Re: [Gnumed-devel] Time for a major re-think in 2005 - opinions please.

[Karsten Hilbert]

Hello Tim,

it's good to have you around.

> My take on this is as an interested bystander who a) has (wearing public 
> health hat) a professional interest in seeing GNUmed suceed as a vehicle 
> and platform for furthering public health in primary care;
That is your "general" use case. I have no problem with
helping to make that happen if you define specific use cases
(eg. your version of minimum requirements to make something
happen for you) and keep pushing them.

> It seems to me that since Aug 2003, there has been a significant 
> increase in activity and code writing, but remarkably little significant 
> progress towards a working prototype.
Actually, *real* progress has only accelerated starting in
early 2004 if you ask me. This was progress in the project
becoming more focussed and slightly better organized. Core
generic code had been written before that (database access,
logging etc) but specific business level logic started to
develop only then in a way that would satisfy me for 0.1. In
the further course of 2004 several "decisions" were taken to
clean up the design/code structure which have since been shown
to be better manageable. Later in 2004 Carlos joined the
coding team which significantly boosted progress towards the
real 0.1 release candidate.

> I regard a working prototype (by 
> which I mean something which has the minimum set of essential features - 
> and defining that minimum set is an essential and early task in itself)
I agree. See below.

> Gnumed seems like a large and ambitious project
Which I for one never had the intention for it to be. We do
carry that message as Richard pointed out and we need to change
that to something saner. I have, however, never harboured any
large ambitions but rather always called for *focussing*
(people who know me personally can attest to the fact that I
can be quite stubborn if I think something should be done
*right* to the point of being "difficult" but that I am not at
all into grand scheming). Lately I have called that differently:
"work against use cases" by which I mean someone saying: "I need
*this* to accomplish *that* task. Let's make GnuMed do that".
One thing I have always urged people to do, however, is to
write *good* code.

> But at the moment, its not even at alpha status. The reasons, I think, 
> is because there is no shared, absolutely explicit list of minimum 
> requirements
There is. Few people, however, stick to working on it. Carlos
does. I do. Not sure about the others. It is here:

http://lists.gnu.org/archive/html/gnumed-devel/2003-06/msg00045.html

Note that this talks about functionality only. The design
specs were always to be intended Richard-space with
sub-standard Richard-alikes for 0.1 where necessary. I have
always told Richard that it may be too involved to completely
flesh out his integrated design in 0.1 but nevertheless use
it as our design guidelines. Lately I have made efforts to
better stick to it and deviate less or in other words actively
work with Richard seeking his input on design issues letting
him be the area-of-interest decision maker. This took some
unconscious "letting go" on my part but I figured GnuMed would
be better off. Here I would suggest not changing the design
specs too much before 0.1...

Also note that we aren't doing too bad on that. The one thing
we don't yet have in a maintainable working state is 5) drug
information browser if data available - eg. interfacing
locally available drug databases for information browsing.

Keeping progress notes (2) is coming along nicely. A maintainable
browser/exporter (3) is there. Input is being worked on actively.

3rd party interfacing (4) works.

The document archive (6) is deployed.

ConfigRegistry browser (7) is 0.1-ready.

Patient search (1) is 0.1-ready. Input/modification are still in
dire straits but Ian started to work on that.

> (and underlying architecture) for Version 1.0
There is but it is not documented outside the code itself
(wherein it *is* documented). We have made efforts to better
transport that documentation for the schema and the API. We
desperately need to get the Wiki back online.

> and no agreement to stick to that minimum list.
I suppose focussing on that is the main problem.

> Inevitably, such a minimum list 
> will never be totally "correct" for everyone
Fully agree.

> or even anyone, but that 
> doesn't matter - at least it enables a reasonable, working Version 1.0 
> to be produced.
Thanks for saying that out aloud. Some people don't seem to
believe this is true.

> Without the "plausible promise" of a working alpha 
> version, Gnumed will never attract the full-time resources (volunteer or 
> paid) needed to allow it to get past the alpha stage, I fear.
I take your word for it.

> So, my (unsolicited) advice is: draw up a set of minimum requirements 
> and an agreed architecture, and then stick to it, and vary the 
> requirements and architecture only reluctantly and after careful 
> consideration of the resource and time implications.
I *try* to do so but cannot be the only one "bickering about"
when new ideas come up. I am not the manager. Just a coder who
happens to stubbornly insist on what he thinks in his
non-CS-educated mind to be good code.

Regards,
Karsten
-- 
GPG key ID E4071346 @ wwwkeys.pgp.net
E167 67FD A291 2BEA 73BD  4537 78B9 A9F9 E407 1346