recent DISCUSSION

• Ken Krechmer says:

  Dear Ajit,

  Not sure I understand your comment, “especially in context of selection of fora in specific domains” Different fora do not have a unique expertise for future-proofing to offer. The goal of future-proofing needs to be made clear to any fora. To my knowledge adaptability standards are the only way to future-proof a very large complex communication system.
• Ajit Jaokar says:

  Ken says under addressing the holes in the current standards systems
  2. Future proof systems:
  There are systems being created now – Internet IPv6, smart grid, electronic health care records – that will have billions of instantiations, tens of thousands of implementations, their standards will change thousands of times and the systems will last for 100s of years. Current technology does not have ways to assure the absolute long term viability of such systems. Commercial organizations are not capable of addressing such problems. Government research and direction is needed.

  I think this is an important point which has been missed especially in context of selection of fora in specific domains. In more complex scenarios, such as smart grids, there is often a hierarchy of standard which have to be collectively addressed. Often in many cases, we are addressing the domain for the first time(like smart grid). In that context, selecting specific fora in a space will be very hard. So, while Ken did not raise this point directly i.e. the relation between futureproofing and fora .. I think it is a critical point which occured to me as I read his post. Perhaps to be addressed in subsequent discussions or posts!
• Ken Krechmer says:

  Dear Peter,

  There is no conflict when it is recognized that adaptability standards (standards which do not pass data or control, but only negotiate data and control capability) are the means to “future proof systems.” Since adaptability is a newer concept, it is not widely understood or practiced. Where it has been practiced, (e.g Group 3 facsimile, telephone modems), very large communications systems have remained compatible and changeable for 40-50 years.

  This is why it is very important for the EU to focus on supporting adaptability standards.
• Istvan Sebestyen says:

  Tim Simcoe: “In principle, couldn’t you allow two working groups within a large consortium to pursue parallel paths to some problem, and then rely on the market to sort things our ex post?”
  Yes, this is very often the case how it works in practice. Usually the overlapping is also partly as requirements are always slightly different, plus usually there are some time differences in standardization… etc… Last, but not least, there is no central standards police that takes such decision and enforces it. It is the members who are deciding usually with compromises. It is usually the market who then makes the selection.
• Tim Simcoe says:

  I have a question about Ken’s point #4, “avoid competition within consortia.” In principle, couldn’t you allow two working groups within a large consortium to pursue parallel paths to some problem, and then rely on the market to sort things our ex post? I always thought that the IETF did a bit of this, and that it was an implicit part of the “rough consensus and running code” philosophy. Perhaps you are referring to smaller more-focused consortia, where the tension between factions could easily produce a deadlock, and it would be relatively straightforward to split the camps into separate organizations?
• Istvan Sebestyen says:

  Peter Gray: I do not think that there is a conflict what Ken says, maybe the formulation might be better. Many standards come in cycles, dynamically adapting and improving (e.g. Facsimile Gr.3 was a good example for this, as it got added added transmission speeds, color, higher resolutions etc.). There it was needed. Now it seem that the technology has been replaced by email largely, and certainly not much is going on in Facs. 3 standardization.
  On the other hand, e.g. we do not have good methods that makes it sure that today a document that has been created in a computer, can be used in say 50-100 years time. The hardware will probably not exist, the storage media player will likely not work anymore, and storage medium will likely be unreadable etc. So, this is an example for an area (and not the only one) where future proof systems will be needed. You may read the information on the Rosetta Stone after 1000s of years, also paper is quite good, but you can not read a 5 1/4 inch floppy disc today that was produced on a computer 20 years ago. Problems both with hardware and software standard. I think this was what Ken has written.
• Tim Simcoe says:

  I think JBC’s post makes a very good point about the often overlooked role of certification and testing in delivering true inter-operability. Perhaps our theories and conversations stop too often at the point where a specification gets published. But let me offer a note of caution about the optimistic claim that economic theory predicts that the “standards market” will deliver the optimal level interoperability. It doesn’t! What makes standards and compatibility a fascinating topic for economists is the joint gains from coordination. Once you introduce increasing returns into a Walrasian economy, you get many possible equilibria, and there is no guarantee that we will find “the best.” While the laissez-faire branch of the economics profession offers some good arguments that we do tend to end up in reasonably good equilibria, there is no strong theoretical basis for those optimistic arguments. All of this goes double for Helen’s arguments that government procurement policies that favor open standards lead to sub-optimal innovation incentives. Open standards may reduce some patent-holders’ profits, but they also lower input prices for innovators that wish to build on the standard. At present, it’s very hard to say where we are relative to the “optimal” level of innovation incentives… since that’s a very hard thing to measure!
• Helen Disney says:

  @Peter – why is one better than the other? Why not use both?
• Niklas Lindblom says:

  @Knut, given the transition into an ever more globalized society, what are your views on the Commission’s approach to international consortia? Also, governments pose as important consumers through public procurement. Is there a risk in preventing international interoperability and compliance with different governments settling for different standards, both within and outside the EU? //Niklas
• Niklas Lindblom says:

  @James, although you are right about the basic economic theories, it seems as if we are deviating from this certain equilibrium. After all, lack of interoperability has repeatedly been pointed out as an issue in standards development and implementation. What appears to me is that demand changes a lot quicker than supply. And that’s not very surprising; it’s easier for consumers to say that they want and need more interoperability now than for developers to instantly comply. What are your views on speeding up e.g. the conformance tests in the development process? //Niklas
• Peter Gray says:

  @Ken Krechmer. You list “future proof systems” as a hole within the EU standardization system. This seems to conflict with number 6 in which you highlight the need for dynamic standardisation with a realisation that standards will evolve over time.
• Peter Gray says:

  @Helen Disney. You describe favouring technology based on open (non-proprietary) standards within government procurement policy as a “worrying trend”. However this is a far more desirable outcome than the alternative, not just from an interoperability perspective.