European Standards and Innovation Policy

I am writing this sitting on the Eurostar from Brussels to London, having just attended an event organised by the Centre for European Policy Studies to discuss “EU Innovation Policy and the Role of Standards”. It was an exciting chance to get to express a view to some of the movers and shaker in and around the European Commission, as well as to take some photos of Brussels and get some last-minute Christmas shopping done (chocolates!).

Brussels by Night #1
Brussels by Night — Église Sainte-Marie.

As the meeting was quite brief, I had decided in advance that any message I wanted to get across would need to attempt to be both honed and compact. I also thought it was likely that the issue of OOXML might be raised – so I was fully ready on that score too, even though I took care to avoid the “single issue politics” approach which seems to have characterised some of the debate on this topic. As it was, the topic was hardly raised and when it was I was glad to be able to put straight some misconceptions floating around about the difference between the PAS and Fast Track procedures.

Over the course of the meeting we heard from Renate Weissenhorn (DG Industry, Head of Standardization Unit) on the importance of standardisation for innovation in the EU, in a presentation which nicely set the scene for what followed. Knut Blind (Berlin University) had carried-out a deep study of some of the subtle interrelations (among other things) between standards and legislation. His presentation, like that of Anne Lehouck (DG Industry, ICT for Competitiveness and Innovation) made me appreciate that if I had thought international standardisation is a complex system, it is as nothing compared to the complexity of trying to set policy in a world with so many disparate standardisation bodies and concerns. Ms Lehouck outlined some of the things which would, and would not, appear in a forthcoming white paper on EU standardisation policy. One thing that particularly interested me was that the EU had decided (of course) on the primacy of IETF specifications in their area, even though IETF is not a formally EU-recognised body. I did say it was complex …

For my own presentation, the need for brevity meant there were quite a few interesting topics which did not survive the triage when preparing my slides – notably the widely-held UK view (though I was representing myself, not the UK) that the European-level standards institutions are, by and large, a waste of space, and that European nations should be using International Standards for all but a very few niche cases where a “European dimension” exists and a regional layer (i.e. a European layer) can bring benefits.

I also confined myself carefully to my own area of knowledge; ISO/IEC (JTC 1) ICT standardisation.

Following a few introductory slides on the functioning of JTC 1, and given the task of predicting the immediate future and describing the challenges ahead, I focussed on four main headings, as set out in the sections that follow.

Resisting vendor encroachment

The points here are:

• Vendors dislike international standardisation (when it does not function in their favour)

Anybody who has read my earlier piece will find the background argument to this familiar: international standardisation is an activity for nations and vendors have no standing. From time to time this causes upset (and in part explains some of the vendor-led assault on the integrity of the European standards institutions following the passage of OOXML).

• Expect continual pressure for a means of “direct participation” by vendors

A corollary of the above is a continued attempt for vendors to participate (i.e. have voting rights) in the international process. My own view on this is quite stark: vendors must never be allowed such rights.

• Governments must strongly resist this and maintain the de jure institutions such as ISO for their own use as bulwarks against corporate tyranny

My point here is that the international standards organisations are made “by governments, for governments”. The use of the words “corporate tyranny” are quite strong – but for myself I am convinced that the power afforded to corporations by the data collection and inspection efforts they may now mount today mean that, more than ever, governments are necessary to keep corporations, rapacious beasts that they are, in check.

Surprisingly (to me) it was this view which gathered most negative reaction, with a view expressed around the table that standards should be made “by industry for industry” and that governments should generally not interfere (much was also made of the distinction between a nation’s view and the view of that nation’s government – which I should treat with more rigour). Of course this is partly right too. I think I need to find a more nuanced way of describing how “industry” may be involved, but without being allowed the final say, at least in international standardisation. I think though, I have a wider view of the social dimension government can bring, and a sharper suspicion of the evils of corporatism, than seems to be currently common in EU circles. I’d never thought of myself as left-wing before. Hmmm.

Effects of economic slowdown

• It is difficult for vendors to commit staff to standardisation activities when under economic pressure

As predictions go, this one doesn’t require much insight. Already the ICT industry is seeing lay-offs in large numbers as the global recession bites. There is always a danger when the pressure is on that standardisation is seen as a luxury, non-gainful activity – and the kinds of gurus and thought-leaders in corporations who do this stuff can find their jobs under threat.

• Health and survival of vendor-led consortia threatened

On a larger level, corporations themselves (those that survive) might come to see participation in standards activities as something which might be given a rest. For consortia which depend on the corporate dollar this can present a challenge. Some commentators, for example, see the W3C’s (perfectly reasonable) recent announcement of validator donation program (“we really can use that money”) as just such evidence of the negative impact of recession on standardisation.

