<?xml version="1.0" encoding="UTF-8"?><!-- generator="wordpress/2.3.3" -->
<rss version="2.0"
	xmlns:content="http://purl.org/rss/1.0/modules/content/"
	xmlns:dc="http://purl.org/dc/elements/1.1/"
	>
<channel>
	<title>Comments on: Pretty as Pie</title>
	<link>http://www.willmcgugan.com/2008/06/29/pretty-as-pie/</link>
	<description>Blog of Will McGugan</description>
	<pubDate>Fri, 21 Nov 2008 00:00:09 +0000</pubDate>
	<generator>http://wordpress.org/?v=2.3.3</generator>
		<item>
		<title>By: Виктор</title>
		<link>http://www.willmcgugan.com/2008/06/29/pretty-as-pie/#comment-71359</link>
		<dc:creator>Виктор</dc:creator>
		<pubDate>Sun, 09 Nov 2008 00:04:22 +0000</pubDate>
		<guid>http://www.willmcgugan.com/2008/06/29/pretty-as-pie/#comment-71359</guid>
		<description>Можно ли взять одну картинку с Вашего блога? Очень понравилась. Линк на Вас есстественно поставлю.</description>
		<content:encoded><![CDATA[<p>Можно ли взять одну картинку с Вашего блога? Очень понравилась. Линк на Вас есстественно поставлю.</p>
]]></content:encoded>
	</item>
	<item>
		<title>By: Marielle Winarto</title>
		<link>http://www.willmcgugan.com/2008/06/29/pretty-as-pie/#comment-31728</link>
		<dc:creator>Marielle Winarto</dc:creator>
		<pubDate>Thu, 24 Jul 2008 10:40:35 +0000</pubDate>
		<guid>http://www.willmcgugan.com/2008/06/29/pretty-as-pie/#comment-31728</guid>
		<description>A beautiful image, but very confusing. A pie chart is used for a visual presentation of data. The size of a part corresponds to the relative size of this part of the data. In your example, multiple interpretations are possible:
angle, height and even volume.

A chart like this could really make sense if you want to visualise multiple factors. Example: sales figures for visitors to a webshop, sorted by country of origin. In that case the circle parts could represent the country of origin, the height of the parts the average amount paid and the volume of each part would represent the total amount of sales to that particular country.</description>
		<content:encoded><![CDATA[<p>A beautiful image, but very confusing. A pie chart is used for a visual presentation of data. The size of a part corresponds to the relative size of this part of the data. In your example, multiple interpretations are possible:<br />
angle, height and even volume.</p>
<p>A chart like this could really make sense if you want to visualise multiple factors. Example: sales figures for visitors to a webshop, sorted by country of origin. In that case the circle parts could represent the country of origin, the height of the parts the average amount paid and the volume of each part would represent the total amount of sales to that particular country.</p>
]]></content:encoded>
	</item>
	<item>
		<title>By: Where there is a Will &#187; Blog Archive &#187; Food File on Ubuntu</title>
		<link>http://www.willmcgugan.com/2008/06/29/pretty-as-pie/#comment-30394</link>
		<dc:creator>Where there is a Will &#187; Blog Archive &#187; Food File on Ubuntu</dc:creator>
		<pubDate>Mon, 14 Jul 2008 17:32:28 +0000</pubDate>
		<guid>http://www.willmcgugan.com/2008/06/29/pretty-as-pie/#comment-30394</guid>
		<description>[...] think it was this project that started my pie chart fetish. I wonder if there is a support group or something&#8230;  Share [...]</description>
		<content:encoded><![CDATA[<p>[&#8230;] think it was this project that started my pie chart fetish. I wonder if there is a support group or something&#8230;  Share [&#8230;]</p>
]]></content:encoded>
	</item>
	<item>
		<title>By: Pretty as Pie</title>
		<link>http://www.willmcgugan.com/2008/06/29/pretty-as-pie/#comment-28877</link>
		<dc:creator>Pretty as Pie</dc:creator>
		<pubDate>Mon, 30 Jun 2008 20:33:21 +0000</pubDate>
		<guid>http://www.willmcgugan.com/2008/06/29/pretty-as-pie/#comment-28877</guid>
		<description>[...] http://www.willmcgugan.com/2008/06/29/pretty-as-pie/ asks Hoosgot, [...]</description>
		<content:encoded><![CDATA[<p>[&#8230;] <a href="http://www.willmcgugan.com/2008/06/29/pretty-as-pie/" rel="nofollow">http://www.willmcgugan.com/2008/06/29/pretty-as-pie/</a> asks Hoosgot, [&#8230;]</p>
]]></content:encoded>
	</item>
	<item>
		<title>By: Toothy</title>
		<link>http://www.willmcgugan.com/2008/06/29/pretty-as-pie/#comment-28858</link>
		<dc:creator>Toothy</dc:creator>
		<pubDate>Mon, 30 Jun 2008 17:28:48 +0000</pubDate>
		<guid>http://www.willmcgugan.com/2008/06/29/pretty-as-pie/#comment-28858</guid>
		<description>Great chart.

