<?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: New link type proposal, a:external {}</title>
	<atom:link href="http://breasy.com/blog/2007/08/07/new-link-type-proposal-aexternal/feed/" rel="self" type="application/rss+xml" />
	<link>http://breasy.com/blog/2007/08/07/new-link-type-proposal-aexternal/</link>
	<description>Musings on design and technology</description>
	<pubDate>Fri, 05 Sep 2008 23:52:56 +0000</pubDate>
	<generator>http://wordpress.org/?v=2.5</generator>
		<item>
		<title>By: Udi</title>
		<link>http://breasy.com/blog/2007/08/07/new-link-type-proposal-aexternal/#comment-6325</link>
		<dc:creator>Udi</dc:creator>
		<pubDate>Tue, 21 Aug 2007 03:25:58 +0000</pubDate>
		<guid isPermaLink="false">http://breasy.com/blog/2007/08/07/new-link-type-proposal-aexternal/#comment-6325</guid>
		<description>Hey Vin,

I was referring to the new startup that I'm currently developing with Dan.  Not this blog.  

Thanks,
Udi</description>
		<content:encoded><![CDATA[<p>Hey Vin,</p>
<p>I was referring to the new startup that I&#8217;m currently developing with Dan.  Not this blog.  </p>
<p>Thanks,<br />
Udi</p>
]]></content:encoded>
	</item>
	<item>
		<title>By: Vin Turk</title>
		<link>http://breasy.com/blog/2007/08/07/new-link-type-proposal-aexternal/#comment-6224</link>
		<dc:creator>Vin Turk</dc:creator>
		<pubDate>Wed, 15 Aug 2007 18:14:44 +0000</pubDate>
		<guid isPermaLink="false">http://breasy.com/blog/2007/08/07/new-link-type-proposal-aexternal/#comment-6224</guid>
		<description>There are code snippets (either in the form of a js that you include, or wordpress/blog plugins) that automatically detect a link not associated with your domain, and style that link differently (defined by you).  The main problem with these are exactly what you mentioned above: they are not implemented at the browser level...its up to the web master, CMS, or platform to identify and style the links differently.

There is another issue when relying on a webmaster: Most large publications have editors that submit content using a defined workflow such as:
1)editor/journalist writes article and uploads it into system
2)article is sent to another user for approval or more editing
3)article is categorized or taxonomy applied
4)article is published to public site.

Relying on an journalist to flag these types of links is a worthless endeavor.

Udi, you mention "The value here comes in picking a simple default to apply when webmasters neglect to do the work on their own. (I’ve already implemented this for our new site, as you know)".

