<?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: How long is a (piece of) string?</title>
	<atom:link href="http://chrisscott.org/software/cms/how-long-is-a-piece-of-string/feed" rel="self" type="application/rss+xml" />
	<link>http://chrisscott.org/software/cms/how-long-is-a-piece-of-string</link>
	<description>On the Semantic Web, Web development and tech in general</description>
	<lastBuildDate>Tue, 19 Jul 2011 11:03:32 +0000</lastBuildDate>
	<sy:updatePeriod>hourly</sy:updatePeriod>
	<sy:updateFrequency>1</sy:updateFrequency>
	<generator>http://wordpress.org/?v=3.3.1</generator>
	<item>
		<title>By: Christopher Hill</title>
		<link>http://chrisscott.org/software/cms/how-long-is-a-piece-of-string/comment-page-1#comment-12</link>
		<dc:creator>Christopher Hill</dc:creator>
		<pubDate>Mon, 27 Apr 2009 16:31:25 +0000</pubDate>
		<guid isPermaLink="false">http://chrisscott.org/?p=13#comment-12</guid>
		<description>Very interesting set of posts on Twitterbots. Publishers right now are at a critical stage with Twitter and because of the short form of the medium aren&#039;t putting enough effort into their Tweet strategy. 

&quot;Quality is what attracts an audience and certainly what keeps them… even on Twitter.&quot;

Right on! And if your content on Twitter sucks it will ultimately detract from what you are trying to do. You&#039;ve shown that automation can be used to product decent robot Tweets that actually meet a consumer need.</description>
		<content:encoded><![CDATA[<p>Very interesting set of posts on Twitterbots. Publishers right now are at a critical stage with Twitter and because of the short form of the medium aren&#8217;t putting enough effort into their Tweet strategy. </p>
<p>&#8220;Quality is what attracts an audience and certainly what keeps them… even on Twitter.&#8221;</p>
<p>Right on! And if your content on Twitter sucks it will ultimately detract from what you are trying to do. You&#8217;ve shown that automation can be used to product decent robot Tweets that actually meet a consumer need.</p>
]]></content:encoded>
	</item>
	<item>
		<title>By: Sarah Bourne</title>
		<link>http://chrisscott.org/software/cms/how-long-is-a-piece-of-string/comment-page-1#comment-11</link>
		<dc:creator>Sarah Bourne</dc:creator>
		<pubDate>Fri, 24 Apr 2009 18:54:21 +0000</pubDate>
		<guid isPermaLink="false">http://chrisscott.org/?p=13#comment-11</guid>
		<description>Well, that&#039;s impressive!

I like your idea of kicking it back to the human, but for a slightly different reason. Far too much of our content is written far above the 8th-grade reading level recommended for web publishing. If the bot can&#039;t fit it into 140 characters, maybe it&#039;s because you used too many big words.

Of course, we&#039;re stuck with &quot;Massachusetts&quot;, but in similar situations, we shorten it to &quot;Mass.&quot; or the postal code &quot;MA&quot;. It might be good to let users set abbreviations for certain words or phrases.</description>
		<content:encoded><![CDATA[<p>Well, that&#8217;s impressive!</p>
<p>I like your idea of kicking it back to the human, but for a slightly different reason. Far too much of our content is written far above the 8th-grade reading level recommended for web publishing. If the bot can&#8217;t fit it into 140 characters, maybe it&#8217;s because you used too many big words.</p>
<p>Of course, we&#8217;re stuck with &#8220;Massachusetts&#8221;, but in similar situations, we shorten it to &#8220;Mass.&#8221; or the postal code &#8220;MA&#8221;. It might be good to let users set abbreviations for certain words or phrases.</p>
]]></content:encoded>
	</item>
	<item>
		<title>By: admin</title>
		<link>http://chrisscott.org/software/cms/how-long-is-a-piece-of-string/comment-page-1#comment-10</link>
		<dc:creator>admin</dc:creator>
		<pubDate>Fri, 24 Apr 2009 18:34:12 +0000</pubDate>
		<guid isPermaLink="false">http://chrisscott.org/?p=13#comment-10</guid>
		<description>Great comments &lt;a href=&quot;http://twitter.com/olegr&quot; rel=&quot;nofollow&quot;&gt;@OlegR&lt;/a&gt;. I particularly like to idea of a dictionary of SMS word truncations. No doubt there are many other things you could do to squeeze extra characters out once you have a decent semantic understanding of the content at the core.</description>
		<content:encoded><![CDATA[<p>Great comments <a href="http://twitter.com/olegr" rel="nofollow">@OlegR</a>. I particularly like to idea of a dictionary of SMS word truncations. No doubt there are many other things you could do to squeeze extra characters out once you have a decent semantic understanding of the content at the core.</p>
]]></content:encoded>
	</item>
	<item>
		<title>By: OlegR</title>
		<link>http://chrisscott.org/software/cms/how-long-is-a-piece-of-string/comment-page-1#comment-9</link>
		<dc:creator>OlegR</dc:creator>
		<pubDate>Fri, 24 Apr 2009 18:28:23 +0000</pubDate>
		<guid isPermaLink="false">http://chrisscott.org/?p=13#comment-9</guid>
		<description>here are some suggestions for the bot that I learned from my personal experience:
1. Twitter (or clients) do not like signs like &amp; / \ &#124; &quot; etc. - these come in often as yada yada and auto truncate meaningful part of the tweet with this kind of rubbish (for ex. #34w%)

2. You might want to abbreviate states (#MA instead of #massachusetts, or #QC instead of #Quebec). I know that Nstein&#039;s text mining engine comes with a geographic location cartridge that has all these abbreviations built in.

3. When you hash tag along name, pleause use underscore as space between words. Smashing all the words together becomes very geek-compatible, but not human-being-compatible if you know what I mean :-)

4. I&#039;ve ran across a dictionary of sms-word truncations - you might want to integrate those to save some twitter real estate. 

Cheers-

Oleg (twitter - OlegR)</description>
		<content:encoded><![CDATA[<p>here are some suggestions for the bot that I learned from my personal experience:<br />
1. Twitter (or clients) do not like signs like &amp; / \ | &#8221; etc. &#8211; these come in often as yada yada and auto truncate meaningful part of the tweet with this kind of rubbish (for ex. #34w%)</p>
<p>2. You might want to abbreviate states (#MA instead of #massachusetts, or #QC instead of #Quebec). I know that Nstein&#8217;s text mining engine comes with a geographic location cartridge that has all these abbreviations built in.</p>
<p>3. When you hash tag along name, pleause use underscore as space between words. Smashing all the words together becomes very geek-compatible, but not human-being-compatible if you know what I mean :-)</p>
<p>4. I&#8217;ve ran across a dictionary of sms-word truncations &#8211; you might want to integrate those to save some twitter real estate. </p>
<p>Cheers-</p>
<p>Oleg (twitter &#8211; OlegR)</p>
]]></content:encoded>
	</item>
</channel>
</rss>

