<?xml version="1.0" encoding="UTF-8"?><!-- generator="wordpress/2.2" -->
<rss version="2.0" 
	xmlns:content="http://purl.org/rss/1.0/modules/content/">
<channel>
	<title>Comments on: Some syntactic considerations</title>
	<link>http://some-antics.com/blog/2007/08/27/some-syntactic-considerations/</link>
	<description>when i say "some", you think "not all"</description>
	<pubDate>Thu, 08 Jan 2009 11:29:39 +0000</pubDate>
	<generator>http://wordpress.org/?v=2.2</generator>

	<item>
		<title>By: simon charlow</title>
		<link>http://some-antics.com/blog/2007/08/27/some-syntactic-considerations/#comment-8125</link>
		<author>simon charlow</author>
		<pubDate>Wed, 29 Aug 2007 03:22:19 +0000</pubDate>
		<guid>http://some-antics.com/blog/2007/08/27/some-syntactic-considerations/#comment-8125</guid>
		<description>couple questions/points on your writeup:

-do you mean the &#124; slash differently than the superscript?

-something to be careful about is that an A^B should distribute just like an A. so having essentially free shifting from A/B to A^B isn't something you want. you'll want to restrict that to cases where A/B actually distributes just like A, i.e., VP ellipsis, since VP/VPs go the same places as VPs (modulo auxiliary stuff):

i [VP went to the store]
i [VP/VP -&#62; VP^VP didn't]</description>
		<content:encoded><![CDATA[<p>couple questions/points on your writeup:</p>
<p>-do you mean the | slash differently than the superscript?</p>
<p>-something to be careful about is that an A^B should distribute just like an A. so having essentially free shifting from A/B to A^B isn&#8217;t something you want. you&#8217;ll want to restrict that to cases where A/B actually distributes just like A, i.e., VP ellipsis, since VP/VPs go the same places as VPs (modulo auxiliary stuff):</p>
<p>i [VP went to the store]<br />
i [VP/VP -&gt; VP^VP didn&#8217;t]</p>
]]></content:encoded>
	</item>
	<item>
		<title>By: simon charlow</title>
		<link>http://some-antics.com/blog/2007/08/27/some-syntactic-considerations/#comment-8120</link>
		<author>simon charlow</author>
		<pubDate>Tue, 28 Aug 2007 21:09:08 +0000</pubDate>
		<guid>http://some-antics.com/blog/2007/08/27/some-syntactic-considerations/#comment-8120</guid>
		<description>yeah, it is left leaning, i just didnt specify (i dont think you can do subscripts in comments, and i didn't want to introduce a notation without context)</description>
		<content:encoded><![CDATA[<p>yeah, it is left leaning, i just didnt specify (i dont think you can do subscripts in comments, and i didn&#8217;t want to introduce a notation without context)</p>
]]></content:encoded>
	</item>
	<item>
		<title>By: emma</title>
		<link>http://some-antics.com/blog/2007/08/27/some-syntactic-considerations/#comment-8117</link>
		<author>emma</author>
		<pubDate>Tue, 28 Aug 2007 17:49:05 +0000</pubDate>
		<guid>http://some-antics.com/blog/2007/08/27/some-syntactic-considerations/#comment-8117</guid>
		<description>I think yes.  Polly's rule (see footnote in the L&#038;P VFS paper in the functional NPs section) just says that those slashes are variable; so presumably, since we're calling &#124; a slash, yes.

The slash on your VP should be a left-leaning, yes?  (S\NP)&#124;NP, if you're following the Steedman notation.</description>
		<content:encoded><![CDATA[<p>I think yes.  Polly&#8217;s rule (see footnote in the L&#038;P VFS paper in the functional NPs section) just says that those slashes are variable; so presumably, since we&#8217;re calling | a slash, yes.</p>
<p>The slash on your VP should be a left-leaning, yes?  (S\NP)|NP, if you&#8217;re following the Steedman notation.</p>
]]></content:encoded>
	</item>
	<item>
		<title>By: simon charlow</title>
		<link>http://some-antics.com/blog/2007/08/27/some-syntactic-considerations/#comment-8104</link>
		<author>simon charlow</author>
		<pubDate>Tue, 28 Aug 2007 01:48:33 +0000</pubDate>
		<guid>http://some-antics.com/blog/2007/08/27/some-syntactic-considerations/#comment-8104</guid>
		<description>can (A/B)/C include (A/B)&#124;C?

just curious because im wondering if you can shift a pro-form hosting VP (i.e., an (S/NP)&#124;NP) to an (S&#124;NP)/NP</description>
		<content:encoded><![CDATA[<p>can (A/B)/C include (A/B)|C?</p>
<p>just curious because im wondering if you can shift a pro-form hosting VP (i.e., an (S/NP)|NP) to an (S|NP)/NP</p>
]]></content:encoded>
	</item>
</channel>
</rss>
