<?xml version="1.0" encoding="UTF-8"?><rss version="2.0"
	xmlns:content="http://purl.org/rss/1.0/modules/content/"
	xmlns:dc="http://purl.org/dc/elements/1.1/"
	xmlns:atom="http://www.w3.org/2005/Atom"
	xmlns:sy="http://purl.org/rss/1.0/modules/syndication/"
		>
<channel>
	<title>Comments on: ActionScript 3 Development Task Contest #1 Entries</title>
	<atom:link href="http://www.mikechambers.com/blog/2009/11/13/actionscript-3-development-task-contest-1-entries/feed/" rel="self" type="application/rss+xml" />
	<link>http://www.mikechambers.com/blog/2009/11/13/actionscript-3-development-task-contest-1-entries/</link>
	<description>code = joy</description>
	<lastBuildDate>Thu, 29 Jul 2010 22:29:12 +0000</lastBuildDate>
	<sy:updatePeriod>hourly</sy:updatePeriod>
	<sy:updateFrequency>1</sy:updateFrequency>
	<generator>http://wordpress.org/?v=3.0</generator>
	<item>
		<title>By: Sharedtut</title>
		<link>http://www.mikechambers.com/blog/2009/11/13/actionscript-3-development-task-contest-1-entries/comment-page-1/#comment-18399</link>
		<dc:creator>Sharedtut</dc:creator>
		<pubDate>Mon, 01 Feb 2010 04:25:25 +0000</pubDate>
		<guid isPermaLink="false">http://www.mikechambers.com/blog/?p=1922#comment-18399</guid>
		<description>the operator is faster.</description>
		<content:encoded><![CDATA[<p>the operator is faster.</p>
]]></content:encoded>
	</item>
	<item>
		<title>By: Brandon</title>
		<link>http://www.mikechambers.com/blog/2009/11/13/actionscript-3-development-task-contest-1-entries/comment-page-1/#comment-17785</link>
		<dc:creator>Brandon</dc:creator>
		<pubDate>Mon, 16 Nov 2009 19:50:15 +0000</pubDate>
		<guid isPermaLink="false">http://www.mikechambers.com/blog/?p=1922#comment-17785</guid>
		<description>@Glassgiant

I&#039;m guessing the bitwise operator is faster than the cast.</description>
		<content:encoded><![CDATA[<p>@Glassgiant</p>
<p>I&#8217;m guessing the bitwise operator is faster than the cast.</p>
]]></content:encoded>
	</item>
	<item>
		<title>By: Glassgiant</title>
		<link>http://www.mikechambers.com/blog/2009/11/13/actionscript-3-development-task-contest-1-entries/comment-page-1/#comment-17781</link>
		<dc:creator>Glassgiant</dc:creator>
		<pubDate>Mon, 16 Nov 2009 15:47:27 +0000</pubDate>
		<guid isPermaLink="false">http://www.mikechambers.com/blog/?p=1922#comment-17781</guid>
		<description>I actually don&#039;t get the real use of the &quot;&#124; 0&quot; ( var pos:uint = (item.x*m&#124;0)*h+item.y*m; ) bitwise operator in the example of G. Skinner. Just a matter of beauty or what&#039;s the difference to simply use int(x)?</description>
		<content:encoded><![CDATA[<p>I actually don&#8217;t get the real use of the &#8220;| 0&#8243; ( var pos:uint = (item.x*m|0)*h+item.y*m; ) bitwise operator in the example of G. Skinner. Just a matter of beauty or what&#8217;s the difference to simply use int(x)?</p>
]]></content:encoded>
	</item>
	<item>
		<title>By: Steve Sedlmayr</title>
		<link>http://www.mikechambers.com/blog/2009/11/13/actionscript-3-development-task-contest-1-entries/comment-page-1/#comment-17749</link>
		<dc:creator>Steve Sedlmayr</dc:creator>
		<pubDate>Sun, 15 Nov 2009 00:49:48 +0000</pubDate>
		<guid isPermaLink="false">http://www.mikechambers.com/blog/?p=1922#comment-17749</guid>
		<description>BTW, Marcin Szczepanski pointed out that the times I got were probably so lengthy because I was using the debug version of Flash Player; actual times should be about a third as much.</description>
		<content:encoded><![CDATA[<p>BTW, Marcin Szczepanski pointed out that the times I got were probably so lengthy because I was using the debug version of Flash Player; actual times should be about a third as much.</p>
]]></content:encoded>
	</item>
	<item>
		<title>By: Steve Sedlmayr</title>
		<link>http://www.mikechambers.com/blog/2009/11/13/actionscript-3-development-task-contest-1-entries/comment-page-1/#comment-17748</link>
		<dc:creator>Steve Sedlmayr</dc:creator>
		<pubDate>Sun, 15 Nov 2009 00:47:50 +0000</pubDate>
		<guid isPermaLink="false">http://www.mikechambers.com/blog/?p=1922#comment-17748</guid>
		<description>It seems to me now you would want to parameterize the enumerable property names you are using to evaluate proximity as Strings in addition to the gridSize and bounds, i.e. something other than x and y. You could then solve a similar problem for any two kinds of data that can be plotted on a two-dimensional Cartesian coordinate system. For instance, say you had a plot of all longitudes and latitudes and wanted to find all of the longitude and latitude coordinates nearby any particular coordinate within a certain distance (as determined by gridSize). You could do the same thing with, say, zip codes.