Please ignore the overly pretentious ones who claim that it distorts anything. 

A piechart's role is to convey an approximate area, it does not matter if it is slightly distorted. One only cares about whether the areas are about equal, twice as big etc.

The image is gorgeous. There is a market for gorgeous stuff. Otherwise we all be using gnuplot ... yikes. 

But as far as charting libraries go, you'll need to have a large number of them, plus customization. It is a tough project.</description>
		<content:encoded><![CDATA[<p>Great chart.</p>
<p>Please ignore the overly pretentious ones who claim that it distorts anything. </p>
<p>A piechart&#8217;s role is to convey an approximate area, it does not matter if it is slightly distorted. One only cares about whether the areas are about equal, twice as big etc.</p>
<p>The image is gorgeous. There is a market for gorgeous stuff. Otherwise we all be using gnuplot &#8230; yikes. </p>
<p>But as far as charting libraries go, you&#8217;ll need to have a large number of them, plus customization. It is a tough project.</p>
]]></content:encoded>
	</item>
	<item>
		<title>By: Bill Karwin</title>
		<link>http://www.willmcgugan.com/2008/06/29/pretty-as-pie/#comment-28854</link>
		<dc:creator>Bill Karwin</dc:creator>
		<pubDate>Mon, 30 Jun 2008 16:16:04 +0000</pubDate>
		<guid>http://www.willmcgugan.com/2008/06/29/pretty-as-pie/#comment-28854</guid>
		<description>I agree with the other folks.  The pic you generated is gorgeous and belongs on the cover of a textbook for statistics or marketing or something.  But it's not useful to convey information, which is the purpose of a graph.

Even without the 3D foreshortening, pie charts are considered "low density and fail to order values along a visual dimension."  Distorting them with 3D, shadows, reflections, height differences, etc. just obscure the information worse.

