Convert your FHIR JSON -> XML and back here. The CDA Book is sometimes listed for Kindle here and it is also SHIPPING from Amazon! See here for Errata.

Friday, February 14, 2014

Easy is Hard

There's a natural tension in standards development between learning the methodology of an SDO and making that methodology easy to use.  The problems we work on are difficult, one would expect that the people who solve them are big-brained, and can learn complex methodologies.  The same can also be said of the practice of medicine.

However, true brilliance in this space makes the complicated look easy and obvious.  You know you've got it right when you can explain it to your mother, or perhaps your teenager.  Unfortunately, it isn't readilly apparent that anybody has found a methodology for that yet, let alone made that easy and obvious to practice.  So, we blunder along, and occasionally something like FHIR comes along.

Yet there is a method in all the madness.  FHIR arose out of one developer's frustration with all that was wrong with HL7 V3.  FHIR rejects complexity as being necessary, and embraces simplicity as one its principles.  One of the most often heard complaints from the congnoscenti is about the "80/20" rule that FHIR applies.  When I hear one or two complaining loudly about how some feature should be in the 80%, I laugh, because the metric is working.  It's causing pain and thought, instead of taking the easy road.  It's easy to add everything including the kitchen sink.  It's hard to figure out what is the essential requirement.  We should be complaining on the boundaries.

But here's the easy metric.  If you are among the 1 or 2 of a dozen who feel that a feature should be in the 80%, you are probably wrong.  If not, you'd be able to get at least one or two others to agree with you. And since you cannot, clearly it isn't there.  Being on the wrong end of that sort of governance is frustrating.  Hell, even being on the right end of it is frustrating.  Frustrating because it's difficult.

Nobody said your job was going to be easy.  After all, that's why they pay you the big bucks, is to do the hard stuff.  So let's do it.


  1. FHIR's not easy. It's just another shiny object.

    It might work in the small. Try scaling it to a national level.

  2. Have you tried it? I have. Starting with nothing. I like it, and I expect will try to scale it to a national level. Have you something better to offer?