1. CD to directory which contains the compiled corpus files.
    cd `corpinfo -p mycorpus`
  1. Create a file with a list of the names of the sketch grammar relations (each name on a separate line).
    tr '\0' '\n' < `corpinfo -g WSBASE mycorpus`.lex > /tmp/mycorpus_sgrel_names
  1. Edit the names of relations in the /tmp/mycorpus_sgrel_names file. Do not change the order of lines.
  1. Backup the current names of sketch grammar relations.
    mkdir ws-backup
    mv `corpinfo -g WSBASE mycorpus`.lex* ws-backup
  1. Compile the new names.
    WSBASE=`corpinfo -g WSBASE mycorpus`
    encodevert -p /tmp/ -a $WSBASE_BASENAME - < /tmp/mycorpus_sgrel_names
    mv /tmp/$WSBASE_BASENAME.lex* .