Comparative data
prepared on another computer can usually be imported into Wordcorr.
So far we have means of bringing data in from
- other Wordcorr users, or your
own Wordcorr backup files
- spreadsheets
- WordSurv and PalmSurv
Import from Wordcorr: When
a colleague or teacher emails you one of their Wordcorr export
files, if the name doesn't conflict with any of your file names,
put it in the same folder where you keep your own export files. If
the name conflicts, store it temporarily somewhere else, rename it
by prefixing the originator's initials or Creator ID and a hyphen,
then move to your working folder. Use File | Import XML. It doesn't
matter if the file is compressed (.zip) or not (.xml).
Import from spreadsheet: On
the Files page of Wordcorr's SourceForge
section there is a group of files called Utilities. Download the
Spreadsheet Utility developed by Maria Faehndrich. This
uses the Python scripting language, so you might want to bring
someone who knows Python into the picture to help. It has to be
done just once.
You transfer your data to
Wordcorr's spreadsheet template, following the pattern in the
example. You fill in any metadata that are missing. Then you save
the results as a Unicode UTF-8 tab-delimited CSV text file, and
give that file to the Python script. What comes out is an XML file
that can be imported into Wordcorr exactly as if it were a Wordcorr
export file.
Older spreadsheet programs don't
handle Unicode. Another Python utility is set up to do this, but
you will probably have to go over it with your Python-speaking
friends to get the details right for your data.
Import from WordSurv, original
format: The early versions of WordSurv put several kinds of
information on a single line, distinguished only by their position
in the line. Import WordSurv on the File menu handles this format
directly.
Import from WordSurv, revised
format, and from PalmSurv: Later versions of WordSurv put each
kind of information on a separate line, preceded by a tag like
"\i ". PalmSurv
developed a variant of WordSurv that handles IPA and does it on a
Palm pocket computer for use on field trips. They stepped into the
gap with the PalmSurv Converter, which changes both WordSurv 2.5
files (the revised format) and PalmSurv XML files into Wordcorr
XML.
Other: If you have
comparative data in some computer format we don't know about, there
are two strategies you can take to get them into Wordcorr:
- Morph your data into a
spreadsheet, then use Spreadsheet Utility,
- Get your Python-speaking
friends to adapt the logic of the Python script to your data.
If it works, let's make
arrangement to share it with the rest of the Wordcorr community. |
When you share a data collection with another linguist via
the Internet, or when you send a collection to your linguistics
students for them to analyze, you export the collection and they
import it. The collection goes out with its creator's ID prefixed
to the collection name, like JG-SulSel12.
They will build up their own views
to make precise how they analyze the data in the collection. So
when they send their work back to you for dialogue (or evaluation),
it still carries the same creator's ID, but it contains more than
you sent out, so it's an expanded version.
You may want to ask people who
work with your collection to prefix their ID to your ID, like
MF-JG-SulSel12, MF's reworking of JG's collection. Then when they
send back to you what they've done, there will be no confusion. |
