<?xml version="1.0" encoding="utf-8" standalone="no"?>
<!DOCTYPE document PUBLIC "-//CNX//DTD CNXML 0.5//EN" "http://cnx.rice.edu/technology/cnxml/schema/dtd/0.5/cnxml_plain.dtd">
<document xmlns="http://cnx.rice.edu/cnxml" xmlns:md="http://cnx.rice.edu/mdml/0.4" xmlns:bib="http://bibtexml.sf.net/" id="new">
  <name>Import from XML</name>
  <metadata>
  <md:version>1.1</md:version>
  <md:created>2007/08/29 23:33:26.969 GMT-5</md:created>
  <md:revised>2007/08/30 00:16:18.149 GMT-5</md:revised>
  <md:authorlist>
      <md:author id="jgrimes">
      <md:firstname>Joseph</md:firstname>
      <md:othername>E.</md:othername>
      <md:surname>Grimes</md:surname>
      <md:email>joe_grimes@sil.org</md:email>
    </md:author>
  </md:authorlist>

  <md:maintainerlist>
    <md:maintainer id="jgrimes">
      <md:firstname>Joseph</md:firstname>
      <md:othername>E.</md:othername>
      <md:surname>Grimes</md:surname>
      <md:email>joe_grimes@sil.org</md:email>
    </md:maintainer>
  </md:maintainerlist>
  
  <md:keywordlist>
    <md:keyword>analysis</md:keyword>
    <md:keyword>backup and restore</md:keyword>
    <md:keyword>collection</md:keyword>
    <md:keyword>transmit data</md:keyword>
    <md:keyword>Wordcorr</md:keyword>
    <md:keyword>XML</md:keyword>
  </md:keywordlist>

  <md:abstract>The main way to get data in and out of Wordcorr, besides typing it in directly, is via an XML file.</md:abstract>
</metadata>
  <content>
    <para id="delete_me">You've already seen that Wordcorr saves everything in the form of XML files. An XML file has a tightly designed structure that keeps metadata, data, and the results of analysis sorted out at all times. A Wordcorr backup file is an XML file. A file for sharing a Wordcorr connection with another scholar or a student is the same thing, an XML file that preserves the structure of everything in a Wordcorr collection. A file for archiving is the same XML file, because it can be adapted to new media and new concepts years into the future.</para><para id="element-176">So whenever you export an XML collection file, you have something that you or a colleague can import into the same computer to restore your work after a malfunction, or into another computer so that another pair of eyes can look at the same data and analysis.</para><para id="element-355">The file you get will not be labeled as an XML file. It will be automatically compressed and presented as a ZIP file. But what is compressed is the XML file. Whether you refer to it as JG-Mind5.XML or JG-Mind5.ZIP, Wordcorr will check to see if it is compressed, and if so, it will extract the XML file from the ZIP file. If it isn't compressed (though it usually will be), Wordcorr knows enough to not try to uncompress it.</para><note>The reason for compressing everything coming out of Wordcorr and uncompressing everything coming into Wordcorr is simple and humane. Most comparative linguists are occasionally out in the field, either without regular access to the Internet or at best with very slow access. That's where the interesting languages are. For them, the smaller their files are, the better.</note><para id="element-440">That makes it easy to back up your work at least once a day. Comparative phonology is complicated enough that you don't want to have to backtrack because of a slipup or a computer crash.</para><para id="element-326">Being able to import Wordcorr files also makes it easy for colleagues to work on large data sets as a team -- though you'll need caution if two people are working on the same part of the same view.</para><note>If there is a chance two people's work on the same collection might get snarled up, the first precaution to take is to have your information technology people set you up with a versioning repository like CVS or SVN. See them for details.</note><para id="element-673">XML is not the only way to take in data, and there are various ways to output data besides XML. The next section sketches some of these. But XML is the workhorse, safe and useful for all the main transport of information.</para>   
  </content>
  
</document>
