<?xml version="1.0" encoding="UTF-8"?><!-- generator="wordpress/2.3" -->
<rss version="2.0"
	xmlns:content="http://purl.org/rss/1.0/modules/content/"
	xmlns:dc="http://purl.org/dc/elements/1.1/"
	>
<channel>
	<title>Comments on: How to fix multi-tap</title>
	<link>http://blog.javia.org/how-to-fix-multi-tap/</link>
	<description>Mihai Preda on mobile applications</description>
	<pubDate>Fri, 21 Nov 2008 21:33:23 +0000</pubDate>
	<generator>http://wordpress.org/?v=2.3</generator>
		<item>
		<title>By: mihai</title>
		<link>http://blog.javia.org/how-to-fix-multi-tap/#comment-12497</link>
		<dc:creator>mihai</dc:creator>
		<pubDate>Thu, 20 Sep 2007 13:35:41 +0000</pubDate>
		<guid>http://blog.javia.org/how-to-fix-multi-tap/#comment-12497</guid>
		<description>Ok, I think I understand. The goal of the change is to get "I" and "U" (vocals) on different keys, and to get "I" and "Y" (both 'accentable') on different keys? I'll think of it, and make some bigram cost calculation to see what would be the impact on English.

One problem is that, while I found single-letter frequency for a few languages (French, Spanish, German, Polish, Romanian), I have bigram-frequency only for English, and thus the second letter was optimized for English only.

But one important thing is that I don't want to put 'accented' characters after the normal ones on the same key (as it is now in multi-tap), because this slows down non-accented typing considerably. I would rather put 'modifiers' (such as: cedilla, grave, acute, etc) on the "*" key, and a special letter is typed by typing the normal letter, followed by the modifier.</description>
		<content:encoded><![CDATA[<p>Ok, I think I understand. The goal of the change is to get &#8220;I&#8221; and &#8220;U&#8221; (vocals) on different keys, and to get &#8220;I&#8221; and &#8220;Y&#8221; (both &#8216;accentable&#8217;) on different keys? I&#8217;ll think of it, and make some bigram cost calculation to see what would be the impact on English.</p>
<p>One problem is that, while I found single-letter frequency for a few languages (French, Spanish, German, Polish, Romanian), I have bigram-frequency only for English, and thus the second letter was optimized for English only.</p>
<p>But one important thing is that I don&#8217;t want to put &#8216;accented&#8217; characters after the normal ones on the same key (as it is now in multi-tap), because this slows down non-accented typing considerably. I would rather put &#8216;modifiers&#8217; (such as: cedilla, grave, acute, etc) on the &#8220;*&#8221; key, and a special letter is typed by typing the normal letter, followed by the modifier.</p>
]]></content:encoded>
	</item>
	<item>
		<title>By: Martin</title>
		<link>http://blog.javia.org/how-to-fix-multi-tap/#comment-12491</link>
		<dc:creator>Martin</dc:creator>
		<pubDate>Mon, 17 Sep 2007 19:35:38 +0000</pubDate>
		<guid>http://blog.javia.org/how-to-fix-multi-tap/#comment-12491</guid>
		<description>I like the proposal, but would change "U" with "M" and "Y" with "X". Rational: In some (south and middle) European languages, accented variants of letters exist, e.g. vocals with accent (grave, acute, circumflex, dieresis), a and o with even more variants like "ae" and "o with /", c with cedilla, n with tilde, sharp s, y with accent grave or dieresis. To avoid "over-assigned" keys, just don't put two vocals or one of c, n, s, y on the same key. For slavic languages things are presumably much more complicated, but at least three of the six official UNO languages (english, french, spanish) would benefit.</description>
		<content:encoded><![CDATA[<p>I like the proposal, but would change &#8220;U&#8221; with &#8220;M&#8221; and &#8220;Y&#8221; with &#8220;X&#8221;. Rational: In some (south and middle) European languages, accented variants of letters exist, e.g. vocals with accent (grave, acute, circumflex, dieresis), a and o with even more variants like &#8220;ae&#8221; and &#8220;o with /&#8221;, c with cedilla, n with tilde, sharp s, y with accent grave or dieresis. To avoid &#8220;over-assigned&#8221; keys, just don&#8217;t put two vocals or one of c, n, s, y on the same key. For slavic languages things are presumably much more complicated, but at least three of the six official UNO languages (english, french, spanish) would benefit.</p>
]]></content:encoded>
	</item>
</channel>
</rss>
