Critique on - On the Criteria To Be used in Decomposing systems into Modules

• Modulariztion -> design decisions ("system level")
• KWIC system ->
• Not a large system -> still used for modularization example -> questionable criteria.
• Four to a word -> less efficiency in data retrieval and storage -> today such small storage volume doesn't matter.
• Ambiguities -> Author mentions the use of CSSETUP -> initial setup -> later says it is not needed in the changeability section
• Modularization 1 -> limited or almost no implementation details
• Modularization 2 -> has functions , variable, conditions
• Modularization 1 -> "system comprehensible as whole" => basic modularization requirement of understanding modules in the system is not fulfilled.
• Indexing and usage of data structures in case of larger systems (involving n->very large number) will affect or not affect efficiency
• Existence of other modularization systems fitting into the same problem domain? Missing??
• Strongest point -> Using modularity to hide decisions.

==> Unfair/incomplete/ comparision