"Base programs"

Grist for the mill of "types of program" here on HNN, ever the most interesting forum out there.

The list:

basic form
form with multi-valued lines
form with data set (multi-valued parent/children)
form with grid
tabbed form
form with skins
batch (loop thru something)
batch update selected records in a table
batch dump table records --> .txt or .xml file
batch .txt or .xml file --> update table records
batch file --> create & populate database table
cut an email
traverse internal tree function
traverse file index(es)
build html from parameters
build javascript from parameters
build .pdf from parameters
build hp esc sequences from parameters
benchmark a process
how does that syntax work?
which way is faster?
batch string parser
batch source code search
batch source code changer
batch parse source code, identify routine for reuse

A follow-up comment mentions that these should be in a library - but I beg to differ. This is a higher level of organization than the library.  This is where we talk about the semantic purpose of the program.  It fits well with the Unix philosophy of many small utilities, actually.

Food for thought!