David Guest <[EMAIL PROTECTED]> wrote:
> 
> Tim Churches wrote:
> > A very good point, which Richard Terry has also made many times as 
> > well. There is a clear difference between the acceptable level of complexity
> > of user interfaces that skilled professionals are expected to use 
> > every single working day, and a user interfaces that neophyte users can
> > assimilate and understand in 30 seconds. Far too many software 
> > designers implement the latter, because they look nicer in the marketing 
> > brochures and during 10 minute demos of the software, and don't produce an
> > initial shock reaction of "uggh, that's so complex, I'll never understand 
> > it".
> > But utility is more important than aesthetics for something that is 
> > used every day and which by necessity becomes as familiar in layout as the
> > inside of your own mouth. That is not to say that complex and
> > information-dense displays can't be made aesthetically pleasing, but 
> > it does require more thought and creativity on the part of the designer.
>
>  We probably need a myghty solution (http://www.myghty.org/).

Not really, that is just a Web templating system (one of many).

What we really need is an iGoogle-style interface. Have a look at 
http://www.google.com.au/ig and, if you have a Google account, sign in. Each of 
the panels can be dragged around the screen, hidden etc. If you sign in, it 
remembers all these customisations and also allows you to choose from hundreds 
of other panel applets which can be added to your home page - these applets are 
written to a standard API - some are done by Google but many are done by third 
parties. 

So, imagine, instead of the Google search box etc at the top of the page, it 
displays some basic patient demographics and a patient search/selection 
mechanism. In the lower part of the page, a configurable array of information 
panels which collect and display information about that patient, including 
latest lab results, reminder lists for screenings and immunisations, 
visualisations of their progress over time (eg BMI, BP or BSL charts etc), 
decision support panels and so on. There could even be some user-defined rules 
which allow some panels to be substituted for others depending on the 
characteristics of each patient. Some of the panels could be editable to allow 
progress notes, problem lists and meds to be updated. All panels work via AJAX 
so there is no need to reload the browser page, except perhaps when changing 
from one patient to another. Keep multiple patients' records open in multiple 
tabs in your browser.

The point is that Google demonstrably has all the technology to do this, and 
all the post-docs and software engineers it needs to be able to hook it all up 
to a solid database back-end. It could sell the thing as a hosted service or as 
a locally installable server as it does with its search engine (see 
http://www.google.com.au/enterprise/index.html - with Google-hosted automated 
back-ups of the local server data). You'd have to pay Google to use it but then 
it wouldn't need to be supported by advertising - and Google's enormous 
economies of scale mean that it could end up being rather cheap for GPs and 
still profitable for Google. Other smart organisations could also do it. All 
that is lacking is will to do it and a few million in funding to get it 
started. And, as Google so clearly understands, there is ginormous advantage in 
providing a standard and open API,t thus allowing third parties to create and 
install plug-ins. That way Jon Patrick's group at Sydney Uni could easil!
 y develop plug-ins that understand natural language English expressions and 
convert them into SNOMED CT codes all related to each other via an ontology, 
and Kuangie could even do a DOCLE plug-in panel, and I could add 
epidemiological analysis panels. And Pfizer could do a Lipitor panel as well - 
but you wouldn't be forced to install it.

Tim C
_______________________________________________
Gpcg_talk mailing list
[email protected]
http://ozdocit.org/cgi-bin/mailman/listinfo/gpcg_talk

Reply via email to