plots' defaults), you can do:

using PyPlot
plot(x, y[:,1], alpha = 0.5, linewidth = 4)
plot(x, y[:,2], alpha = 0.5, "r+")
plt[:hist](y[:,3], orientation = "horizontal", alpha = 0.5, normed = true)
grid()
title("My title")
legend(["y1","y2"])

[stevengj]

(Plots is a great package, but I don't think the current comparison to PyPlot is informative.)

[tbreloff]

I'll merge a PR which changes this

…

On Mon, Mar 6, 2017 at 2:31 PM, Steven G. Johnson ***@***.***> wrote: (Plots is a great package, but I don't think the current comparison to PyPlot is informative.) — You are receiving this because you are subscribed to this thread. Reply to this email directly, view it on GitHub <#719 (comment)>, or mute the thread <https://github.com/notifications/unsubscribe-auth/AA492lh7BilwHKEu7RjGMq8kppPP_a0Kks5rjF79gaJpZM4MUlPh> .

[tbreloff]

And to be clear... that specific section was to be explicit about what is happening "inside" Plots, in that it's not "just calling PyPlot's plot method". It was not meant to be "look at how much better Plots is". So accepting PyPlot's defaults would be completely against the intention of the section.

…

On Mon, Mar 6, 2017 at 3:10 PM, Tom Breloff ***@***.***> wrote: I'll merge a PR which changes this On Mon, Mar 6, 2017 at 2:31 PM, Steven G. Johnson < ***@***.***> wrote: > (Plots is a great package, but I don't think the current comparison to > PyPlot is informative.) > > — > You are receiving this because you are subscribed to this thread. > Reply to this email directly, view it on GitHub > <#719 (comment)>, > or mute the thread > <https://github.com/notifications/unsubscribe-auth/AA492lh7BilwHKEu7RjGMq8kppPP_a0Kks5rjF79gaJpZM4MUlPh> > . >

[stevengj]

Fair enough, thanks!