<?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: Stochastic, Nerdtastic Restaurant Bill Splitting</title>
	<atom:link href="http://messymatters.com/expectorant/feed/" rel="self" type="application/rss+xml" />
	<link>http://messymatters.com/expectorant/?utm_source=rss&#038;utm_medium=rss&#038;utm_campaign=expectorant</link>
	<description>Bring your own data</description>
	<lastBuildDate>Thu, 16 May 2013 16:01:02 +0000</lastBuildDate>
	<sy:updatePeriod>hourly</sy:updatePeriod>
	<sy:updateFrequency>1</sy:updateFrequency>
	<generator>http://wordpress.org/?v=3.5.1</generator>
	<item>
		<title>By: SirPumpkinLongshanks</title>
		<link>http://messymatters.com/expectorant/#comment-172977</link>
		<dc:creator>SirPumpkinLongshanks</dc:creator>
		<pubDate>Fri, 23 Nov 2012 07:52:00 +0000</pubDate>
		<guid isPermaLink="false">http://messymatters.com/?p=1611#comment-172977</guid>
		<description><![CDATA[If you found this article a tiny bit sesible, then you must definitlely check out http://apps.facebook.com/thepayup - it&#039;s a natural continuation of whats in this post
cheers!]]></description>
		<content:encoded><![CDATA[<p>If you found this article a tiny bit sesible, then you must definitlely check out <a href="http://apps.facebook.com/thepayup" rel="nofollow">http://apps.facebook.com/thepayup</a> &#8211; it&#8217;s a natural continuation of whats in this post<br />
cheers!</p>
]]></content:encoded>
	</item>
	<item>
		<title>By: Bittner vs Hughes &#8211; Nerdtastic Restaurant Bill Splitting &#124; Splitwise Blog</title>
		<link>http://messymatters.com/expectorant/#comment-172582</link>
		<dc:creator>Bittner vs Hughes &#8211; Nerdtastic Restaurant Bill Splitting &#124; Splitwise Blog</dc:creator>
		<pubDate>Thu, 09 Aug 2012 15:03:44 +0000</pubDate>
		<guid isPermaLink="false">http://messymatters.com/?p=1611#comment-172582</guid>
		<description><![CDATA[[...] The problem is that the math you are talking about is pretty hard, even for me. So here is this article on messymatters.com about a way to split your restaurant bill using the [...]]]></description>
		<content:encoded><![CDATA[<p>[...] The problem is that the math you are talking about is pretty hard, even for me. So here is this article on messymatters.com about a way to split your restaurant bill using the [...]</p>
]]></content:encoded>
	</item>
	<item>
		<title>By: Daniel Reeves</title>
		<link>http://messymatters.com/expectorant/#comment-172553</link>
		<dc:creator>Daniel Reeves</dc:creator>
		<pubDate>Thu, 26 Jul 2012 05:36:00 +0000</pubDate>
		<guid isPermaLink="false">http://messymatters.com/?p=1611#comment-172553</guid>
		<description><![CDATA[Doh! I just proposed this when out to eat with my sister and brother-in-law and they pointed out that this totally doesn&#039;t work!  (We did the normal version of Expectorant and they liked it. :))

The problem is that you can&#039;t wait till you&#039;re on the spot to post the collateral. Consider the extreme case that the mechanism passes up most of the items and now you&#039;re up with a 99% chance of paying the bill. If you can now decide to post collateral then you&#039;ve totally shirked your fair probability!]]></description>
		<content:encoded><![CDATA[<p>Doh! I just proposed this when out to eat with my sister and brother-in-law and they pointed out that this totally doesn&#8217;t work!  (We did the normal version of Expectorant and they liked it. :))</p>
<p>The problem is that you can&#8217;t wait till you&#8217;re on the spot to post the collateral. Consider the extreme case that the mechanism passes up most of the items and now you&#8217;re up with a 99% chance of paying the bill. If you can now decide to post collateral then you&#8217;ve totally shirked your fair probability!</p>
]]></content:encoded>
	</item>
	<item>
		<title>By: Daniel Reeves</title>
		<link>http://messymatters.com/expectorant/#comment-172550</link>
		<dc:creator>Daniel Reeves</dc:creator>
		<pubDate>Mon, 23 Jul 2012 23:36:00 +0000</pubDate>
		<guid isPermaLink="false">http://messymatters.com/?p=1611#comment-172550</guid>
