<?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: Unicode in URIs makes my head hurt</title>
	<atom:link href="http://therning.org/magnus/archives/297/feed" rel="self" type="application/rss+xml" />
	<link>http://therning.org/magnus/archives/297</link>
	<description>Incoherent mumblings</description>
	<lastBuildDate>Wed, 08 Aug 2012 00:21:01 +0000</lastBuildDate>
	<sy:updatePeriod>hourly</sy:updatePeriod>
	<sy:updateFrequency>1</sy:updateFrequency>
	<generator>http://wordpress.org/?v=3.5.1</generator>
	<item>
		<title>By: Dan Knapp</title>
		<link>http://therning.org/magnus/archives/297/comment-page-1#comment-67584</link>
		<dc:creator>Dan Knapp</dc:creator>
		<pubDate>Thu, 21 Jun 2007 11:10:05 +0000</pubDate>
		<guid isPermaLink="false">http://therning.org/magnus/archives/297#comment-67584</guid>
		<description><![CDATA[&lt;p&gt;The halfwidth/fullwidth thing has to do with the history of Japanese encodings.  See, on old-style text terminals, Japanese characters occupied two adjacent cells.  If you tried to intermingle English text with that, it would look awkward; so they have the extra-wide versions of the Latin characters.  Of course some of these look pretty badly distorted to the western eye, but at least they don&#039;t leave awkward gaps in the text.  I&#039;m not sure why the basic Japanese characters have halfwidth forms, though; possibly to solve the inverse problem?&lt;/p&gt;

&lt;p&gt;I guess it&#039;s not totally accurate to say that this is only a historical thing, because Japanese input methods still provide a way to type the fullwidth forms, and people still use them.  Most users are aware of the difference and know, for example, that you can&#039;t type the fullwidth numerals into a form field that&#039;s going to try to parse it as an integer.&lt;/p&gt;

&lt;p&gt;If you&#039;re setting your text vertically, you would rarely want the regular (halfwidth) Latin characters, because it would mess up what would otherwise be a very regular grid; also they are probably in a different font and there probably aren&#039;t metrics included for laying them out that way.  But if it were trailing off at the end of the line, it might look okay...  So at least to some people, there is a semantic difference between the halfwidth and fullwidth forms and it&#039;s wrong to automatically map between them.&lt;/p&gt;
]]></description>
		<content:encoded><![CDATA[<p>The halfwidth/fullwidth thing has to do with the history of Japanese encodings.  See, on old-style text terminals, Japanese characters occupied two adjacent cells.  If you tried to intermingle English text with that, it would look awkward; so they have the extra-wide versions of the Latin characters.  Of course some of these look pretty badly distorted to the western eye, but at least they don&#8217;t leave awkward gaps in the text.  I&#8217;m not sure why the basic Japanese characters have halfwidth forms, though; possibly to solve the inverse problem?</p>

<p>I guess it&#8217;s not totally accurate to say that this is only a historical thing, because Japanese input methods still provide a way to type the fullwidth forms, and people still use them.  Most users are aware of the difference and know, for example, that you can&#8217;t type the fullwidth numerals into a form field that&#8217;s going to try to parse it as an integer.</p>

<p>If you&#8217;re setting your text vertically, you would rarely want the regular (halfwidth) Latin characters, because it would mess up what would otherwise be a very regular grid; also they are probably in a different font and there probably aren&#8217;t metrics included for laying them out that way.  But if it were trailing off at the end of the line, it might look okay&#8230;  So at least to some people, there is a semantic difference between the halfwidth and fullwidth forms and it&#8217;s wrong to automatically map between them.</p>
]]></content:encoded>
	</item>
</channel>
</rss>
