panel_13

[hCraker]

Closes #130

[hCraker]

[clyne]

Rose, the arrow glyphs in the Python version are slightly shorter, and thicker than the NCL version. As a result many of the smaller magnitude glyphs - in the equatorial region, for example - degenerate to blobs that don't really have a discernible orientation. Could you try tweaking these a bit and see if you can improve the clarity?

[hCraker]

I tried to make the arrows more legible. Is this good @clyne?

[clyne]

That's better, but some of the arrows are still pretty much blobs unless you zoom in a lot. I see what the problem is; if you set the scale factor to something like 200 the smallest arrows are visible, but then the others are too long. I believe NCL has a notion of minimum arrow size, and I think @michaelavs or @clairefio may have run into this. I'm not sure if there is a simple solution, but perhaps they have some thoughts. If there is not a simple workaround, how about adding a comment that some arrows aren't legible, and opening an issue on geocat-viz to provide a means to set a minimum arrow size?

[hCraker]

@clyne I have found a simple way to set minimum lengths using some trig. It works fine, but I'm afraid it is misleading because I had to manipulate the data to produce the desired lengths. Is this best practice? I could add a comment to the documentation explaining how this may not be a desirable feature when trying to make truly accurate plots.

[jukent]

@clyne I have found a simple way to set minimum lengths using some trig. It works fine, but I'm afraid it is misleading because I had to manipulate the data to produce the desired lengths. Is this best practice? I could add a comment to the documentation explaining how this may not be a desirable feature when trying to make truly accurate plots.

What type of data manipulation are you referring to?

[hCraker]

@clyne I have found a simple way to set minimum lengths using some trig. It works fine, but I'm afraid it is misleading because I had to manipulate the data to produce the desired lengths. Is this best practice? I could add a comment to the documentation explaining how this may not be a desirable feature when trying to make truly accurate plots.

What type of data manipulation are you referring to?

I'm searching through the data for any wind vector with a magnitude less than minLength and then changing those vectors' components to make their magnitude equal to minLength

[jukent]

I'm searching through the data for any wind vector with a magnitude less than minLength and then changing those vectors' components to make their magnitude equal to minLength

It might be more honest to not display wind vectors below that minimum length than to change their size, but an argument could be made for still displaying wind direction. I would feel better about this if it were somehow indicated in a key (size X represents wind of value X and below, for example). Thoughts @clyne ?

[clyne]

I'm searching through the data for any wind vector with a magnitude less than minLength and then changing those vectors' components to make their magnitude equal to minLength

It might be more honest to not display wind vectors below that minimum length than to change their size, but an argument could be made for still displaying wind direction. I would feel better about this if it were somehow indicated in a key (size X represents wind of value X and below, for example). Thoughts @clyne ?

I don't like the idea of misrepresenting the arrow length without clearly indicating in the plot that not all arrows are drawn to scale. Ideally, a different glyph would be drawn to indicate this. I don't think we're going to solve this problem with the time remaining and suggest falling back to the illegible glyphs and a comment in the documentation for the plot.

[hCraker]

@clyne @jukent I have opened an issue regarding this discussion so we remember to fix this later. Please let me know if the added comment in the latest commit needs clarification.

[clyne]

Thanks, Rose.

[clyne]

Thanks, Rose.

Rose, I don't see a comment in the example about the glyph legibility issue.

[hCraker]

Rose, I don't see a comment in the example about the glyph legibility issue.

I was having issues with git, but I just pushed the change. Sorry about that @clyne

[clyne]

Terrific!

[erogluorhan]

Nice work!