<?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"
	>
<channel>
	<title>Comments on: Proving There are Only Six Dudeney Numbers</title>
	<atom:link href="http://hostilefork.com/2009/12/24/six-dudeney-numbers-proof/feed/" rel="self" type="application/rss+xml" />
	<link>http://hostilefork.com/2009/12/24/six-dudeney-numbers-proof/</link>
	<description>a disgruntled developer taking a stand in the information multiverse</description>
	<pubDate>Fri, 18 May 2012 04:55:26 +0000</pubDate>
	<generator>http://wordpress.org/?v=2.5</generator>
		<item>
		<title>By: Hostile Fork</title>
		<link>http://hostilefork.com/2009/12/24/six-dudeney-numbers-proof/#comment-1265</link>
		<dc:creator>Hostile Fork</dc:creator>
		<pubDate>Tue, 21 Jun 2011 11:41:52 +0000</pubDate>
		<guid isPermaLink="false">http://hostilefork.com/?p=131#comment-1265</guid>
		<description>@Cloudy0: Such formalism is absolutely valuable, so thanks!  Though once you start saying "derivative" and "ln" many people's eyes glaze over.  I don't know if you've read Lockhart's Lament or not, but his point about the "excessive" proof about intersecting lines and angles is something I'll mention:

http://www.maa.org/devlin/LockhartsLament.pdf

It would be nice if we could drill down into an argument to the level of our curiosity.  So if you want to know the exact zero you could click to see how to find it with the right equations, but if you're willing to accept a "good enough for the purpose" approximation then you could just keep reading.

Some day they'll finish &lt;i&gt;A Young Lady's Illustrated Primer&lt;/i&gt; :)

http://en.wikipedia.org/wiki/The_Diamond_Age</description>
		<content:encoded><![CDATA[<p>@Cloudy0: Such formalism is absolutely valuable, so thanks!  Though once you start saying &#8220;derivative&#8221; and &#8220;ln&#8221; many people&#8217;s eyes glaze over.  I don&#8217;t know if you&#8217;ve read Lockhart&#8217;s Lament or not, but his point about the &#8220;excessive&#8221; proof about intersecting lines and angles is something I&#8217;ll mention:</p>
<p><a href="http://www.maa.org/devlin/LockhartsLament.pdf" rel="nofollow" class="lipdf">http://www.maa.org/devlin/LockhartsLament.pdf</a></p>
<p>It would be nice if we could drill down into an argument to the level of our curiosity.  So if you want to know the exact zero you could click to see how to find it with the right equations, but if you&#8217;re willing to accept a &#8220;good enough for the purpose&#8221; approximation then you could just keep reading.</p>
<p>Some day they&#8217;ll finish <i>A Young Lady&#8217;s Illustrated Primer</i> <img src='http://hostilefork.com/wp-includes/images/smilies/icon_smile.gif' alt=':)' class='wp-smiley' /> </p>
<p><a href="http://en.wikipedia.org/wiki/The_Diamond_Age" rel="nofollow" target="_blank" class="liwikipedia">http://en.wikipedia.org/wiki/The_Diamond_Age</a></p>
]]></content:encoded>
	</item>
	<item>
		<title>By: Cloudy0</title>
		<link>http://hostilefork.com/2009/12/24/six-dudeney-numbers-proof/#comment-1263</link>
		<dc:creator>Cloudy0</dc:creator>
		<pubDate>Sun, 19 Jun 2011 14:47:52 +0000</pubDate>
		<guid isPermaLink="false">http://hostilefork.com/?p=131#comment-1263</guid>
		<description>"88 is the last number for which the right hand side is bigger than the left, and from then on we know that log(value)—even multiplied by a constant and added to another constant—will not have the possibility of catching up with value again."

Clarification:
Using Wolfram Alpha, you can find that the derivative of (36+27*log10(x))-x is 27/(x ln(10))-1, and this function stays at the negative part forever beyond a certain point. Therefore (36+27*log10(x))-x never goes up again.</description>
		<content:encoded><![CDATA[<p>&#8220;88 is the last number for which the right hand side is bigger than the left, and from then on we know that log(value)—even multiplied by a constant and added to another constant—will not have the possibility of catching up with value again.&#8221;</p>
<p>Clarification:<br />
Using Wolfram Alpha, you can find that the derivative of (36+27*log10(x))-x is 27/(x ln(10))-1, and this function stays at the negative part forever beyond a certain point. Therefore (36+27*log10(x))-x never goes up again.</p>
]]></content:encoded>
	</item>
	<item>
		<title>By: Hostile Fork</title>
		<link>http://hostilefork.com/2009/12/24/six-dudeney-numbers-proof/#comment-1259</link>
		<dc:creator>Hostile Fork</dc:creator>
		<pubDate>Mon, 11 Apr 2011 19:43:56 +0000</pubDate>
		<guid isPermaLink="false">http://hostilefork.com/?p=131#comment-1259</guid>
		<description>@steven - Sure!  From http://hostilefork.com/legal/

