[Top][All Lists]
[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
Re: modeling tools for developping AND implementing
From: |
Derek Neighbors |
Subject: |
Re: modeling tools for developping AND implementing |
Date: |
Mon, 22 Apr 2002 21:02:12 -0700 |
> So, 2 questions
sorry we have a one question limit. just kidding. ;)
> 1/ business process modeling
> to execute components or services from differents sources, one need of
> course standardisation of business objects and language to describe them.
> this technical aspect is easy to decide and weel-known (XML and dialects...)
>
> But, from an organisationnal (roles, business process..) point of view, it's
> more complex
> You have to take in account that it's not easy to understant how is working
> a standard component, what is the real adequation with existing applications
>
> Often, the culture of the component designer or developper is very different
> from those of the user... specially regarding business culture
> process modeling is mandatory a way to compare, understand, explain... HOW a
> tools can be used
Im not sure I follow the question.
> 2/ modeling for desing and modeling for use
> Of course, we all know different modelisation technics, as SADT, MERISE (for
> oldiers !) and UML...
>
> But is process modeling the same activity whan one model a process to desing
> a future tools, and when one model a process ton understant who he can use a
> tools ?
>
> From my experience, i think that it is two different activities... because,
> when i model a process for training, simulating, consulting in an
> implementation project, i don't want to know how the package is working
> internally, but what human actors of the system have to do... I do'nt want
> to model the system.. but the use of the system...
>
> One can say that there is a difference because my experience is on classic
> packages, and that free approach let me know exactly HOW the system is
> working internally, so i can explain and decide exactly what i want.. and if
> needing, that i can customize the tools...
>
> What's your opinion ?
Well for now I think we need to get GNUe working. Then grow it into a build a
UML diagram and press some buttons and your 'objects/schemas' are updated on
the fly. I agree to a degree that 'developers' and 'users' might see the
process slightly different. I think these are good questions and I want to
disvalue them, but at this point unless someone wants to step up to the plate,
we are over worked on the stuff we have already bitten off.
I still think it would relatively elementary to make a dia xml out put be
transformed by xslt into our database schema format and someday directly into
our class file definitions for the Application Server. Just I dont have the
time to tackle right now, nor does anyone else currently contributing. Any
volunteers?
> 3/ Suggestion
> whatever is the correct answer... I think that a process modeling tools to
> define what services the gnue packages can deliver would be very
> interesting..
I agree.
> I can imagine a project next year with student on such a subject...
That would be great. :)
-Derek