I’ve had a chance to play around with the Open Web Apps module, and it’s awesome. I think it’s going to be transformative for OpenMRS as a community, since it lets someone with a standard HTML+JS skill set write code the way they are used to, on top of the standalone, without knowing anything about historic OpenMRS technologies.
I’ve written a tutorial, and improvements are very welcome since I’m not fluent in best JavaScript dev practices. (If you want to see a bit of this in action, join tomorrow’s Dev Forum.)
I propose that we include the Open Web Apps module in the Platform 2.0 release (obviously we’ll miss the Alpha, but we can include it in the Beta). And we should make this a community-supported module, and transfer the repository from @sunbiz to OpenMRS. Any reason not to do this?
I agree with the transfer of the repository to OpenMRS… and the transformative power of choice that OWA provides. In fact it should allow individuals with more than HTML+JS, and including php, asp and what have you, since it allows separating the UI layer.
Tangential thread:
I feel that as a paradigm shift, we should try and think of RESTful services, and build UI components. My biggest gripe for the systems we build IMO is that we know that implementations are different, yet we are doing very little to solve this. Users without programming skills should be able to customize every part of the EMR. I think this can be done with flexible UI components… and another part of it is taking advantage of browser/device APIs. So online-offline hybrids in UI.
+1 to bundling OWA as a community-supported module.
+1 for evolving UI components… ideally a set of useful examples to seed a library of components into which community members could contribute new ones.
I’m a little weary about promoting use of device APIs that are inherently insecure (e.g., localStorage) unless we can provide easy ways to use them with at least some level of security. People can, of course, do whatever they please, but I believe we have a responsibility to promote and teach safe use of patient data. For example, I believe @jdick has developed some methods to encrypt identifiable data stored locally with user’s salted password. Maybe these concerns will be moot as device technology becomes more secure.