Governance issues regarding the “Internet of things”: comments about a chronicle from Jacques Attali announcing the upcoming French ONS root (English version of the previous post).
Par Philippe GAUTIER le vendredi 30 novembre 2007, 11:50 - Economic Warfare and Competitive Intelligence - Lien permanent
In a chronicle (Europe 1 radio - 25th of November 2007:
http://www.europe1.fr/informations/chronique-t2.jsp?idboitier=767123#)
Jacques Attali points out a very good news: GS1 France is about to
launch an independent ONS root, which
announcement will occur on the 03th of December.
However there are still some important remaining questions that need to be addressed if the "EPC network" wants to be the "Internet of Things"
In a chronicle (Europe 1 radio - 25th of November 2007:
http://www.europe1.fr/informations/chronique-t2.jsp?idboitier=767123#)
Jacques Attali points out a very good news: GS1 France is about to
launch an independent ONS root, which
announcement will occur on the 03th of December.
Actually, this ONS root (*) will help the “EPCGlobal European community” to protect its own
numeric sovereignty; and subsequently its economic one as well since ONS is a
cornerstone of the EPCGlobal design for the Internet of things (*).
We sincerely congratulate GS1 France for it: as a think tank,
SIAIGE has, amongst others, strongly lobbied (*) for it
(French Government, European commission, GS1, EPCGlobal, press, etc).
We also have supported the very first independent ONS root in Europe, the
one used in the operational EPCGlobal pilot for BENEDICTA (pallets traceability
in the food industry).
However there are still some important remaining questions that need to be
addressed if the "EPC network" wants to be the
"Internet of Things":
- Will the ongoing “EPCGlobal standardization” be the only one to support the
Internet of Things (although the most advanced)?
- EPCGlobal has, potentially, a wider scope than GS1 (many industries are
participating) but the potential target is bigger than the only manufactured
products
- What about the future of the existing “EPC naming system”, fully controlled by
EPCGlobal (monopoly)?
- EPCGlobal is a JV between GS1 US and GS1 WW and is, accordingly, controlled by GS1. Subscribing to EPCGlobal is not free of charge.
- GS1 is mostly representing FMCG companies (manufacturers, retailers, etc.) and local organizations (by countries) can be, sometimes, private. Either Public entities or consumers/citizen representatives are not directly represented within GS1.
- There are some raising voices asking for a full open model: some working
groups (IETF, public entities) are also considering that question.
The purpose of this post is definitely not to criticize what have been done
so far: what we are experiencing today should be considered as a very
satisfying beginning for something bigger.
Actually, to go on in that hopeful direction, we could consider different
alternatives (non-exhaustive list):
- The opening of GS1 to some others players (consumers, public entities, etc.)
- EPCGlobal’s cut off from GS1 to make it independent and fully representative of every involved stakeholder with a dedicated governance
- Transfer of the EPCGlobal standards to an open source project without precisely addressing the governance issues (e.g.: sourceforge, etc.)
- a cheaper subscription model
- Etc.
We will not give the relevant answer right now (is SIAIGE really capable of
this?) to those open questions: we just want to make concerned people paying
attention to what we are observing today should be considered as a major step
for a wider process.
PG
(*) see former posts on this site (mainly in French)
Commentaires
During the last four years since EPCglobal was formed in late 2003, significant progress has been made by the EPCglobal community in the development and ratification of user-driven open standards for many elements of the EPC Network architecture. These include software standards, as well as standards relating to hardware, such as the UHF Class 1 Gen 2 air interface, which has even been adopted by ISO as ISO 18000-6 C. They have also been active in promoting the technology and helping to drive adoption. This in turn has helped lead to increased demand and falling prices for low-cost RFID tags etc.
Upon ratification, EPCglobal standards are freely available to everybody to download from http://www.epcglobalinc.org/standar... . Recently, the associated files (XML schema, WSDL files etc.) have also been made available. In addition, there are open source projects for the EPC Network, in Java ( http://www.accada.org ) and C#/.NET ( http://source.mit.edu ).
