<?xml version="1.0" encoding="UTF-8"?><!-- generator="wordpress/2.3.1" -->
<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 for Wine Marketing + Law Canada</title>
	<link>http://winemarketing.ca/blog</link>
	<description>A blog on wine law and marketing in Canada emphasizing British Columbia</description>
	<pubDate>Wed, 08 Sep 2010 13:42:57 +0000</pubDate>
	<generator>http://wordpress.org/?v=2.3.1</generator>
		<item>
		<title>Comment on The Future of Wine Buying and Research? by pbrosen</title>
		<link>http://winemarketing.ca/blog/?p=39#comment-46</link>
		<dc:creator>pbrosen</dc:creator>
		<pubDate>Mon, 05 Jan 2009 17:05:38 +0000</pubDate>
		<guid>http://winemarketing.ca/blog/?p=39#comment-46</guid>
		<description>Mark - 

Thanks for the "shout-out"! We're really excited about Drync (makes sense - we wrote it).

We've looked closely at the scanning barcodes thing. Turns out to be very difficult for two reasons:

1. there is no centralized wine bar code dbase or standards body, and wineries re-use barcodes. 

2. given low-light conditions and bottle curvature, the actual "recognition" piece is extremely difficult, making the likelihood of getting the code right low.

However, there is hope for label recognition (also a super hard technical problem), and possibly "mechanical turk" or other human based recognition scenarios. We are looking at all, because of course it makes perfect sense to be able to snap a photo and instantly get back reviews, prices, pairings etc.

Again, thanks for the mention.

Best,
Brad</description>
		<content:encoded><![CDATA[<p>Mark - </p>
<p>Thanks for the &#8220;shout-out&#8221;! We&#8217;re really excited about Drync (makes sense - we wrote it).</p>
<p>We&#8217;ve looked closely at the scanning barcodes thing. Turns out to be very difficult for two reasons:</p>
<p>1. there is no centralized wine bar code dbase or standards body, and wineries re-use barcodes. </p>
<p>2. given low-light conditions and bottle curvature, the actual &#8220;recognition&#8221; piece is extremely difficult, making the likelihood of getting the code right low.</p>
<p>However, there is hope for label recognition (also a super hard technical problem), and possibly &#8220;mechanical turk&#8221; or other human based recognition scenarios. We are looking at all, because of course it makes perfect sense to be able to snap a photo and instantly get back reviews, prices, pairings etc.</p>
<p>Again, thanks for the mention.</p>
<p>Best,<br />
Brad</p>
]]></content:encoded>
	</item>
</channel>
</rss>