Of course, those are both still coordinates plotted on a physical plane, but it could apply to any 2 data points; for instance morbidity rates for smallpox by month. If you used 12 as your gridSize, you would get similar infection rates for smallpox within roughly 18 months before or after your data point. If each object in the Vector had a third piece of data, say the name of a clinic or hospital where the rate occurred, you would now have a list of all locations that experienced similar morbidity rates during a particular time period to the clinic or hospital you passed in. If this value or another one were a guid, you could then look up more detailed information about each clinic in a database: addresses, contact info, case notes, and so on. Speaking of which, in another example, your axises could simply be columns and rows in a large database table in order to find nearby records.

It also seems like, for large enough sets, you could nest grids within grids, arbitrarily deep, at some optimal size per grid to focus in on data in a way that as the overall amount of points increases by some exponent, the amount of time to solve for neighboring data only increases arithmetically in proportion to the number of levels in your hierarchy. It would only work if each grid collection had metadata about the range of data it contained, though, so it could be treated as a point itself. Otherwise it would be no different than a multidimensional array. I&#039;m not sure if this corresponds to some existing data algorithm or what you would use it for; it would have to be a huge amount of data.</description>
		<content:encoded><![CDATA[<p>It seems to me now you would want to parameterize the enumerable property names you are using to evaluate proximity as Strings in addition to the gridSize and bounds, i.e. something other than x and y. You could then solve a similar problem for any two kinds of data that can be plotted on a two-dimensional Cartesian coordinate system. For instance, say you had a plot of all longitudes and latitudes and wanted to find all of the longitude and latitude coordinates nearby any particular coordinate within a certain distance (as determined by gridSize). You could do the same thing with, say, zip codes.</p>
<p>Of course, those are both still coordinates plotted on a physical plane, but it could apply to any 2 data points; for instance morbidity rates for smallpox by month. If you used 12 as your gridSize, you would get similar infection rates for smallpox within roughly 18 months before or after your data point. If each object in the Vector had a third piece of data, say the name of a clinic or hospital where the rate occurred, you would now have a list of all locations that experienced similar morbidity rates during a particular time period to the clinic or hospital you passed in. If this value or another one were a guid, you could then look up more detailed information about each clinic in a database: addresses, contact info, case notes, and so on. Speaking of which, in another example, your axises could simply be columns and rows in a large database table in order to find nearby records.</p>
<p>It also seems like, for large enough sets, you could nest grids within grids, arbitrarily deep, at some optimal size per grid to focus in on data in a way that as the overall amount of points increases by some exponent, the amount of time to solve for neighboring data only increases arithmetically in proportion to the number of levels in your hierarchy. It would only work if each grid collection had metadata about the range of data it contained, though, so it could be treated as a point itself. Otherwise it would be no different than a multidimensional array. I&#8217;m not sure if this corresponds to some existing data algorithm or what you would use it for; it would have to be a huge amount of data.</p>
]]></content:encoded>
	</item>
	<item>
		<title>By: Arnaud Gatouillat</title>
		<link>http://www.mikechambers.com/blog/2009/11/13/actionscript-3-development-task-contest-1-entries/comment-page-1/#comment-17744</link>
		<dc:creator>Arnaud Gatouillat</dc:creator>
		<pubDate>Sat, 14 Nov 2009 18:18:31 +0000</pubDate>
		<guid isPermaLink="false">http://www.mikechambers.com/blog/?p=1922#comment-17744</guid>
		<description>@Grégory

yes they&#039;re equivalent, but only if lengths are constrained within [0,127]. If not the signifiant bits of pos and length could overlap.</description>
		<content:encoded><![CDATA[<p>@Grégory</p>
<p>yes they&#8217;re equivalent, but only if lengths are constrained within [0,127]. If not the signifiant bits of pos and length could overlap.</p>
]]></content:encoded>
	</item>
	<item>
		<title>By: Andrew Traviss</title>
		<link>http://www.mikechambers.com/blog/2009/11/13/actionscript-3-development-task-contest-1-entries/comment-page-1/#comment-17743</link>
		<dc:creator>Andrew Traviss</dc:creator>
		<pubDate>Sat, 14 Nov 2009 18:12:14 +0000</pubDate>
		<guid isPermaLink="false">http://www.mikechambers.com/blog/?p=1922#comment-17743</guid>
		<description>I didn&#039;t do quite as badly as I expected. I made the mistake of comparing my time to Grant&#039;s first and thought I&#039;d been completely flattened.