"All short code samples presented in articles, unless otherwise marked, do not require attribution to when used in your own software.  However, I will point out that it's often very nice (for maintenance reasons) to link to the source of such information in a comment."</description>
		<content:encoded><![CDATA[<p>@steven - Sure!  From <a href="http://hostilefork.com/legal/" rel="nofollow" class="liinternal">http://hostilefork.com/legal/</a></p>
<p>&#8220;All short code samples presented in articles, unless otherwise marked, do not require attribution to when used in your own software.  However, I will point out that it&#8217;s often very nice (for maintenance reasons) to link to the source of such information in a comment.&#8221;</p>
]]></content:encoded>
	</item>
	<item>
		<title>By: steven</title>
		<link>http://hostilefork.com/2009/12/24/six-dudeney-numbers-proof/#comment-1258</link>
		<dc:creator>steven</dc:creator>
		<pubDate>Sun, 10 Apr 2011 02:42:50 +0000</pubDate>
		<guid isPermaLink="false">http://hostilefork.com/?p=131#comment-1258</guid>
		<description>May I borrow your program?
And I must say your proof is quite wonderful!</description>
		<content:encoded><![CDATA[<p>May I borrow your program?<br />
And I must say your proof is quite wonderful!</p>
]]></content:encoded>
	</item>
	<item>
		<title>By: Hostile Fork</title>
		<link>http://hostilefork.com/2009/12/24/six-dudeney-numbers-proof/#comment-1241</link>
		<dc:creator>Hostile Fork</dc:creator>
		<pubDate>Mon, 20 Dec 2010 23:06:16 +0000</pubDate>
		<guid isPermaLink="false">http://hostilefork.com/?p=131#comment-1241</guid>
		<description>@Zomega: Indeed, it seems choosing that order for the substitutions gives fewer cases to test!</description>
		<content:encoded><![CDATA[<p>@Zomega: Indeed, it seems choosing that order for the substitutions gives fewer cases to test!</p>
]]></content:encoded>
	</item>
	<item>
		<title>By: Zomega</title>
		<link>http://hostilefork.com/2009/12/24/six-dudeney-numbers-proof/#comment-1238</link>
		<dc:creator>Zomega</dc:creator>
		<pubDate>Sat, 20 Nov 2010 01:20:19 +0000</pubDate>
		<guid isPermaLink="false">http://hostilefork.com/?p=131#comment-1238</guid>
		<description>I very much like your proof, but I have a minor improvement (unless I've missed something) that reduces the bound. Simply put, you substitute in for DigitCount(value^3) when you don't have to. The log term in your bound on DigitCount will nicely remove the power for you, which decreases the constant term involved from 36 to 9, and the bound to about 55.

DigitSum(value^3)
        &#60;= 9 * DigitCount(value^3) 
        &#60;=9*(log(value^3) + 1)
        &#60;=9*(3*log(value) + 1)
        &#60;=27*log(value) + 9

&lt;a href="http://www.wolframalpha.com/input/?i=%289+%2B+27+*+log_10%28x%29%29+-+x+from+1+to+100" rel="nofollow"&gt;www.wolframalpha.com/input/?i=%289+%2B+27+*+log_10%28x%29%29+-+x+from+1+to+100&lt;/a&gt;

In terms of your proof, this bound doesn't matter much, but it starts to be important if, like me, you're looking at the equivalent of Dudeney numbers with powers much higher than 3.

