Re: [xsl] dubious certification questions

[tcn at melvaig dot co dot uk (Trevor Nash)]

>It was fun to take the test, but just to demonstrate how worthless
>certifications can be,

Yep.  I took MCSD (because my employer at the time paid me to do so).
IMO not worth the paper its written on.

> here are 2 questions from that heavily MSXML and
>Xalan oriented Brainbench XSL exam. I'm posting them here for reasons that
>should be obvious when you read them, if you know the correct answers. Oh,
>they say there's one -- and only one -- correct answer per question.
>
This was my gut reaction to the MCSD stuff - what a lot of rhubarb.
But then I sat back and thought about it a bit, and begin to see some
subtle techniques, of which you have provided examples.

>=========================================================
>
>  The Saxon XSLT Processor takes what kind of input?  
>    Choice 1     URI  
>    Choice 2     Stream of SAX events  
>    Choice 3     Stream of bytecode  
>    Choice 4     Single DOM object  
>    Choice 5     Text stream  
>
This is called diguising the question.
OK so you *can* pass a DOM or whatever to Saxon if you want.  What the
question os obviously looking for is, do you know that Saxon was
basically written as a SAX filter, and therefore that you should use
SAX if you get the choice?  So not all the answers offered are wrong -
but if you have to pick one, its clear which one it is.

'stream of bytecode' is there to weed out anyone answering at random,
in particular some sort of robot trying to work out what the right
answers are.  

>
>=========================================================
>
>  <xsl:template match="text()"> 
>    <xsl:value-of select="."/> 
>  </xsl:template>
>  
>  Referring to the sample code above, what is the result
>  of applying the XSL fragment?  
>
>    Choice 1    It copies all parent nodes to output that
>                 has text nodes as children.  
>
>    Choice 2    It copies all child nodes resident under
>                 any text node.  
>
>    Choice 3    It copies all descendent to the current
>                 node text nodes to the output.  
>
>    Choice 4    It changes all nodes to text nodes from
>                 the current location on the document tree.  
>
>    Choice 5    It matches all nodes and copies them to output.  
>
>
This is a combination of leaving out half the question, and phrasing
the answers in unusual terms.  If you make reasonable assumptions
about the context, and then remove all answers which have something
wrong with them (e.g. choice 1 is about parents, but the code has '.'
not '..') you are left with just one.  While it isn't a *good*
description of the result, its not actually wrong.

Of course these tricks are indistinguishable from sloppy question
setting.

>Have fun picking the right answer.
>
>   - Mike
>_____________________________________________________________________________
>mike j. brown, software engineer at  |  xml/xslt: http://skew.org/xml/
>webb.net in denver, colorado, USA    |  personal: http://hyperreal.org/~mike/
>
> XSL-List info and archive:  http://www.mulberrytech.com/xsl/xsl-list

Trevor Nash

 XSL-List info and archive:  http://www.mulberrytech.com/xsl/xsl-list