Re: [Gnumed-devel] Re: get something done

[Sebastian Hilbert]

On Wednesday 30 November 2005 12:57, Richard Hosking wrote:

Hi Richard,

> I am glad someone has said this - I have been reluctant as a newbie as
> it may already have been said and done before.
> I have difficulty visualizing Gnumed as a model. The Wiki does have some
> stuff, but the ideal would be a visual model starting at the "context"
> level diagram - ie the inputs and outputs to the system as a whole. The
Personally I lack the knowledge to follow you but please elaborate.
> model is developed in progressive detail, drilling down to a module level.
> Lack of upfront modelling and design effort is a weak point in OSS
> software as a whole. The software eventually evolves into something
> useful as nothing is lost, but it might take a few million years.. :)
>
> I agree there is a real risk that the work so far will be lost, and that
> this risk could be reduced by a phase of planning/design according to
> one of the accepted methodologies in software design
While I mostly seem to justify the status quo I am willing to work with you on 
this. Please tell me more what you would like to see.

Some graphical representation of GNUmed. Do you have a specific idea ?
Since I have never attended any CS classes you guys need to help me out on 
this. Are there any tools to do this, any language or is it just drawing 
boxes , circles arrows on a piece of paper ?

I really have no idea.
Sebastian Hilbert

>
> Could the students be better employed in gathering requirements and
> modelling, rather than rushing to code?
They can be employed in anything we offer. They will works with us not the 
other way around, I am afraid. I can only guess but telling them to gather 
and model requirement would leave them lost. Heck, I couldn't do it if I 
wanted. 

On a side note. Ask any question you want. There is no such thing as has been 
asked before. 
>
> Richard
>
> Hilmar Berger wrote:
> >On Mon, 28 Nov 2005 21:52:24 +0100
> >
> >Sebastian Hilbert <address@hidden> wrote:
> >>>I do not think that GNUmed sucks currently.  But I'm afraid it will suck
> >
> >at
> >
> >>>one day if we continue working as we do currently because the way it is
> >>>done currently does not scale for a larger project (IMHO).
> >>
> >>Please suggest improvements. Privately if neccessary. Thanks.
> >
> >What I'm missing most in GnuMed is a proper specification of what we want
> > to achieve, i.e. project vision, detailed requirements for all planned
> > modules, design documents (what is/should be done in which way),
> > assignment of work items etc.
> >
> >I am aware of the fact that there is a lot of information distributed in
> > the mailing list and in pieces in the wiki. Still, if I should point
> > somebody (say, an informatics student going to do some work for us) to
> > some place were he can find all I wouldn't know what to say.
> >
> >Over the last years I learned that large and complex projects without a
> >final specification *before* starting to work will either have
> > difficulties or fail completely. I'm quite convinced that GnuMed is
> > complex enough to need a specification. I'm sure Karsten knows exactly
> > how he wants to build GnuMed (as knows Ian, as knows Richard, ...),
> > however, I believe we would be able to get more collaborators if we could
> > only show a well-thought plan what milestones we want to achieve next.
> >
> >I therefore suggest that at least some energy should be spent in
> > documenting and discussing our plans for the nearer future.
> >The specification should probably go into the Wiki, once this is up and
> >running again.
> >
> >>My bad, I admit. IT students, course assignments for one semester. 2
> >
> >students
> >
> >>in each group. That's all I know and it's not even certain.
> >
> >My limited experience with these students is that they will need tight &
> >frequent interaction with their project manager, detailed instruction on
> >what exactly they should try to do and if possible strict checks if they
> >conform to the design principles of the software they are coding for.
> >Usually these guys don't have much experience nor in medicine neither in
> >delivering quality parts for already existing software, so that's
> > something you must try to enforce by some way. If not, we will end up
> > with a lot of semi-functional prototypes that might need more work to fix
> > than to write from scratch.
> >
> >Hilmar
>
> _______________________________________________
> Gnumed-devel mailing list
> address@hidden
> http://lists.gnu.org/mailman/listinfo/gnumed-devel

-- 
Sebastian Hilbert 
Leipzig / Germany
[www.openmed.org]  -> PGP welcome, HTML ->/dev/null
ICQ: 86 07 67 86   -> No files, no URL's
VoIP: callto://address@hidden
My OS: Suse Linux. Geek by Nature, Linux by Choice