Over the past few months, I reviewed a range of (but by no means all) national eGIFs, charted out some of the resulting data and shared it at a workshop at the FutureGov2009 conference in Indonesia. What jumped off the page at me? First, the review suggests that policymakers are focusing too much on technical interoperability, which although significant a decade ago, have increasingly been worked out. Governments (and consultants) are often turning to policy tools that are largely modeled on what the UK did in 2000 (largely focused on the technical domain), to the detriment of much more substantial semantic and organizational issues that today are the main barriers to eGovernment interoperability. It’s not hard to see the common ancestry that many of these technically-oriented eGIFs share. Just one example is the so-called “8µ Law” standard, a non-existent standard that initially appeared on the UK list and then somehow wound up under consideration in at least 7 other countries in the context of their eGIF. The existence of this “standard” across so many other countries raises some important questions about the utility and effectiveness of these policy tools, nearly a decade after they were first rolled out.

Even more intriguing, one of the most touted benefits of these technical focused frameworks– that they give governments a mechanism to mandate specific technical standards which in turn result in better interoperability– doesn’t appear supported by the data. Instead, the data appear to present a case of “the tail wagging the dog.” Rather than dictating the standards that will lead to better interoperability, these standards lists largely capture the commonly used standards that the market has already embraced and agreed to use for interoperability purposes. If you look at the group of standards that are shared across roughly 80% of the standards lists (a dozen or so standards), you don’t find many surprises. What you find are core networking and interconnection standards (HTML, HTTP, XML, etc) that have long been supported by any product that wants to have any chance at achieving widespread marketshare. While there are some interesting questions raised by the many, many other standards that appear on only one or two lists, there is little evidence that eGIFs were/are being used by governments to drive adoption and use (through requirements or mandates) of various standards, with an aim toward improving technical interoperability. It’s possible that the technically-focused approach contributed a whole lot less to improvements in eGovernment interoperability over the last decade than some governments and practitioners believe.

There is a ton of work to be done in the next decade to make eGovernment work better. It is critical as we start out the next decade that we make sure we have the right mix of policy tools at hand. My preliminary look back suggests that we need to take a more thoughtful look at whether the current incarnation of eGIFs is the right starting point for the next decade.

LET is an emerging domain. Everything technological is in flux. And then we have standards that are supposed to be stable. It sounds strange, and as a matter of fact, it does not make sense. In the European project I am meeting a strong respect for what is classified as a standard (related to formal standard setting bodies). And I observe a less strong respect for what is classified as specifications (often related to user communities). Having been part of the process of producing standards I know too well that you cannot analyse standards by reputation of the organisation behind it.

European eGovernment initiatives give standards boards a prominent role in the governance of standards, also LET standards. One of the instruments of governance is a standards catalogue, which is intended to guide users towards appropriate standards to implement. Again, it goes without saying. Formal standards are more likely to end up in standards catalogues than community based standards.

Together with Paul Hollins of the UK Centre for Educational Technology and Interoperability Standards I have analysed the standards catalogue approach against a horizon scan report of current standardisation projects in the sector. First, given the emergent nature of the LET domain, we are not sure that the right candidates for recommendation are put forward. Second, we also question if bad decisions or obsolete recommendations would be withdrawn from the catalogue. The process models behind these initiatives have weak feedback mechanisms, and point more or less in one direction: towards the registry. There is a growing awareness in the LET standards community that we are not good at dispensing some of our earlier, less strong ideas, i.e., putting some of our standards into the bin.

However, what is more important is the effect the standards catalogues will have on what we would call the standards discourse. There is a growing awareness in standards bodies, e.g., demonstrated recently through a number of work method seminars organised prior to standardisation meetings in CENWS-LT, of the need for a meta level discussion on “the way we do standards”. The standards catalogue approach may steer discussions in the direction of the standards that are on the list, and what level of mandation these standards are assigned.

Therefore, we suggest that eGovernment standards boards should focus on semantic, organisational, cultural, political and legal interoperability, in preference to attempting to stabilise practice around a limited number of technical interoperability standards.

In conclusion, “unstable standards” might be more in line what the LET stakeholders need than stifled standards that are not serving innovation in the domain. Then we have just to talk about standards – and give up the hidden quality denotation in keeping up the categorisation of standards vs. specifications.