Re: Fwd: Re: [Gnumed-devel] GNUmed brochures (was When will GNUmed be ready)

[Sebastian Hilbert]

On Thursday 15 December 2005 16:44, Hilmar Berger wrote:
> Hi,
>
> > --- Ursprüngliche Nachricht ---
> > Von: Sebastian Hilbert <address@hidden>
> >
> > > So what is GNUmed ? A collection of medical plugins ? A
> > > framework for developing health care applications for doctors ?
> >
> > Depends on you audience
> >
> > > What are you trying to achieve ?
> >
> > Depends on you audience
>
> I don't believe that your goals depend on your audience in any way. Maybe
> you will describe them differently, the goals itself should remain the
> same.
That is what I meant. 

>
>  > I really doubt that you will attract anyone without telling him what
> >
> > GNUmed
> >
> > > is supposed to be (project vision).
> >
> > True. We disagree on the scope or definition of vision.
> > I envision short term goals.
>
> Which is only possible because someone (I guess it were Horst, Karsten,
> Ian, Richard ) a long time ago had a vision of what they wanted to build.
Quite possible. I have seen no vision having been written down. That doesn't 
mean it does not exist. The problem we had and still have that those visions 
have never been written down. It became apparent that those visions exist are 
quite different. 

I can only guess here but here is what I seem to know. Guys, correct me if I 
am wrong.
Richard:
- is looking for an all or nothing for AU
- envisions a development style which involves gui mockups and talking to 
clients before coding

Horst:
- has a need for GNUmed right now
- forked / rewrote something himself because GNUmed does not proceed fast 
enough
- seems to have found a way to use his version of GNUmed as well as a legacy 
software

Karsten:
- doesn't support quick hacks to support 
        - avoid the need to fork like Horst did
- does not support the  all or nothing approach
- uses GNUmed or parts of it already
- codes with his need in mind but keeps it open as much as possible

Sebastian:
- believes there is no all or nothing approach when seeing GNUmed as a 
community project
- has no problem marketing parts of GNUmed
- collects feedback from potential users
- thinks about extending GNUmed via commercial coding effort
>
> > > I agree that we will have to tell what parts are ready. Still, IMHO the
> > > vision is the most important part of the message.
> >
> > We still disagree. Vision is good for investors and academic people.
>
> How can I convince you that having a clear vision of what you want to build
> is no marketing buzz or academic nonsense ?
You don't have to convince me. All I say is you can envision all you want and 
describe and model as much as you want. This does not replace the work that 
needs to be done. I have come to believe there is no or little community that 
can *finish* GNUmed. We are a talented team but I don't see any developers 
hiding in the dark.

Spelling out the vision is helpful. It helps interested parties to see where 
some people think GNUmed shold be going. One has to decide. Either believe in 
the coder myth and try to attract coders or code it yourself because there 
are no coders in the dark. In that case one does not have to write down a 
vision. 

There is absolutely no use in trying to convince each other that there is or 
is no secret coder army. The beauty of this project is that perceptions can 
coexist. You can write down your visions and Karsten can code. None of the 
two approaches is worse than the other.

There is only one thing that sucks. Doing neither but talking about doing 
either.
> If you build a house, you will sit down first and make a plan of what you
> need exactly. Then you ask an architect to make a plan of how the house
> will look like and how it will be built.
Yes and no. It depends. I can be architect  and builder if I don't expect 
anyone to help. While you plan the mansion with the pool I build the garage.
Don't tary to discuss which is better. Only time will tell. I will build the 
garage in a fashion that in ten years noone will see my mansion started with 
a garage. You might suffer from the problem that your beautiful drawing of 
the mansion will never see one brick put on top of the other. No matter how 
excellent your plan is.

Only time will tell.

> The same goes for software engineering. First there is a vision (a goal),
> then you make up your specs, then you start to code.
No. Absolutely not. There *are* different models. Both have the potential to 
fails as well as to succeed. You live in Germany. Reads the facts how many 
huge scale projects have failed over the last two years. Can you say 
Herkules ? Can you say AlG II ? I am pretty sure they did any excellent job 
modeling. They probabyl still do. Open your eyes. There is more than one 
approach.

> Without a clear vision, your specs will be wrong (written down or not),
> which will finally result in useless code.
Absolutely not true. I have no idea what gives you this impression. A vision 
can develop while you code. It takes longer but it is way better than a 
vision that is screwed.

>
> > Doesn't
> > help doctors a bit. Just read facharzt.de (German) and you will get a
> > feeling
> > for the issues that bother doctors. They want solutions now.
>
> There is no way to good solutions without hard thinking before, I guess.
This is true but it does not support your hypothesis in any way.
>
> > They are fed up with vaporware. Universities produce vapoware all the
>
> time.
> Sure, Universities are supposed to produce good science, not good software
> applications.
They are tryng to sell us the idea that we have to plan a software product in 
a certain way. They imply that software writes itself once it has been 
planned good enough. Forget this. If we plan academically and still fail they 
will thell us that we were bad academics rather than admitting that their 
model might be screwed.
>
> > Ask any doctor in Leipzig. They don't care sh.. about GNUmed's long term
> > goals.
>
> I guess the doctors do care for what Gnumed will be able to do finally.
Yes but they don't care what you call it. Call it GNUmed or blahmed. This has 
nothing to do with GNUmed itself. I can come up with ideas all day long. This 
does not help GNUmed the least. Many many people are fooled by the idea that 
thinking replaces coding or other work.

> They just want to know if right now GNUmed matches their idea (=vision) of
> what a medical software should look like.
Maybe. As long as they don't say what they want I couldn't care less. I 
support people who say what they want. I talk to doctors all the time. They 
want to hear what GNUmed is and then tell me why GNUmed sucks. They are not 
puuting any effort into telling me what they think what it should look like.
>
As much as I enjoy out little academic yes no maybe discussion. Neither you or 
me know the truth. So why don't you ask the people we talk about what they 
want. *Then* you know what they want. I guess right now you only know what 
you think it is they want. Prove me wrong. 

Thinking about it I get the impression when you say *they want* you mean I 
want ... That makes sense and is ok with me.

I am not trying to convince you to stop thinking that way. I would be more 
than delighted to see your ideas and vision being put down in our Wiki for 
other people to see.
-- 
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