Im looking at all the blog comments and most user's links are external, yet its the same styling for all the links.</description>
		<content:encoded><![CDATA[<p>There are code snippets (either in the form of a js that you include, or wordpress/blog plugins) that automatically detect a link not associated with your domain, and style that link differently (defined by you).  The main problem with these are exactly what you mentioned above: they are not implemented at the browser level&#8230;its up to the web master, CMS, or platform to identify and style the links differently.</p>
<p>There is another issue when relying on a webmaster: Most large publications have editors that submit content using a defined workflow such as:<br />
1)editor/journalist writes article and uploads it into system<br />
2)article is sent to another user for approval or more editing<br />
3)article is categorized or taxonomy applied<br />
4)article is published to public site.</p>
<p>Relying on an journalist to flag these types of links is a worthless endeavor.</p>
<p>Udi, you mention &#8220;The value here comes in picking a simple default to apply when webmasters neglect to do the work on their own. (I’ve already implemented this for our new site, as you know)&#8221;.</p>
<p>Im looking at all the blog comments and most user&#8217;s links are external, yet its the same styling for all the links.</p>
]]></content:encoded>
	</item>
	<item>
		<title>By: Jon Williams</title>
		<link>http://breasy.com/blog/2007/08/07/new-link-type-proposal-aexternal/#comment-6201</link>
		<dc:creator>Jon Williams</dc:creator>
		<pubDate>Wed, 15 Aug 2007 04:32:21 +0000</pubDate>
		<guid isPermaLink="false">http://breasy.com/blog/2007/08/07/new-link-type-proposal-aexternal/#comment-6201</guid>
		<description>CSS level 3 selectors can accomplish some of this functionality. I matched on links beginning with "mailto:" and the "after:" selector to type email links.</description>
		<content:encoded><![CDATA[<p>CSS level 3 selectors can accomplish some of this functionality. I matched on links beginning with &#8220;mailto:&#8221; and the &#8220;after:&#8221; selector to type email links.</p>
]]></content:encoded>
	</item>
	<item>
		<title>By: Udi</title>
		<link>http://breasy.com/blog/2007/08/07/new-link-type-proposal-aexternal/#comment-6087</link>
		<dc:creator>Udi</dc:creator>
		<pubDate>Tue, 07 Aug 2007 21:12:43 +0000</pubDate>
		<guid isPermaLink="false">http://breasy.com/blog/2007/08/07/new-link-type-proposal-aexternal/#comment-6087</guid>
		<description>The value here comes in picking a simple default to apply when webmasters neglect to do the work on their own.  (I've already implemented this for our new site, as you know)</description>
		<content:encoded><![CDATA[<p>The value here comes in picking a simple default to apply when webmasters neglect to do the work on their own.  (I&#8217;ve already implemented this for our new site, as you know)</p>
]]></content:encoded>
	</item>
	<item>
		<title>By: Udi</title>
		<link>http://breasy.com/blog/2007/08/07/new-link-type-proposal-aexternal/#comment-6086</link>
		<dc:creator>Udi</dc:creator>
		<pubDate>Tue, 07 Aug 2007 21:09:31 +0000</pubDate>
		<guid isPermaLink="false">http://breasy.com/blog/2007/08/07/new-link-type-proposal-aexternal/#comment-6086</guid>
		<description>No, keep it simple.  External means different top-level domain.  That should be the default and I think that will suffice for most cases where this would be useful.  

After that, you're welcome to not style a:external and do something more complex on your own with a custom style.</description>
		<content:encoded><![CDATA[<p>No, keep it simple.  External means different top-level domain.  That should be the default and I think that will suffice for most cases where this would be useful.  </p>
<p>After that, you&#8217;re welcome to not style a:external and do something more complex on your own with a custom style.</p>
]]></content:encoded>
	</item>
	<item>
		<title>By: Dan McKinley</title>
		<link>http://breasy.com/blog/2007/08/07/new-link-type-proposal-aexternal/#comment-6085</link>
		<dc:creator>Dan McKinley</dc:creator>
		<pubDate>Tue, 07 Aug 2007 21:02:58 +0000</pubDate>
		<guid isPermaLink="false">http://breasy.com/blog/2007/08/07/new-link-type-proposal-aexternal/#comment-6085</guid>
		<description>What about subdomains? Are those external? They are for some people but not for others. Same domain, but https:// instead of http://? Different ports? What about http://foo.com vs. http://www.foo.com?

I would like the idea but the implementation seems more than trivial. And if that's true the odds that results from all of the major browsers would match seem low. 

However you definitely do know which links are external when you're coding your application. With the concerns above it winds up being easier for you to code a control (or whatever this concept is called in your web framework of choice) that renders an anchor with class="external" on it.</description>
		<content:encoded><![CDATA[<p>What about subdomains? Are those external? They are for some people but not for others. Same domain, but <a href="https://" rel="nofollow">https://</a> instead of <a href="http://?" rel="nofollow">http://?</a> Different ports? What about <a href="http://foo.com" rel="nofollow">http://foo.com</a> vs. <a href="http://www.foo.com?" rel="nofollow">http://www.foo.com?</a></p>
<p>I would like the idea but the implementation seems more than trivial. And if that&#8217;s true the odds that results from all of the major browsers would match seem low. </p>
<p>However you definitely do know which links are external when you&#8217;re coding your application. With the concerns above it winds up being easier for you to code a control (or whatever this concept is called in your web framework of choice) that renders an anchor with class=&#8221;external&#8221; on it.</p>
]]></content:encoded>
	</item>
</channel>
</rss>
