[version 0.5.1 12/18/2006]: [Abhaya]
- No stoplist is used by default now.
- Handle the case of reference system IDs and test systemIDs
  being same.
- Output can be genertaed in two formats, one a sgm file with
 the original segs annotated with score and corresponding refID
 and other a stripped down version with no sgml using plainOutput
 option. Details in README.txt.
- Assign a score of 1 when the hypothesis and reference
  are identical by explicitly checking number of matches and
  number of chunks.
- Added --keepPunctuation option to retain the punctuation during
 the normalization (to match the NIST normalization). Default is 
 to remove everything non alphanumeric from the data.

[version 0.5 10/20/2006]: [Abhaya]
- Internal changes to make wn_modules faster.
- Accept test files with nbest lists. For details, see README.txt
 
[version 0.4.3, 4/18/2005]: [Bano]
- Expanded the segment reading regular expressions in the readSGMfile function
to incorporate unexpected characters beween the tokens of the tag <seg id=1>.
This was in response to the problem where meteor.pl would refuse to read in
files with the tags <seg id="1">.

[version 0.4.2, 4/11/2005]: [Bano]
- Changed switch "-d <debug level>" to "-l <log file name>" so that user can
specify the name of the output sentence-by-sentence log file. 

[Version 0.4.1, 4/1/2005]: [Bano] 
- Fixed some bugs that got inserted with code clean up in version 0.4 regarding
reading in the sgm files. 

[Version 0.4, 3/27/3005]: [Bano]
- Did miscellaneous code clean-ups in main meteor.pl code to improve
maintainability. Cleaned up output.