~Zomega</description>
		<content:encoded><![CDATA[<p>I very much like your proof, but I have a minor improvement (unless I&#8217;ve missed something) that reduces the bound. Simply put, you substitute in for DigitCount(value^3) when you don&#8217;t have to. The log term in your bound on DigitCount will nicely remove the power for you, which decreases the constant term involved from 36 to 9, and the bound to about 55.</p>
<p>DigitSum(value^3)<br />
        &lt;= 9 * DigitCount(value^3)<br />
        &lt;=9*(log(value^3) + 1)<br />
        &lt;=9*(3*log(value) + 1)<br />
        &lt;=27*log(value) + 9</p>
<p><a href="http://www.wolframalpha.com/input/?i=%289+%2B+27+*+log_10%28x%29%29+-+x+from+1+to+100" rel="nofollow" target="_blank" class="liexternal"></a><a href="http://www.wolframalpha.com/input/?i=%289+%2B+27+" rel="nofollow">http://www.wolframalpha.com/input/?i=%289+%2B+27+</a>*+log_10%28x%29%29+-+x+from+1+to+100</p>
<p>In terms of your proof, this bound doesn&#8217;t matter much, but it starts to be important if, like me, you&#8217;re looking at the equivalent of Dudeney numbers with powers much higher than 3.</p>
<p>~Zomega</p>
]]></content:encoded>
	</item>
	<item>
		<title>By: DideC</title>
		<link>http://hostilefork.com/2009/12/24/six-dudeney-numbers-proof/#comment-1033</link>
		<dc:creator>DideC</dc:creator>
		<pubDate>Wed, 13 Jan 2010 16:46:45 +0000</pubDate>
		<guid isPermaLink="false">http://hostilefork.com/?p=131#comment-1033</guid>
		<description>Parentheses, yes there is useless one but I imagine it is for clarity purpose.

Anyway, nice demonstration !

Here is just another implementation for interleave-block :

interleave-block: func [block [block!] item /local pos] [
    loop subtract length? block 1 [
    	block: insert next block item
    ]
    return head block
]

if block length is 1 or 0, then no loop occurs, so no item is inserted.</description>
		<content:encoded><![CDATA[<p>Parentheses, yes there is useless one but I imagine it is for clarity purpose.</p>
<p>Anyway, nice demonstration !</p>
<p>Here is just another implementation for interleave-block :</p>
<p>interleave-block: func [block [block!] item /local pos] [<br />
    loop subtract length? block 1 [<br />
    	block: insert next block item<br />
    ]<br />
    return head block<br />
]</p>
<p>if block length is 1 or 0, then no loop occurs, so no item is inserted.</p>
]]></content:encoded>
	</item>
	<item>
		<title>By: Hostile Fork</title>
		<link>http://hostilefork.com/2009/12/24/six-dudeney-numbers-proof/#comment-1024</link>
		<dc:creator>Hostile Fork</dc:creator>
		<pubDate>Thu, 24 Dec 2009 22:52:59 +0000</pubDate>
		<guid isPermaLink="false">http://hostilefork.com/?p=131#comment-1024</guid>
		<description>Thanks Gregg!

It was interesting for me to try this.  I didn't study proofs of this kind in school (at least not that I remember :P).  But I've read enough Wikipedia math articles on map coloring and distribution of primes to now grok the strategy of attack.  Reading up on hard problems might not get you measurably closer to solving other unsolved problems, but it makes little puzzles like this look trivial by comparison!

I fixed up the code a little...but if anyone has ideas to improve it (that &lt;i&gt;don't&lt;/i&gt; involve taking out parentheses) let me know!</description>
		<content:encoded><![CDATA[<p>Thanks Gregg!</p>
<p>It was interesting for me to try this.  I didn&#8217;t study proofs of this kind in school (at least not that I remember :P).  But I&#8217;ve read enough Wikipedia math articles on map coloring and distribution of primes to now grok the strategy of attack.  Reading up on hard problems might not get you measurably closer to solving other unsolved problems, but it makes little puzzles like this look trivial by comparison!</p>
<p>I fixed up the code a little&#8230;but if anyone has ideas to improve it (that <i>don&#8217;t</i> involve taking out parentheses) let me know!</p>
]]></content:encoded>
	</item>
	<item>
		<title>By: Gregg Irwin</title>
		<link>http://hostilefork.com/2009/12/24/six-dudeney-numbers-proof/#comment-1023</link>
		<dc:creator>Gregg Irwin</dc:creator>
		<pubDate>Thu, 24 Dec 2009 18:06:19 +0000</pubDate>
		<guid isPermaLink="false">http://hostilefork.com/?p=131#comment-1023</guid>
		<description>Great article. It's wonderful to see the thought process explained, so others can really learn from it.</description>
		<content:encoded><![CDATA[<p>Great article. It&#8217;s wonderful to see the thought process explained, so others can really learn from it.</p>
]]></content:encoded>
	</item>
</channel>
</rss>