		<description><![CDATA[Good question. Benford&#039;s Law -- http://en.wikipedia.org/wiki/Benford&#039;s_law -- might be at play. I bet that reversing the serial number would suffice though.]]></description>
		<content:encoded><![CDATA[<p>Good question. Benford&#8217;s Law &#8212; <a href="http://en.wikipedia.org/wiki/Benford&#039;s_law" rel="nofollow">http://en.wikipedia.org/wiki/Benford&#039;s_law</a> &#8212; might be at play. I bet that reversing the serial number would suffice though.</p>
]]></content:encoded>
	</item>
	<item>
		<title>By: Marcel Nijman</title>
		<link>http://messymatters.com/expectorant/#comment-172542</link>
		<dc:creator>Marcel Nijman</dc:creator>
		<pubDate>Sun, 22 Jul 2012 14:37:00 +0000</pubDate>
		<guid isPermaLink="false">http://messymatters.com/?p=1611#comment-172542</guid>
		<description><![CDATA[I guess if you add the time required to explain the procedure (and the validity of it) you&#039;d be much faster just calculating the exact split...]]></description>
		<content:encoded><![CDATA[<p>I guess if you add the time required to explain the procedure (and the validity of it) you&#8217;d be much faster just calculating the exact split&#8230;</p>
]]></content:encoded>
	</item>
	<item>
		<title>By: David Pennock</title>
		<link>http://messymatters.com/expectorant/#comment-172540</link>
		<dc:creator>David Pennock</dc:creator>
		<pubDate>Sat, 21 Jul 2012 11:01:00 +0000</pubDate>
		<guid isPermaLink="false">http://messymatters.com/?p=1611#comment-172540</guid>
		<description><![CDATA[&#039;dish of doom&#039; -- love it. 

Are bill serial numbers uniform?]]></description>
		<content:encoded><![CDATA[<p>&#8216;dish of doom&#8217; &#8212; love it. </p>
<p>Are bill serial numbers uniform?</p>
]]></content:encoded>
	</item>
	<item>
		<title>By: Daniel Reeves</title>
		<link>http://messymatters.com/expectorant/#comment-172537</link>
		<dc:creator>Daniel Reeves</dc:creator>
		<pubDate>Fri, 20 Jul 2012 23:24:00 +0000</pubDate>
		<guid isPermaLink="false">http://messymatters.com/?p=1611#comment-172537</guid>
		<description><![CDATA[Check out Carlos Bueno&#039;s comment here from 2 days ago, with a variant of credit card roulette using serial numbers on dollar bills. I think that would work nicely for lmm&#039;s spinner method. Put a decimal point in front of the serial number -- or better, at the end of the serial number and read it backwards -- on a randomly chosen dollar bill and multiply that by the subtotal.


I&#039;m also eager to try this in the field! Maybe best to avoid the mental sort and go strictly by the order on the bill. Then when you identify the Dish of Doom you have to make sure to identify any other Dish of Doom Duplicates. Let&#039;s say we order people clockwise starting on the gamemaster&#039;s left. Then we can assign the Dish of Doom to the right person among those who ordered Dish of Doom Duplicates.
]]></description>
		<content:encoded><![CDATA[<p>Check out Carlos Bueno&#8217;s comment here from 2 days ago, with a variant of credit card roulette using serial numbers on dollar bills. I think that would work nicely for lmm&#8217;s spinner method. Put a decimal point in front of the serial number &#8212; or better, at the end of the serial number and read it backwards &#8212; on a randomly chosen dollar bill and multiply that by the subtotal.</p>
<p>I&#8217;m also eager to try this in the field! Maybe best to avoid the mental sort and go strictly by the order on the bill. Then when you identify the Dish of Doom you have to make sure to identify any other Dish of Doom Duplicates. Let&#8217;s say we order people clockwise starting on the gamemaster&#8217;s left. Then we can assign the Dish of Doom to the right person among those who ordered Dish of Doom Duplicates.</p>
]]></content:encoded>
	</item>
	<item>
		<title>By: David Pennock</title>
		<link>http://messymatters.com/expectorant/#comment-172536</link>
		<dc:creator>David Pennock</dc:creator>
		<pubDate>Fri, 20 Jul 2012 22:34:00 +0000</pubDate>
		<guid isPermaLink="false">http://messymatters.com/?p=1611#comment-172536</guid>
		<description><![CDATA[I love Arthur&#039;s &quot;post collateral&quot; addition. Brilliant indeed.
