Weird Looking: In Rome's proud steps...

Last weekend, I drove over to Uvalde to help out with the JCL Area B convention.  It was a lot of fun.

I love JCL.  I love the JCL song.  I love the Latin classroom.  I love mythology, and history, and showing up for a convention at 7:30am having gotten no sleep.

Test grading is terrible, though.  The workflow is something like:  scantron -> access database -> saved to fixed-width record flat files -> data cleaned up with DOS “edit” -> (magical grading and reporting) -> BLUE RIBBON!

The flat file editing was terrible.  The records look something like:
[last name]      [first name]    [student id][answer-1][answer-2][answer-3]…

The answers are A, B, C, D, or a “?” if the scantron machine wasn’t able to decide which answer the student marked.  Then they sit about 20 people down at computers with DOS screens.  These people go through the files until they find a “?”, get its column number, look at a hard copy lookup table to find out what question number is associated with that column, then find it in the test and see if they can figure out what the student meant to mark.

It’s incredibly fair to students who are too stupid to correctly bubble in scantron forms.

And you’re supposed to always work with edit in insert mode, or the “you know, where the thingy is a white square, not a little, like, (sideways hand motion)” mode, as they called it.  Please note that edit has no undo function, which, when coupled with insert mode and my habitual attempts to use vi commands, meant I pretty much wiped out vast swaths of unrecoverable data every time I got near a keyboard.

If grading works the same way at state, I’m going to be all PERL!  I actually thought very seriously about banging out a little perl this time, but the grading organizers didn’t even like me pulling out my laptop to edit with vi.

Sometimes I think it’d be good to keep in mind that Rome fell.

Name

Email

Link

Comment

Accepts BBCode with a few enhancements.