[Gnumed-devel] demographics & identity (was: describing use cases)

[J Busser]

At 8:35 AM +1100 2/13/05, E Dodd wrote:
>  > names stores all the present and past names of an individual.
> >  So when Jane Smith marries Fred Jones, her herpes PCR from
> >  two months ago (yes some labs take that long around here)
> >  will still match in the database.
> so i don't have to do a search on dob and medicare number to find the match??
> a very good idea

I'm sorry I don't yet have a linux desktop in case some of the 
behaviour is already viewable (though I may have a linux desktop in a 
week or so).

There are a few stages to what we would like to do:

1) develop use cases & decide how we want the functions to "work"
2) get the software written
3) try it out and feed back into the process

in deciding how we want functions to work we probably
need to identify what most people would feel to be
a) very useful
b) useful in some situations, without impeding general usability
c) useful to some people but the implementation would give grief to others

a) should, & probably will, be implemented, but not all at once
- anyone is welcome & encouraged to implement any & all parts
- these parts would all eventually go into the main trunk
- consensus on the most desirable path forward is desirable
   to better enable the 'active" people to collaborate

b) anyone who *really* wants to get this done is welcome to
- it may just be their passion of theirs, or vital to their use case
- it may not help others until later if ever, but it is up to the developer
- if it does not *interfere* with anything, just extends it,
   it will be welcome into the main trunk??

c) like (b), and can be developed in test-area, but should *not* go into
    the main trunk or should only go into some clearly identified subfolder
    in which optional widgets can be kept, and activated at the level of
    an individual gnumed *installation*, perhaps to replace default
    widgets. Kind of like the {richard-space, horst-space} choice.

I drafted the above only because I cannot recall any description of 
how a demographic widget might work other than the "x# of characters 
from the surname, firstname" discussions. So maybe I am missing 
something, unless people think it a worthy point in time to now 
*compile* ideas on what would be desirable/useful.

It is certainly of interest to me because I hope by May-June to have 
someone local look at where gnumed is at, and identify for me how I 
might effectively hire them to contribute code. Maybe I am 
unrealistic but if it proves possible for me to push functional 
requirement questions ahead of where things are at (to try to help 
get 0.1 "done" at least to the extent that I could use it in the 
clinic) it might stimulate more people's helpful suggestions and 
encourage as-yet unmade decisions for 0.1?

PS Would 1,2,3 a b c be worth wiki-ing i.e. would it help lurkers & 
new people identify *how* their needs might "fit" inside gnumed?