You might want to read Edward Tufte's books, or the "Ask E.T" forum:
http://www.edwardtufte.com/bboard/q-and-a?topic_id=1</description>
		<content:encoded><![CDATA[<p>I agree with the other folks.  The pic you generated is gorgeous and belongs on the cover of a textbook for statistics or marketing or something.  But it&#8217;s not useful to convey information, which is the purpose of a graph.</p>
<p>Even without the 3D foreshortening, pie charts are considered &#8220;low density and fail to order values along a visual dimension.&#8221;  Distorting them with 3D, shadows, reflections, height differences, etc. just obscure the information worse.</p>
<p>You might want to read Edward Tufte&#8217;s books, or the &#8220;Ask E.T&#8221; forum:<br />
<a href="http://www.edwardtufte.com/bboard/q-and-a?topic_id=1" rel="nofollow">http://www.edwardtufte.com/bboard/q-and-a?topic_id=1</a></p>
]]></content:encoded>
	</item>
	<item>
		<title>By: wcyee</title>
		<link>http://www.willmcgugan.com/2008/06/29/pretty-as-pie/#comment-28853</link>
		<dc:creator>wcyee</dc:creator>
		<pubDate>Mon, 30 Jun 2008 15:58:36 +0000</pubDate>
		<guid>http://www.willmcgugan.com/2008/06/29/pretty-as-pie/#comment-28853</guid>
		<description>I'll add to the chorus, but go further: pie charts (2D included) rarely express anything about data that can't be expressed in far better ways through bars or stacked bars.  They're just visually appealing, which is why you really need to target business/marketing powerpoint types, looking to dress up their online presentations.  They love this stuff.</description>
		<content:encoded><![CDATA[<p>I&#8217;ll add to the chorus, but go further: pie charts (2D included) rarely express anything about data that can&#8217;t be expressed in far better ways through bars or stacked bars.  They&#8217;re just visually appealing, which is why you really need to target business/marketing powerpoint types, looking to dress up their online presentations.  They love this stuff.</p>
]]></content:encoded>
	</item>
	<item>
		<title>By: kib2</title>
		<link>http://www.willmcgugan.com/2008/06/29/pretty-as-pie/#comment-28839</link>
		<dc:creator>kib2</dc:creator>
		<pubDate>Mon, 30 Jun 2008 12:08:51 +0000</pubDate>
		<guid>http://www.willmcgugan.com/2008/06/29/pretty-as-pie/#comment-28839</guid>
		<description>Same as Nick, 
these diagrams are really beautiful, but totally loose their meaning in 3d.</description>
		<content:encoded><![CDATA[<p>Same as Nick,<br />
these diagrams are really beautiful, but totally loose their meaning in 3d.</p>
]]></content:encoded>
	</item>
	<item>
		<title>By: Will</title>
		<link>http://www.willmcgugan.com/2008/06/29/pretty-as-pie/#comment-28813</link>
		<dc:creator>Will</dc:creator>
		<pubDate>Mon, 30 Jun 2008 08:14:34 +0000</pubDate>
		<guid>http://www.willmcgugan.com/2008/06/29/pretty-as-pie/#comment-28813</guid>
		<description>John, maybe one day!

Nick, granted a 3D pie chart makes it a little harder to determin relative proportions than a 2D one, but I don't think they are useless -- people are used to viewing real world objects with foreshortening. I'm also looking in to ways of displaying the actual percentages over the pie chart...</description>
		<content:encoded><![CDATA[<p>John, maybe one day!</p>
<p>Nick, granted a 3D pie chart makes it a little harder to determin relative proportions than a 2D one, but I don&#8217;t think they are useless &#8212; people are used to viewing real world objects with foreshortening. I&#8217;m also looking in to ways of displaying the actual percentages over the pie chart&#8230;</p>
]]></content:encoded>
	</item>
	<item>
		<title>By: Nick Moffitt</title>
		<link>http://www.willmcgugan.com/2008/06/29/pretty-as-pie/#comment-28774</link>
		<dc:creator>Nick Moffitt</dc:creator>
		<pubDate>Sun, 29 Jun 2008 21:31:50 +0000</pubDate>
		<guid>http://www.willmcgugan.com/2008/06/29/pretty-as-pie/#comment-28774</guid>
		<description>This image makes me very sad.  Pie charts that are distorted along one axis (commonly for whizbang foreshortened 3d effects) do not have any correlation between the area of the slice and the value being displayed.  I was very disappointed when the google graphing URLs included 3D pie charts, as they're useless at conveying information in a way that can be understood by the reader.</description>
		<content:encoded><![CDATA[<p>This image makes me very sad.  Pie charts that are distorted along one axis (commonly for whizbang foreshortened 3d effects) do not have any correlation between the area of the slice and the value being displayed.  I was very disappointed when the google graphing URLs included 3D pie charts, as they&#8217;re useless at conveying information in a way that can be understood by the reader.</p>
]]></content:encoded>
	</item>
</channel>
</rss>