I love the HN debate.
I like lmm&#039;s faster method but I think it has a few more drawbacks than Daniel identified:
  * It requires a trusted random number generator. Expectorant does not since the person picks a number from 1-20.
  * Unless you always scan from top to bottom (which is inefficient) or always scan *exactly* in descending price order (which requires a mental sort) it&#039;s easier to game. Whoever is picking items can &quot;accidentally&quot; skip their own item if they are near the threshold.
 * If you scan in descending price order, you have to group items by equal price, not just by identical items. In other words, if there are two coffees and a tea, all the same price, then you have to treat them as a group of 3 and use your predefined method of ordering the 3 people. You can&#039;t treat the coffees as a group of 2, then the tea, otherwise gaming could be going on. (Or you need a predefined sort on items too, like coffee before tea (alphabetical), but this gets kind of ridiculous in terms of number of tie breaking rules.
Still, it&#039;s definitely worth trying &quot;in the field&quot; to see if it works well/better.]]></description>
		<content:encoded><![CDATA[<p>I love Arthur&#8217;s &#8220;post collateral&#8221; addition. Brilliant indeed.<br />
I love the HN debate.<br />
I like lmm&#8217;s faster method but I think it has a few more drawbacks than Daniel identified:<br />
  * It requires a trusted random number generator. Expectorant does not since the person picks a number from 1-20.<br />
  * Unless you always scan from top to bottom (which is inefficient) or always scan *exactly* in descending price order (which requires a mental sort) it&#8217;s easier to game. Whoever is picking items can &#8220;accidentally&#8221; skip their own item if they are near the threshold.<br />
 * If you scan in descending price order, you have to group items by equal price, not just by identical items. In other words, if there are two coffees and a tea, all the same price, then you have to treat them as a group of 3 and use your predefined method of ordering the 3 people. You can&#8217;t treat the coffees as a group of 2, then the tea, otherwise gaming could be going on. (Or you need a predefined sort on items too, like coffee before tea (alphabetical), but this gets kind of ridiculous in terms of number of tie breaking rules.<br />
Still, it&#8217;s definitely worth trying &#8220;in the field&#8221; to see if it works well/better.</p>
]]></content:encoded>
	</item>
	<item>
		<title>By: Zed</title>
		<link>http://messymatters.com/expectorant/#comment-172526</link>
		<dc:creator>Zed</dc:creator>
		<pubDate>Thu, 19 Jul 2012 11:26:00 +0000</pubDate>
		<guid isPermaLink="false">http://messymatters.com/?p=1611#comment-172526</guid>
		<description><![CDATA[The notion is that it&#039;s much slower (or more boring). However, I suspect that though the algorithmic complexity may be worse, it parallelizes a lot better than having one chump tap away at his phone.]]></description>
		<content:encoded><![CDATA[<p>The notion is that it&#8217;s much slower (or more boring). However, I suspect that though the algorithmic complexity may be worse, it parallelizes a lot better than having one chump tap away at his phone.</p>
]]></content:encoded>
	</item>
	<item>
		<title>By: Daniel Reeves</title>
		<link>http://messymatters.com/expectorant/#comment-172524</link>
		<dc:creator>Daniel Reeves</dc:creator>
		<pubDate>Thu, 19 Jul 2012 04:46:00 +0000</pubDate>
		<guid isPermaLink="false">http://messymatters.com/?p=1611#comment-172524</guid>
		<description><![CDATA[Yeah, if you can decide that tax+tip is, say, 25% then when the person puts the $15 on the table, you enter &quot;17-15/1.25&quot; as their number for expectorant. Not so bad.


Then if you actually put down $17 + 25% = $21.25 in cash then what you type into expectorant is 17-21.25/1.25 which is zero, as it should be. That person covered their item and has no risk of paying the whole bill.]]></description>
		<content:encoded><![CDATA[<p>Yeah, if you can decide that tax+tip is, say, 25% then when the person puts the $15 on the table, you enter &#8220;17-15/1.25&#8243; as their number for expectorant. Not so bad.</p>
<p>Then if you actually put down $17 + 25% = $21.25 in cash then what you type into expectorant is 17-21.25/1.25 which is zero, as it should be. That person covered their item and has no risk of paying the whole bill.</p>
]]></content:encoded>
	</item>
</channel>
</rss>
