[Top][All Lists]

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: [Gnumed-devel] Gui-Designers was the id_name debate

From: Karsten Hilbert
Subject: Re: [Gnumed-devel] Gui-Designers was the id_name debate
Date: Thu, 9 Sep 2004 16:41:04 +0200
User-agent: Mutt/

> Should we maybe take the view that even if though the organization 
> and content of "Richard-space" widgets will be different from 
> Horst's, the method of building widgets (wxGlade etc) that Horst used 
> can/should be more closely followed?
"Our" way of building widgets has been like this: Sebastian or
I came up with a use case. Sebastian designed the plugin widget
in wxGlade with some random suggestions thrown in by me. He
did or did not add in relevant calls to business objects or
even the backend. I reviewed the code, changed, broke
it, complained, had him fix my mistakes and checked it
in :-)

The one thing we should take a closer look at is what Horst
claimed he does to decouple widget coding ("bone code") from
"meat code" (business logic). I am not sure I entirely
understand what he does *exactly* and whether (and if how)
that helps any.

> This could allow Horst, though 
> he does not have time to do the coding, to comment when he can on any 
> posted problems he has already had a chance to figure out.
That would be helpful.

> Moreover, 
> by pursuing the same approach, we may catch up to places where Horst 
> had to stop, and may together dig up solutions that will serve the 
> needs of both sets of widgets (Horst space and Richard space)
Unfortunately, Horst had (found it necessary to) decided that
he would not be using the existing framework for his
Better-MDW, sometimes dubbed Mini-GnuMed. Hence we'll have a
bit more difficulty carrying over actual code. Nonetheless
concepts and techniques may be learned from it.

> If the above has any merit, do we still need 2 things (adding to Karsten):
> 1) a statement by individuals or groups of what it would take (the 
> minimum requirements) for gnumed to begin to serve *some* limited but 
> useful *real* function, i.e. while continuing to use other 
> applications for what gnumed cannot do (iwhich BTW is what Horst did)
Absolutely. That is IMO *the* way to deployment.

> 2) seeing what these people are willing to do to get gnumed to meet 
> the minimum requirements (do = coding and/or financially supporting 
> coding and/or helping with  the legwork to get it done)
I am hesitant as concerns funding efforts before some sort
of legal entity is established that can handle such matters

> In the past 3 months, purely as an experiment to try to move toward 
> an EMR (and for no other reason) I have spent about $Can 5,000 on a 
> server and hardware firewall and $Can 4,000 on support for server 
> set-up (Debian, packages, Apache/Tomcat and the firewall, also 
> security/performance/reliability testing, documentation), 
> trouble-shooting my pesky ISP and helping me to learn what I need to 
> co-administer the server.
This is indeed one of the things potential users can do.
Enable themselves. This is also the business opportunity for
someone else. Enable potential users. Find their use case.
Solve their use case. I am willing to help with that last

> and certainly want to pitch in whatever support I can to get 
> gnumed implemented. This includes at least *trying* gnumed once it is 
> closer to minimum function and seeing what can be done to help it get 
> to minimum function.
So, what IS your "minimum function" ? I remember you have a
need for path lab handling. Can you put it in writing and
research what needs to be done to GnuMed to make your
particular use case happen ?

> I would contribute to code funding to get a 
> solution to work for me,
Even if you don't *fund* the indispensable first step is to
clearly write down your target incentive, your requirements
and your time frame. So you can point people to that Wiki or
to that posting's URL like "look, here, that's what I need".

> but I would prefer to have some realistic 
> possibility that the time and money will not be wasted.
Absolutely, see above. Open books would IMO be used for that
legal entity.

> I cannot yet 
> tell for example, even if I were to ask local people to code 
> something for me, if their product can truly contribute usefully to a 
> growing code base. While it may be possible, I am concerned that 
> these people would need some active direction on what and how to 
> build the modules and it is direction that I am unable to give,
I have not seen any of them locals asking for any sort of
directions or guidance ? Also see the answer to Syan's post on
"throwaway clients".

GPG key ID E4071346 @
E167 67FD A291 2BEA 73BD  4537 78B9 A9F9 E407 1346

reply via email to

[Prev in Thread] Current Thread [Next in Thread]