using Plots; plotly()
xs = [ "x$i" for i in 1:10 ]
p = bar(xs, -5:4, orientation = :h)

julia> p[1][1][:x]
10-element Array{String,1}:
"x1"
"x2"
"x3"
"x4"
"x5"
"x6"
"x7"
"x8"
"x9"
"x10"
julia> p[1][1][:y]
10-element Array{Float64,1}:
0.5
1.5
2.5
3.5
4.5
5.5
6.5
7.5
8.5
9.5
There is also a disagreement with axis[:discrete_values]
julia> p[1][:yaxis][:discrete_values]
10-element Array{Any,1}:
"x1"
"x2"
"x3"
"x4"
"x5"
"x6"
"x7"
"x8"
"x9"
"x10"
julia> p[1][:xaxis][:discrete_values]
0-element Array{Any,1}
which is making ticks = :native output both x and y as the array `["x1", "x2", ... , "x10"] and gives

This was a fun one. It happens in here because the x-axis does not get properly updated when the orientation is horizontal. It affects Plotly/PlotlyJS because they rely on native support for :bar whereas the others (GR, PyPlot, and PGFPlots, at least) use the step recipe.
A fix is on the way!
Most helpful comment
This was a fun one. It happens in here because the x-axis does not get properly updated when the orientation is horizontal. It affects Plotly/PlotlyJS because they rely on native support for
:barwhereas the others (GR, PyPlot, and PGFPlots, at least) use thesteprecipe.A fix is on the way!