This question was kicked about in December. A student of mine stumbled on
a syntax highlighting related "conceptual bug" today. Thought I'd
share/revisit a dead thread.

On that thread, Alan suggested we use all our spare time to write a
reasonable program editor, Frank called for the SDI of the text editor
world and asked what we could do for Emacs (not what it could do for us),
some colourful images were kicked around, and Thomas asked in

http://www.mail-archive.com/[EMAIL PROTECTED]/msg00641.html

<QUOTE>
So, a question for all: if Alan gets something going, what structures
would be highlighted in a programming editor? How would they vary with the
type of language?
</QUOTE>

This came back to me because of a question an adult (non-traditional, as
they are sometimes referred to) student in a course I'm teaching
asked regarding syntax highlighting in BlueJ. His question highlights the
inconsistency or, perhaps, conceptual opacity of snytax highlighting as it
currently is (typically) implemented:

<QUESTION>
As you know when we enter say  public void or int the void or int appear
red on the screen. string refuses to be any colour except black it`s the
same symptom as the backspace issue but thats not the cause this time.
</QUESTION>

The context of the student's question is the header for a method in a
class, eg.

public void setFoo(int x) {
 ...
}

where BlueJ colors the 'void' and 'int' in red. However,

public String setFoo(int x) {
 ...
}

will result in 'String' being black. I'm guessing from an expert
perspective that BlueJ is highlighting Java's primitive types in *red*,
and anything that is a *class* is not. My student seems to have expected
things to be colored based on (perhaps)  their position, or perhaps their
*correctness*---by getting the type "String"  correct, BlueJ should
acknowledge the existance of this datatype, and therefore color it.

I'll ask him when I see him on Friday what he was expecting, so I can
compose a "bug report" to the BlueJ developers. However, it's a conceptual
bug from the perspective of a novice, and it's with respect to something
that we (apparently) have no real research on---the effect of syntax
highlighting on the comprehension of code. While this may not matter
significantly to experts, right here we have an example of a novice who is
confused by the (apparent) inconsistency of the cue.

Apologies for ressurecting a dead thread. I guess I'm just wonndering at
you how I might argue in support of my "bug report."

M

________________________________________________________________________
Matt Jadud                             http://www.cs-ed.org/blogs/mjadud/
Canterbury Weather: High 48 F / 9 C, Low 44 F / 7 C, Low-level cloud
 
----------------------------------------------------------------------
PPIG Discuss List ([EMAIL PROTECTED])
Discuss admin: http://limitlessmail.net/mailman/listinfo/discuss
Announce admin: http://limitlessmail.net/mailman/listinfo/announce
PPIG Discuss archive: http://www.mail-archive.com/discuss%40ppig.org/

Reply via email to