The EPC Network is currently positioned primarily as a business-to-business infrastructure that enables businesses and supply chains to operate more efficiently through the automated capture and exchange of serial-level data about the location and movement of individual objects. Although an Object Name Service (ONS) provides a lookup mechanism to find authoritative information about an individual object or EPC (usually information provided by the object's manufacturer), its implementation in DNS is not well suited to providing dynamic track and trace information in a scalable and secure manner. For this purpose, complementary lookup services, known as 'Discovery Services' are under development. EPCglobal is currently gathering user requirements for these, which will then feed into the development of a technical standard. In parallel, three IETF internet drafts on Extensible SupplyChain Discovery Services (ESDS) have been posted - and a mailing list esds@ietf.org has been established to promote discussion on the design of the technical interfaces and data model. Over the last 18 months, the BRIDGE project (a 3 year integrated project within the European Commission's 6th framework programme (FP6)) has also undertaken significant work on the requirements, design and prototyping of Discovery Services. The BRIDGE team has also contributed two of their reports to both EPCglobal's Data Discovery Joint Requirements Group and has also posted the reports to the ESDS@IETF.ORG mailing list, in order to contribute to current standardization efforts.
Discovery Services will probably provide much greater business value than an Object Name Service. It is already apparent that a number of solution providers are already providing services targeting this role - and it is hoped that we can work towards convergent interfaces between them, in order to achieve interoperability and a genuinely competitive multi-vendor marketplace for Discovery Services.
Before this point is reached, EPCglobal may need to rethink their current business model, in which subscribers are provided with entries in their root ONS, as one of the subscriber benefits (in addition to the opportunities to network with other early adopters and actively participate and influence the development of user-driven standards). Of course, a future 'Internet of Things' has a much broader scope than the current 'EPC Network'; we can imagine several location-based services for citizens - effectively interfacing the World Wide Web much more closely to objects and real-world locations.
Much of the innovation in the World Wide Web has been driven by the creative and altruistic ideas of enthusiastic individuals - and it is essential that citizens are not deterred from full participation in the future 'Internet of Things'. In the world wide web, there are very low economic barriers to citizens who wish to create new websites and services - the costs of registering a domain name and hosting websites are very affordable - and there are also several providers of free hosting and free website addresses. The availability of freely available web browsers and many free or affordable software tools for creating information content and services means that there are few economic barriers to participation. The free availability of relevant web standards from W3C and related network standards from other bodies, such as IETF enables global interoperability.
Turning to the future 'internet of things', the open standards from EPCglobal may provide a strong foundation for the technical standards - and the availability of freely available open source software for the EPC network should enable the prototyping and development of new and innovative applications and services by enthusiastic citizens. Falling costs of RFID tags (economies of scale driven by the large volumes to be used by business) is also a welcome development.
However, at present, the EPC identifier namespace does not yet cater for tagging of individual objects by citizens, nor the tagging of arbitrary locations anywhere on the planet (such as signposts and viewpoint information boards within the landscape and national parks, which are not commercial 'business locations'). The good news is that the Electronic Product Code (EPC) is really a framework that can support multiple identifier schemes - and there are already standards such as Tag Data Translation and the Tag Data Standard, which facilitate the mapping of any identifier system into an EPC format. It is therefore not a difficult technical challenge to develop an EPC identifier scheme that can encode the latitude, longitude and altitude of any place on earth (so that it could be read via a mobile phone and linked to location-based services (e.g. photos, video, reviews) on the web). Likewise, it would technically be a very trivial matter for an 8-bit EPC header code to be allocated for private / internal-use identifiers generated by citizens or companies. It is even conceivable that citizens might use the short codes generated by URL-shortening services (such as snipurl.com, tinyurl.com, purl.org) to encode an RFID tag and link it to a particular resource on the web). Most of the EPCglobal technical standards would continue to function even if the EPC namespace were opened up to serve an expanded 'internet of things' that is open for innovation and active participation by all citizens.
Mark,
I do agree with everything in your comment.
We're on the same line !
Regards,
PG