Friday, August 16, 2013

Why standards should be ignored

There are many good reasons to follow standards but, I would argue, also as many to ignore them. It all depends on the task on hand and the circumstances…

An example often quoted to illustrate the importance of standards, and the need to follow them rigorously, is the case of a joint US-European project to build a satellite – scientists on both sides of the Atlantic worked to a common blueprint but each group used different units of measurement, resulting in incompatibility of finished components.

But this story also illustrates that existence of standards and blind following without applying a common sense, is not enough to prevent disasters. Standards offer framework to work with, not the formula for success (unfortunately, this is how the benefit of using standards if often sold to the public). As a framework, they offer useful guiding principles but if all standards are followed rigorously all the time, there would be no progress… Just think about it for a moment:

  • There would be no Google Maps and myriad of slippy maps that followed because there was already a good standard that could do the same, and more – I am talking about much underrated Web Map Service (WMS) standard here.
  • There would be no KML because there was already in place a very comprehensive GML standard for 2D spatial data, and 3D version was on agenda.
  • There would be no GeoJSON data format, that powers myriad of online mapping applications, and no REST service because there was already comprehensively documented Web Feature Service (WFS) standard.

Map tiles, KML and GeoJSON became de facto standards and have much bigger following than their officially sanctioned counterparts. There is your answer.