<!DOCTYPE html PUBLIC "-//W3C//DTD HTML 4.01 Transitional//EN">
<html>
<head>
<meta content="text/html;charset=ISO-8859-3" http-equiv="Content-Type">
</head>
<body bgcolor="#ffffff" text="#000000">
Hamish wrote:
<blockquote cite="mid:610518.92776.qm@web110005.mail.gq1.yahoo.com"
type="cite">
<pre wrap="">Luigi Ponti wrote:
</pre>
<blockquote type="cite">
<pre wrap="">I just ran into this page
<a class="moz-txt-link-freetext" href="http://geography.uoregon.edu/datagraphics/color_scales.htm">http://geography.uoregon.edu/datagraphics/color_scales.htm</a>
that includes, among others, precipitation color tables. I
don't know if that can be useful.
The page of this Lab also provides an interesting paper titled "End of
the Rainbow" which elaborates on why one should not use continuously
varying color schemes (and absolutely no rainbow color table). The way
to go seems to be banded color schemes -- a color scheme with equally
sized bands of constant color [1].
</pre>
</blockquote>
<pre wrap=""><!---->...
</pre>
<blockquote type="cite">
<pre wrap="">[1] Borland D, Taylor II RM (2007) Rainbow color map (still) considered
harmful. IEEE Computer Graphics and Applications, 27, 14-17.
</pre>
</blockquote>
<pre wrap=""><!---->
In matters of human perception the individual experience is not subject
to hard rules. Of course it is important to think about how the color-
blind will see things and how it will look printed in greyscale, etc.
An interesting subject to me, thanks for the link/look forward to reading
it. It reminds me that I still need to read "How to lie with maps":
<a class="moz-txt-link-freetext" href="http://www.eyrie.org/~eagle/reviews/books/0-226-53421-9.html">http://www.eyrie.org/~eagle/reviews/books/0-226-53421-9.html</a>
(in the spirit of "How to lie with statistics")
</pre>
</blockquote>
<br>
I found the refs I posted while browsing the net in search for good
color schemes: it puzzled me that the majority of the maps (including
mine) would use a rainbow, continuous color bar despite good arguments
against this technique were available in the literature. I have no bias
since my experience in very limited (nor do I want to lie with maps).<br>
<br>
<blockquote cite="mid:610518.92776.qm@web110005.mail.gq1.yahoo.com"
type="cite">
<pre wrap="">
It is really quite amazing/scary how much changing the colors affects
the interpretation and is able to hide/highlight features. After spending
a fair bit of time with the issue I personally feel it is most honest to
use a color scale which can be described as a continuous mathematical
function. e.g. log()+BCYR or something directly based on the univariate
stats of the data. -> Take the biased "what looks nicest" human out of
the picture. </pre>
</blockquote>
<br>
Thanks: I was about to ask your opinion on this. Intuitively, I also
feel that continuously varying stuff should be represented as such.<br>
<br>
<blockquote cite="mid:610518.92776.qm@web110005.mail.gq1.yahoo.com"
type="cite">
<pre wrap="">Choosing arbitrary band limits on a continuous value dataset
just compounds the opportunities for arbitrary human influence. </pre>
</blockquote>
<br>
The argument I read is that color bands act as contours lines do in
elevation maps. It is true that choosing arbitrary band limits on a
continuous value dataset is simply... ...arbitrary. However, when you
have a continuously varying color bar in the map, you will have label
numbers next to the bar and you will try to associate the bar color
next to the label value with a color in the map. The argument Borland
& Taylor (2007) make is that if each label of the color bar is
located between two constant color bands in the bar, you will have that
value easily located in the map along the boundary between the two
contiguous areas of constant color. What do you think about this?<br>
<br>
I have no bias given my very limited experience on the matter -- just
trying to make an informed decision. Thanks again for discussing this
and sorry if this is way off topic for the list (please advice).<br>
<br>
<blockquote cite="mid:610518.92776.qm@web110005.mail.gq1.yahoo.com"
type="cite">
<pre wrap="">The
statistical equivalent is to slowly vary the number bins in a histogram
while the peaks seemingly double and halve.
</pre>
</blockquote>
<br>
Yes, you lose information that way.<br>
<br>
Kind regards,<br>
<br>
Luigi<br>
<br>
<br>
</body>
</html>