The attachments are the final versions of mKR contexts for Sjir's example.
Composers.mkr = known facts about Composers
Famous.Composers.mkr = rules for deciding who is famous
Find.Famous.Composers.mkr = mKR script which applies rules & prints results
Famous.mkr = results hierarchy

The rules are greatly simplified using the Current Concept "."
(see Tutorial II.15) with its characteristics stored in $variables.
Here are the exact Linux shell commands used to produce the results
ke -s Find.Famous.Composers.mkr
deletecomment *.out >Famous.mkr
cat Famous.mkr

You can see that none of the Composers were found to be
Famous Composers, due to lack of detailed information about
the community. The classification results are obtained using
the concept formation verbs isi, isd (integrate, differentiate)
described in Tutorial III.19. For example:

rhm@lenovo8 /home/mactest$ ke -s
ke$ do read from Famous.mkr done;
ke$ Wolfgang Amadeus Mozart isd Famous Composer;
ke$ Composer isc* ?;
Composer;
/ Famous Composer;
// Wolfgang Amadeus Mozart;
/ Guisseppe Verdi;
/ Gustav Mahler;
ke$ exit;
rhm@lenovo8 /home/mactest$

Dick McCullough
Context Knowledge Systems
mKE and the mKR language
mKR/mKE tutorial