Will too many standards cooks spoil the IoT broth?
Too Many Cooks? photo via Flickr © Kecko (CC BY 2.0)
As far as the global comms industry is concerned, 2014 was partly characterised by speculation about the potentialities if the Internet of Things (IoT). Much of the discourse on the subject was PR-engendered wishful thinking but some work was firmly based in the deployment of real-life, real-world services and applications as de facto IoT standards began to emerge.
It seems that IoT, as has so often been the case in the history of the development of so many other comms technologies, is yet another example of official standards bodies tending to lag so far behind the rate of development of the very technologies they are supposed to be helping to formalise, that de facto standards become the norm because of the absence of anything else. Indeed, they are often so entrenched that by the time industry groups and standards bodies actually get round to codifying things the "official" specifications are rendered all but redundant.
Of course, there are notable and laudable exceptions to this trend, such as SDN, NFC and MEC, where standards have been quickly and successfully debated, agreed, specified, codified, tested and applied. However, it looks as though IoT is unlikely to be one of the standards success stories - unless things change very soon.
IoT is huge, diffuse and is either still predominantly experimental or in its commercial infancy. It involves the linking of devices that have never, ever, been connected before in any sort of open and coherent comms network. Deploying and managing an IoT network is a massive challenge not least because devices, objects and apps will have to be able to work together and communicate between one another at some point and at some level even though the products that enable the network and services will come from a plethora of manufacturers and vendors.
That's the challenge. For IoT to reach its much-vaunted potential of connecting homes, offices, enterprises, factories, towns and cities, billions upon billions of increasingly smart but also increasingly inexpensive devices will be needed because no-one is going to pay premium prices for something like a gizmo in a fridge that will tell a user when it is time to buy some more milk. To succeed IoT devices and networks will have to be cheap to supply, monitor and to manage.
Naturally enough, manufacturers and vendors want to capitalise on the growing interest in smart devices and the Internet of Things and they want to do so now, not at some indeterminate date in the future when standards bodies will have done their work. This date is mooted by most analysts to be sometime in mid-2017!. Hence the coming together over the course of 2014 of many interested parties in the IoT space and the emergence of de facto IOT standards.
Over the course of last year (and late 2103) six more or less informal but powerful groupings attempted to impose some formal shape on the IoT. Several of the efforts overlapped and duplicated work and this caused considerable confusion, a confusion could well get worse before it gets better.
New Players at the start of 2015
Here's a list of the main IoT standards players that emerged last year and will be active this.
1) The AllSeen Alliance
This one was actually born in December 2103 but gained traction in the industry some months later. Members of the Alliance (which number over 100) include the likes of Qualcomm and Cisco as well as Panasonic and other consumer electronics manufacturers. Founding member Qualcomm is probably the most important in the Alliance as its open-source software framework, AllJoyn, is based on Qualcomm code and technology platform.
The goal of the AllSeen Alliance is to permit residential and business devices that use different operating systems and communications networks protocols to co-ordinate and work with one another. Last month Microsoft began shipping AllJoyn-enabled kit and is embedding it into Windows 10.
2) IEEE P2413.
Representing the ancien regime in standards, the Institute of Electrical and Electronics Engineers, is IEEE P2413. The IEEE has a long and venerable industry but is regarded in some quarters and by some manufacturers as being a part of the problem rather than the solution. The main criticism is that the body is too hidebound and slow to be relevant in the fast-moving new world of rapidly developing and changing technologies. Nonetheless the IEEE has instituted a Working Group whose purpose is to prevent duplication of effort and overlap evident in parts of the sprawling IoT specifications being developed by industry organisations, formal or informal.
Back in July last year IEEE P2413 held it's first meeting with 23 vendors and other interested parties and has announced that it "hopes" to have a fully defined IoT standard written and published by sometime in 2016. That's an awfully long time terms of the expected rate of development and uptake of IoT and that, say the manufacturers and vendors is why de facto standards will dominate the industry by the time the IEEE's standards solution is available.
3) The Industrial Internet Consortium.
This body which official began work in March 2014 comprised originally of AT&T, Cisco (again), General Electric, IBM and Intel. It now has in excess of 100 members including the powerful likes of Huawei, Microsoft (again) and Samsung. The Consortium focuses on IoT in the enterprise and its policy is not to set standards but rather to work with standards bodies ensure IoT technologies will interoperate together across business sectors.
The basic idea is to engender interworking and co-operation between entities that hitherto have developed both IoT and M2M technologies in isolation from one another. This involves the definition of base requirements for standards, reference architectures and the proof of concepts.
4) The Open Interconnect Consortium.
Also launched in July last year, the Open Internet Consortium counts Dell, HP, Intel, Lenovo and Samsung amongst its 50+ members. It is working on the writing of a series of open-source specifications that will enable devices to find, isolate and recognise one another and engender communication, inter-working and data exchange. Plans are for the first source code to be released to developers by the end of 2015.
5) The Thread Group.
This consortium, also formed in July 2014, counts ARM Holdings, Nest (Google's smart thermostat and smoke alarm buyout of last year) and Samsung among its 50 members. It is pushing a mesh networking protocol for low-power devices in domestic environments. What gives Thread and early and important edge is that its protocol works on a chip that is already on the market and gives every device an IPv6 address. What's more, Thread defines nothing more than networking, thus paving the way open for higher layer specifications such as AllSeen and OIC to be embedded in Thread products. It will begin to certify products in the first half of this year.
The Fresh Five listed above are in addition to other bodies and industry consortia that have been around for more than a year, including the likes of oneM2M, and the International Society for Automation. It is also a strong possibility that other new confederations, consortia and interest crops will emerge over the course of this year
So, there are already a lot of standards bodies out there, some working on complementary solutions and others on competing ones - and, it should be noted, some of those competing and overlapping are actually members of the same body.
This is confusing, especially for the domestic and residential market, so expect the first really big I0T success stories to emanate from the enterprise arena where open internecine conflict is less likely.