<?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: SwiftRiver Web Services Launches</title>
	<atom:link href="http://blog.ushahidi.com/index.php/2010/05/12/swiftriver-web-services-launches/feed/" rel="self" type="application/rss+xml" />
	<link>http://blog.ushahidi.com/index.php/2010/05/12/swiftriver-web-services-launches/</link>
	<description>Thoughts and Lessons from an African Open-Source Project</description>
	<lastBuildDate>Tue, 07 Feb 2012 20:20:38 +0000</lastBuildDate>
	<sy:updatePeriod>hourly</sy:updatePeriod>
	<sy:updateFrequency>1</sy:updateFrequency>
	<generator>http://wordpress.org/?v=3.0.4</generator>
	<item>
		<title>By: Jon Gosier</title>
		<link>http://blog.ushahidi.com/index.php/2010/05/12/swiftriver-web-services-launches/comment-page-1/#comment-5224</link>
		<dc:creator>Jon Gosier</dc:creator>
		<pubDate>Sat, 29 May 2010 16:41:47 +0000</pubDate>
		<guid isPermaLink="false">http://blog.ushahidi.com/?p=2044#comment-5224</guid>
		<description>Jaro, SwiftRiver already filters out duplicate content.  This feature is also an open web service that anyone can use for their applications.  It&#039;s called SiCDS (Swift Content Duplication Service).  It looks for exact duplicates, re-tweets, reposts and more.

As far as marking an entire channel (SMS, Twitter, Email, News) as trusted, no.  However, you can automatically trust sources so as you add phone numbers into the system, you could just automatically trust those numbers.  It&#039;s what you&#039;re asking for but it takes a few more steps to do (trusting each number individually).</description>
		<content:encoded><![CDATA[<p>Jaro, SwiftRiver already filters out duplicate content.  This feature is also an open web service that anyone can use for their applications.  It&#8217;s called SiCDS (Swift Content Duplication Service).  It looks for exact duplicates, re-tweets, reposts and more.</p>
<p>As far as marking an entire channel (SMS, Twitter, Email, News) as trusted, no.  However, you can automatically trust sources so as you add phone numbers into the system, you could just automatically trust those numbers.  It&#8217;s what you&#8217;re asking for but it takes a few more steps to do (trusting each number individually).</p>
]]></content:encoded>
	</item>
	<item>
		<title>By: Jaro</title>
		<link>http://blog.ushahidi.com/index.php/2010/05/12/swiftriver-web-services-launches/comment-page-1/#comment-5218</link>
		<dc:creator>Jaro</dc:creator>
		<pubDate>Fri, 28 May 2010 23:06:08 +0000</pubDate>
		<guid isPermaLink="false">http://blog.ushahidi.com/?p=2044#comment-5218</guid>
		<description>Hi, thanks for the post and great to hear about these developments. I would like to ask two questions:

Once we started processing text messages coming from Haiti and consequently submitted reports based on these messages to haiti.ushahidi.com, we created an unique public source of information. before that we were combing any other available sources for such reports manually. And it wasn&#039;t unusual that some information was circulating the web (twitter, FB,...)for many days/weeks as people actually didn&#039;t know how actual this information is, but at the same time they felt it is important to share such information because it sounded urgent. Once we opened the stream of reports based on text messages, it was suddenly very &quot;fresh&quot; source of information on the web and wild retweeting started. People even started submitting this information back to ushahidi through web forms not realizing that ushahidi is actually the original source. It took some effort of our team to ensure that we are not approving these reports in order not to create duplications in the system. My point is, if we can automatize this process, is there a way how to distinguish the original report from it&#039;s retweets/copies? because if one information appears in many sources it doesn&#039;t necessarily mean that it increases its importance and/or reliability. But, if I understand it correctly, automatic combing will add importance to one information if it comes from variety of sources? But I might be wrong.

Second question: at one point we decided that information coming through text messages is the best and most reliable source of information at the moment (serious emergency). Is there a way that user of Swift River could temporarily tag SMS based sources as something like &quot;more reliable than others&quot; in order to give them higher verification rating?

Thanks

Jaro</description>
		<content:encoded><![CDATA[<p>Hi, thanks for the post and great to hear about these developments. I would like to ask two questions:</p>
<p>Once we started processing text messages coming from Haiti and consequently submitted reports based on these messages to haiti.ushahidi.com, we created an unique public source of information. before that we were combing any other available sources for such reports manually. And it wasn&#8217;t unusual that some information was circulating the web (twitter, FB,&#8230;)for many days/weeks as people actually didn&#8217;t know how actual this information is, but at the same time they felt it is important to share such information because it sounded urgent. Once we opened the stream of reports based on text messages, it was suddenly very &#8220;fresh&#8221; source of information on the web and wild retweeting started. People even started submitting this information back to ushahidi through web forms not realizing that ushahidi is actually the original source. It took some effort of our team to ensure that we are not approving these reports in order not to create duplications in the system. My point is, if we can automatize this process, is there a way how to distinguish the original report from it&#8217;s retweets/copies? because if one information appears in many sources it doesn&#8217;t necessarily mean that it increases its importance and/or reliability. But, if I understand it correctly, automatic combing will add importance to one information if it comes from variety of sources? But I might be wrong.</p>
<p>Second question: at one point we decided that information coming through text messages is the best and most reliable source of information at the moment (serious emergency). Is there a way that user of Swift River could temporarily tag SMS based sources as something like &#8220;more reliable than others&#8221; in order to give them higher verification rating?</p>
<p>Thanks</p>
<p>Jaro</p>
]]></content:encoded>
	</item>
	<item>
		<title>By: Tweets that mention SwiftRiver Web Services Launches – The Ushahidi Blog -- Topsy.com</title>
		<link>http://blog.ushahidi.com/index.php/2010/05/12/swiftriver-web-services-launches/comment-page-1/#comment-4817</link>
		<dc:creator>Tweets that mention SwiftRiver Web Services Launches – The Ushahidi Blog -- Topsy.com</dc:creator>
		<pubDate>Wed, 12 May 2010 19:25:24 +0000</pubDate>
		<guid isPermaLink="false">http://blog.ushahidi.com/?p=2044#comment-4817</guid>
		<description>[...] This post was mentioned on Twitter by SwiftRiver. SwiftRiver said: SwiftRiver Web Services Launches: The SwiftRiver Web Services platform offers RESTful apps that live in the cloud ... http://bit.ly/bokeL3 [...]</description>
		<content:encoded><![CDATA[<p>[...] This post was mentioned on Twitter by SwiftRiver. SwiftRiver said: SwiftRiver Web Services Launches: The SwiftRiver Web Services platform offers RESTful apps that live in the cloud &#8230; <a href="http://bit.ly/bokeL3" rel="nofollow">http://bit.ly/bokeL3</a> [...]</p>
]]></content:encoded>
	</item>
</channel>
</rss>

