Re: [school-discuss] Static vs dynamic EduML

[Dominique Broeglin <broeglin@essi.fr>]

On Mon, 2002-02-11 at 18:16, Bruno Vernier wrote:
> On Mon, Feb 11, 2002 at 09:53:57AM +0100, Dominique Broeglin wrote:
> 
> > 	Maybe the XML namespaces could be used ? The way the openoffice format
> > was build is very intersting (http://www.openoffice.org). They used a
> > lot of the existing W3C schemas and assembled them in different
> > namespaces (SVG, MathML, HTML, FO, etc...)
> 
> Thanks Dom: that is exactly what I am thinking of for Eduml. 
> But since W3C does not have any Schemas for Education Interopability, we are
> forced to either:
> 
> 1. look at the Commercial "open" standards which we can slightly
> modify to fit our "opensource/freedom" culture. (which is doable now)
> 
> 2. Make our own standard and submit to W3C for adoption (which is no small feat)
> 
	Yes, the W3C will only launch a work group on strong industry support.
Maybe it would be easier to approach one of the groups that has already
connections to the learning community (I can only think of MathML for
now...)

> > 	Also, why sould we adapt other formats ? As long as they are open
> > formats I think we should just use them in the definition of EduML. This
> > would considerably lower the cost of writing the applications because a
> > lot of open tools already exist for those formats.
> 
> which is why I now favor option 1 above. (after having spent considerable
> time with option 2 in 98-99)
>  
	I'm also favorable to this option, but as long as we can, I would stick
to the existing formats, even if they may not be free in the future. I
have two reason for this:
	1. the attempt that takes place here to produce a open format needs an
enormous effort. Thus as long as possible I would try to be very
pragmatic and reuse as much as possible.
	2. Even with that new format existing, and even if it is far better
than other existing formats, I'm not quite sure it will be used by the
industry (here it's just my opinion, but based on my experience). Thus,
we will still have to be interoperable with the industry standards. The
more we stick to them, the easier this will be.

	What do you think ?

...
> 
> If instead a WIMS server exported pre-cooked randomized math exercises
> with RPC (remote procedure call), then the management of marks
> and scoring methods would devolve to the local online management tool, while
> not re-inventing Xiao Gang's wonderful wheel ... the part that generates
> dynamic math exercises, together with the answer.  
> 
> Now all we need is a driver (an RPC call, a CGI-BIN URL if you wish, let's
> say http://wims.unice.fr/oefdriver?item=trig2 which calls for a randomized rendering 
> of exercise trig2 and its answer (if it is static, otherwise another RPC
> call will be needed for the answer given the student's response as input:
> http://wims.unice.fr/oefdriver2?item=trig2&response=b)
> 
> The results from calling those URLs would be (Eduml) XML and so a driver for
> dealing with those XML responses for each different online management course
> system.
> 
> I hope that providing these semi-concrete examples is useful to you all.
> 

	Yes, I misunderstood the goal of the dynamic part. I was thinking about 
a way to represent dynamic exercises in a static format (i.e. for transmission 
or storing). What you propose is very interesting. I would put the 
example a little farther. Using a granularity of one question implies a lot
of coupling between the user application and the providing application (WIMS).
Maybe, the user application should just send a request for a test session. Then
the student goes to WIMS (in an automated manner) and responds to the questions.
Eventually, WIMS transfers the results to the user application.

	I hope this helps the discussion.
	
	Am I too quick with such discussions ? Should I wait until the different 
exercises are ended ?

	Dom