[Year 12 SofDev] Pseudocode standards

[Andrew Shortell]

Hi Mark

Before we get too carried away with producing a complex definitive set of
constructs I would like you (and everyone else too) to carefully read the
examiners¹ reports since 1991 about the question(s) involving writing pseudo
code or reading pseudo code and flowcharts and Nasty spiderman diagrams and
match them up to the questions themselves.

Then consider carefully the level at which these questions have had to be
pitched to elicit a reasonable spread of results. (an average of 50% of the
available marks is nice )

One needs to carefully consider the purpose of VCE

CATs/SACs/projects are very different to exams

My 5c worth (given that the face value is nearly less than the metal in it!)

Andrew
-- 
Andrew Shortell

mailto:shortell at get2me.net
Heidelberg Teaching Unit
Ph 9470 3403
Fax  9470 3215



On 18/05/11 11:16 AM, "Mark KELLY" <kel at mckinnonsc.vic.edu.au> wrote:

> Hi Claudia and all
> 
> Is there any news on what pseudocode we can expect to be "standard" in SD
> exams?  Until recently there wasn't even a FOR structure in pseudocode, but
> it's there now (Q4 in the sample questions) - a good move.
> What other structures and syntaxes does VCAA consider appropriate?
> 
> Just as the lack of complex structures in flowcharts result in primitive and
> bulky representations of design, a pseudocode lacking important programming
> primitives must hamstring the potential of pseudocode-based questions.
> I suggest we need some more powerful data and control structures in our
> pseudocode that better reflect the standards seen in all modern languages.
> The exact naming is neither here nor there as long as they're intelligible to
> everyone...
> 
> = for assignment.  Why do we persist with  <- when it's not used in any
> language and is so hard to typeset for printing??
>  
> IF-ELSEIF-ELSE-ENDIF (including AND/OR as logical operators)
> 
> The three basic loops...
> FOR-NEXT
> DO-WHILE
> LOOP-UNTIL (or REPEAT-UNTIL like last year's exam had)
> 
> I'd like pseudocode to be able to declare variables and arrays.  It's a vital
> programming concept...
> DECLARE var AS type
> DECLARE array(a,b) AS type
> 
> Add some decent I/O operators...
> INPUT (from keyboard) and READ (from a file)
> DISPLAY (onscreen), WRITE (to file) and PRINT (to paper)
> OPEN - CLOSE text file
> MESSAGEBOX  - hell, our language has to be GUI but there's no GUI support in
> the pseudocode. 
> 
> One major omission in current pseudocode is the concept of modularity: I've
> never seen a subprogram or function call. 
> We could really improve the power and reality of algorithms by allowing
> modularity with
> CALL subprogram(parameters)
> SUBPROGRAM(parameters)-END SUBPROGRAM
> FUNCTION(parameters)-END FUNCTION (with RETURN value)
> 
> And if the examiners want to introduce new functions all they have to do is
> explain them before using them. It's not too hard.
> 
> Have I left out any important programming concepts you think we could use in
> pseudocode?

-- 

-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://www.edulists.com.au/pipermail/sofdev/attachments/20110518/03ddd7cf/attachment.html