I&#039;m surprised more entries didn&#039;t use lazy evaluation of objects. If update() gets called multiple times without a getNeighbors() call between them, you&#039;ve wasted your time.</description>
		<content:encoded><![CDATA[<p>I didn&#8217;t do quite as badly as I expected. I made the mistake of comparing my time to Grant&#8217;s first and thought I&#8217;d been completely flattened.</p>
<p>I&#8217;m surprised more entries didn&#8217;t use lazy evaluation of objects. If update() gets called multiple times without a getNeighbors() call between them, you&#8217;ve wasted your time.</p>
]]></content:encoded>
	</item>
	<item>
		<title>By: Grégory Pelletier</title>
		<link>http://www.mikechambers.com/blog/2009/11/13/actionscript-3-development-task-contest-1-entries/comment-page-1/#comment-17741</link>
		<dc:creator>Grégory Pelletier</dc:creator>
		<pubDate>Sat, 14 Nov 2009 12:44:00 +0000</pubDate>
		<guid isPermaLink="false">http://www.mikechambers.com/blog/?p=1922#comment-17741</guid>
		<description>Participate to this contest was very interesting! I haven&#039;t seen all the versions but I learn a lot of those I looked. Now I feel like coding for FP 10 !
I have a small question, in the class of Grant, I wonder if &quot;pos&lt;&lt;7&#124;lengths[pos]++&quot; is a equivalent to &quot;pos*128 + lengths[pos]++&quot; ?</description>
		<content:encoded><![CDATA[<p>Participate to this contest was very interesting! I haven&#8217;t seen all the versions but I learn a lot of those I looked. Now I feel like coding for FP 10 !<br />
I have a small question, in the class of Grant, I wonder if &#8220;pos&lt;&lt;7|lengths[pos]++&quot; is a equivalent to &quot;pos*128 + lengths[pos]++&quot; ?</p>
]]></content:encoded>
	</item>
	<item>
		<title>By: Jonas</title>
		<link>http://www.mikechambers.com/blog/2009/11/13/actionscript-3-development-task-contest-1-entries/comment-page-1/#comment-17740</link>
		<dc:creator>Jonas</dc:creator>
		<pubDate>Sat, 14 Nov 2009 09:56:28 +0000</pubDate>
		<guid isPermaLink="false">http://www.mikechambers.com/blog/?p=1922#comment-17740</guid>
		<description>Does 4 getNeighbors() correspond to a real use case ? If displayObject are not moving (less update) I think no. If they are moving probably yes...

Héhé my entry is better with 100.</description>
		<content:encoded><![CDATA[<p>Does 4 getNeighbors() correspond to a real use case ? If displayObject are not moving (less update) I think no. If they are moving probably yes&#8230;</p>
<p>Héhé my entry is better with 100.</p>
]]></content:encoded>
	</item>
	<item>
		<title>By: Mihai Alexandru Bîrsan</title>
		<link>http://www.mikechambers.com/blog/2009/11/13/actionscript-3-development-task-contest-1-entries/comment-page-1/#comment-17739</link>
		<dc:creator>Mihai Alexandru Bîrsan</dc:creator>
		<pubDate>Sat, 14 Nov 2009 09:46:19 +0000</pubDate>
		<guid isPermaLink="false">http://www.mikechambers.com/blog/?p=1922#comment-17739</guid>
		<description>I expected my implementation to be among the last, since I&#039;m accounting for DO coordinate changes between calls to getNeighbors, but I guess some people did worse according to Steve Sedlmayr&#039;s test run. http://ssedlmayr.wordpress.com/2009/11/13/as3dtc-p0wn3d-me/

Next contest I promise not to be lazy and update my implementation when presented with new information. :D</description>
		<content:encoded><![CDATA[<p>I expected my implementation to be among the last, since I&#8217;m accounting for DO coordinate changes between calls to getNeighbors, but I guess some people did worse according to Steve Sedlmayr&#8217;s test run. <a href="http://ssedlmayr.wordpress.com/2009/11/13/as3dtc-p0wn3d-me/" rel="nofollow">http://ssedlmayr.wordpress.com/2009/11/13/as3dtc-p0wn3d-me/</a></p>
<p>Next contest I promise not to be lazy and update my implementation when presented with new information. :D</p>
]]></content:encoded>
	</item>
</channel>
</rss>
