<?xml version="1.0" encoding="UTF-8"?>
<!-- generator="bbPress/1.0.2" -->
<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>Dipsters.org Forum &#187; Topic: USAK and Social Networks</title>
		<link>http://www.dipsters.org/forum/topic/usak-and-social-networks</link>
		<description>First please read the post, &quot;Welcome! Using the Forum&quot;.</description>
		<language>en-US</language>
		<pubDate>Sun, 05 Sep 2010 03:51:29 +0000</pubDate>
		<generator>http://bbpress.org/?v=1.0.2</generator>
		<textInput>
			<title><![CDATA[Search]]></title>
			<description><![CDATA[Search all topics from these forums.]]></description>
			<name>q</name>
			<link>http://www.dipsters.org/forum/search.php</link>
		</textInput>
		<atom:link href="http://www.dipsters.org/forum/rss/topic/usak-and-social-networks" rel="self" type="application/rss+xml" />

		<item>
			<title>Chris Babcock on "USAK and Social Networks"</title>
			<link>http://www.dipsters.org/forum/topic/usak-and-social-networks#post-187</link>
			<pubDate>Sat, 06 Mar 2010 19:56:58 +0000</pubDate>
			<dc:creator>Chris Babcock</dc:creator>
			<guid isPermaLink="false">187@http://www.dipsters.org/forum/</guid>
			<description>&#60;p&#62;Become a fan of USAK on Facebook and follow USAK on Twitter:&#60;br /&#62;
   &#60;a href=&#34;http://www.facebook.com/pages/Diplomacy-on-USAK/328122573511&#34; rel=&#34;nofollow&#34;&#62;http://www.facebook.com/pages/Diplomacy-on-USAK/328122573511&#60;/a&#62;&#60;br /&#62;
   &#60;a href=&#34;http://twitter.com/USAK_Diplomacy&#34; rel=&#34;nofollow&#34;&#62;http://twitter.com/USAK_Diplomacy&#60;/a&#62;&#60;/p&#62;
&#60;p&#62;The Facebook page is a trial to promote the judge and see if any particular feature is a desirable for implementation on the local judge site.&#60;/p&#62;
&#60;p&#62;The Twitter stream is a feed of game notices - results for all hosted games and openings for listed games. I expect it to help fill openings faster and improve player retention. If it becomes widely used then I can add other features like authentication and individual notices.&#60;/p&#62;
&#60;p&#62;I've also tested XMPP notifications. One of the weaknesses of play by web Diplomacy is an inability to push notices when players forget to check the site. XMPP is a good push mechanism, especially for Gmail users, because it is almost as much a part of many players' online routine as email, but it does not depend on polling to deliver the message when the player is online. XMPP addresses take the form of xmpp:user@example.com, which matches the judge's definition of an email address. While XMPP can usually handle a message the size of a judge email, that can be a little inconvenient. The best way to play over XMPP will be to make reports and press available on the web site and send notices when new playing content is available.&#60;/p&#62;
&#60;p&#62;One advantage of XMPP and services like Twitter over play by email is that it is possible to know if the registered account is still in use. The judge can clean out unused accounts and publish abandoned positions the minute a player severs a reciprocal 'friendship' with the judge's bot and the timing of a player's decision to abandon can be taken into account in a judgekeeper's treatment of the account. A player who breaks contact with the judge after he has been replaced in a game has suffered from an accident or poor time management while a player who leave the judge when he sees games results has gone deliberately abandoned.&#60;/p&#62;
&#60;p&#62;Long term, I'm looking at using forensic social network analysis and lexical analysis to protect against collusion in games that are any-site. Forensic SNA compares real social networks with disclosed social networks to discover when users are trying to hide relationships - e.g. duplicates or collusion. Lexical analysis looks at streams of text, like press or orders, to flag those that have certain characteristics that do not necessarily need to be very well defned - e.g. lacking organic development, i.e. game development that too neatly exploits a nominal opponents' mistakes or press contrived to mask a predetermined alliance. At least those are the things that I would like to work on if I can get some help with user interface elements. Statistical methods depend on having enough players for statistics to become meaningful. I'm sure it will happen, but the issue is whether it takes 3 years or 10.&#60;/p&#62;
&#60;p&#62;Chris
&#60;/p&#62;</description>
		</item>

	</channel>
</rss>
