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

Re: [Gnumed-devel] Re: How to build GNUmed packages for Ubuntu from Debian

On 2-Jul-09, at 6:18 PM, Rogerio Luz Coelho wrote:

Ok it says to upload only packages with a good userbase ... so where are we now ?
Or can Debian-Med give us a hand with this ?

The challenge of "GNUmed for the masses" gets discussed about every 6-12 months, and I am happy to refresh my thoughts.

Free medical software would like to build a user base but only AFAICT has not figured out to do so. It may be partly an issue of timing until (I think) the world is "flatter", in terms of its economics, and ideology, and its willingness to share for altruistic purposes.

In the meantime I think there is no way I would have believed GNUmed could be made to work other than by and for ridiculously expert people, ***if it had not been for the packages***. The (now very old) Mac .dmg allowed me to connect, however imperfectly (and painfully slowly) to the public database. That, combined with my familiarity and dissatisfaction with the commercial medical softwares that I have seen, allowed me to know that a persistent, unafraid (but only very very modestly/minimally computer-skilled doctor) like myself could get it to work.

I think at one point I did also manage to get the server installed on Mac using a Mac Postgres package which I think took an unorthodox approach (Postgres was made to be a user that showed up in the GUI logins). I do not remember it being a happy experience.

When I found a spare machine on which to install Debian and I managed without too much difficulty (a bit of newbie clumsiness) to get sudo and the postgres config sorted out, and -- after learning things like the need to make things executable, and the required prefix ./ -- the Debian packages (server and client) made things a *breeze*.

Therefore, I do believe that there is a benefit to producing packages even before the user bases yet exists, as part of fostering the development of a user base.

Eventually, a critical mass of people -- who would do the tech support for groups of clinicians (doctors) -- would probably *want* to build and maintain the packages, if only to make their work of updating their own user (customer) base more manageable.

In the meantime it remains useful to think how to build the ecosystem:

(1) I do not think vendors will be of any help until GNUmed offers enough functionality to run a praxis. I believe that once GNUmed is able to manage and print medication lists -- modifiable to serve as prescriptions -- that time will be very close. It may depend additionally on:
- a workgroup appointment manager that can interoperate with gnumed
- a finance (billing) program that can interoperate with GNUmed
- a few extra bits of GNUmed functionality such as:
- - an expanded patient creation widget in which to also input a primary external id
- - maintaining who within the praxis is a patient's "default" provider
- - maintaining who else (outside the praxis) among a database of other doctors (GPs and specialists) have previously or currently do or have been asked in the upcoming days/weeks/months to provide care - - attaching codes to sOAp rows and/or health issues (ICD9, Reid, SNOMED)

(2) Doctors who, as a group, can see immediate value -- to their group -- from some of the limited (e.g. archiving) functionality already available in GNUmed. Karsten knows of one or a few of these.

(3) Doctors who, as individuals, can see real daily value in using GNUmed (particularly its notes and problem list) as a "shadow system" to supplement care they are already, as individuals, giving to patients. Rogerio, is this you?

(4) Doctors who can see the *potential* of GNUmed and want to help it to achieve (1). I am in this group, I am also *close* to being able to see it having value to me as an individual, I am only afraid of depending on it if and when I did not have in place an adequate infrastructure.

re Debian-Med I did already get a question offlist from someone if I knew of any way to fund people's time to help. So I think Debian-Med may remain short of help, until one or both of the following:

a) scenario (1) above comes to pass, and the IT support business, or subscribing clinicians

b) Debian-Med grows free help, through focusing of enthusiasm. Right now I suspect the help received on Debian-Med is from pure goodwill without the contributors being able to see their progress. I think if at some stage people believed there were specific packages within Debian Med that were very important to deliver into target audiences (and that the other packages in Debian Med would be a help) then I could see the possibility of more / better success,

Reply to: