Re: [school-discuss] EduML/EML/ and others

[mikee <meschman@engima.com>]

hello,

designs have to be prototyped to be successful.
module level prototyping is cost effective
and well suited to collaboration.

writing a 5000 line specification without benefit of feedback
from working code is risky at best.  that's why, in some cases,  w3c builds 
reference implementation code throughout the development of the 
specification. [ ref. jigsaw for http and amaya for browsers ].

sometimes committee members spend the whole session reviewing 
prototype code.

we ported jigsaw to s/390 unix services and
wrote a redbook on this together with ibm.

the redbook can be downloaded from our HISTORY page. (.pdf).

mike eschman, etc...
"Not just an afterthought ...
http://www.engima.com



On Wednesday 10 April 2002 08:22 pm, you wrote:
> On Wed, 10 Apr 2002, Les Richardson wrote:
> > When you have this kind of thing, you usually have a "model"
> > implementation. I would say "Write the app and make it work. Get it
> > used." Then Model the markup.
> >
> > Anything else I would say is premature...
>
> I would disagree, by writing the app to "model" the markup so to speaou
> leave alot to be desired. Is it truly able to be extended to other
> applications, does your one application truly show everything you want in
> the schema. In other words lets say you create a grade book. You use XML
> to do so you are creating EduML or some variation. Now you have this
> application and you begin working with it and blah blah blah. Now it comes
> time to model the data. Now all you have modeled with your application is
> the grading process and data. Well there is more to the proposed EDU XML
> then just grading or anything that one application can take into account
> for. Also you have developers who don't like to do non-standard work, so
> they will "just wait till there is a standard". I believe that if you sit
> down and abstract it get an idea of where you want to take it before you
> write any code then you are ahead of the game.
>
> Maybe its because I dont have a Computer Science Degree that I think this
> way. But I am a firm believe in knowing the way before you right the code.
> A lot of the "bad code" is written by people who dont know where they are
> going with the application.
>
> I am rambling. But I feel that it is time to get up and be heard, its time
> to do the work, and stop the dream. You know I understand dreams are good
> but what good is a dream if you dont work towards it with full force?
>
> Sorry for the ramble