• Standardisation needs to be recognised as of much as a market-enabler as pure innovation; governments may assist

This is the nub: standardisation is (when it is done well) a first-class form of market enablement. Yet the myopic corporations are not generally in a position to see this, and even if they can they have no interest (they tend to be interested only in markets in which they are confident they can win, rather than in market creation in the general sense). And so this is a perfect example of where governments, with their (one hopes) wider and longer-term view, can intervene by continuing to support standards activities through supporting their National Standards Bodies.

Reform / modernisation

• The publishing business model of many European standards bodies (“selling pages”) is out of tune with the realities of modern ICT standardisation

The broken business model of National Standards Bodies is a serious problem. With ICT standards often (by demand) being given away free-of-charge the traditional means by which NBs can recoup the cost of making standards has gone away. Why, then, should they bother? I have no quick answer to this question, but I expect an answer would involve the need to have both governments and vendors contributing more to the financial cost of creating international ICT standards.

• Some European bodies assist experts in standardisation in their duties, but this is far from universal

Related to this is the fact the individual experts who are not employed by corporations can find it difficult to contribute to international standardisation – especially when international travel is required. Some enlightened standards bodies (the UK for example) do help defray these expenses but it would be worthwhile to see this kind of support more widely and consistently deployed throughout the European nations. The danger is that if independents cannot be funded, the void will be entirely filled by well-funded corporate employees, and the ensuing lack of balance would be regrettable.

• The Directives governing JTC 1 standardisation are archaic and confusing and in need of improvement

Another kind of reform that is needed is that of the JTC 1 Directives themselves. This dovetails with the above two points as securing this kind of reform requires dedicated experts – it will not just happen by itself.

IPR reform

• The patent spectres haunting innovation in ICT are also at work in the standards arena

The discussion around IPR, and particularly patents, in the EU is a vibrant one. My own view is that software patents are A Bad Thing but if we are to have them then standardisation could have a particular role to play in the standardisation landscape.

• Ideally, an International Standard should provide a guarantee of freedom from IPR encumbrance

This idea (first raised on this blog by André Rebentisch) is that a certain class of International Standards could provide a “safe haven” for implementers, who should feel secured against legal action for any implementations that arise directly from the use of that standard.

• Governments (the EU) could usefully legislate in this area

The way this could be practically achieved is for the EU to legislate that for certain de jure standards (JTC 1 ones, in my examples) which are labelled as unencumbered, there was an absolute defence against patent actions centred around IPR embodied therein. IANAL, but this kind of thing could usefully protect innovation and further enhance and clarify the role of de jure standards organisations, and the special relationship they have with governments/nations.

Again, here I sensed I was a little out-of-tune with the consensus round the table, which seemed to hold that patents were a given, and a useful aspect of the standards landscape (of course, it difficult to tell here how much that view applied to the area of ICT on which I was focussed, and which brings its own particular difficulties). On the other hand, there was general agreement that this area was one in which a lot of further work is required.

Brussels by Night #1
Brussels by Night — Hotel Frontage.

Comments (3) -

  • Chris Ward

    12/25/2008 3:13:40 AM |

    Interesting point about 'vendors'.

    I guess you are always welcome to use meeting space at an IBM Development Laboratory to agree on a standard. But then you will come up with SNA, rather than TCP; or System/360 Floating Point, rather than IEE754 Floating Point; or whatever the modern-day equivalent is. That is, it will be an IBM vendor-owned standard, rather than an open vendor-neutral standard.

    So what's the better way "For the world" ? And how do you get the vendors to buy in to the process ?

  • IMFreakz

    1/2/2009 10:14:25 PM |

    I do not know much about the article, but I love the master picture. It look great

  • Alex

    1/5/2009 1:45:41 AM |


    It's tricky.

    I don't think using IBM facilities would lead to IBM-biased standards. In general, kudos to corporations which support standardisation processes. Standards orgs generally want from corporations: experts brains, facilities and money. But not undue corporate influence. As with most imporant things in life, it's not a question of avoiding competing interests, but controlling them.

    Corporations might be motivated to participate in standards creation for many reasons, but primarily because it can be a first-class form of (in the pure sense of the word) marketing -- i.e. making sure products are aligned with the market in such a way as to boost profitability over the longer term. When a standards org responds to the market, this is true at any rate!


    Thanks! Brussels certainly looked more interesting at night than during the very grey December day time.

    - Alex.

Comments are closed