REGISTER | LOGIN
Breaking News
News & Analysis

Why Japan Hasn't Led IoT

11/19/2015 10:30 AM EST
8 comments
NO RATINGS
Page 1 / 2 Next >
More Related Links
View Comments: Newest First | Oldest First | Threaded View
y_sasaki
User Rank
Author
Re: Zigbee
y_sasaki   11/20/2015 4:00:18 PM
NO RATINGS
It is true that iTRON is essentially "Japanese local standard", despite it is free-license, open standard, English document available, having 30 years of successful history. Several supported commercial products and free open-source implementations available (almost all in Japan though). Having said that, it is also true that iTRON did not spread outside of Japan, meanwhile much newer FreeRTOS (introduced in 2002) quickly outgrew iTRON and becoming almost international de-facto standard of small-footprint open-source RTOS.

iTRON may be not the same category of full-featured fully-supported RTOS like VxWorks, ThreadX or Integrity. But FreeRTOS is a bare-bone RTOS Kernel - just like iTRON. Why iTRON could not spread outside of Japan but FreeRTOS did is tantalizing question to me. It may be hold the answer to Dr Sakamura's frustration, why "initiated in Japan" standard rarely be internationally sucessful.

docdivakar
User Rank
Author
Re: Japan is just being Japan...
docdivakar   11/20/2015 3:15:39 PM
NO RATINGS
Junko, I agree... being a 'maker' myself, I wouldn't claim the maker community is responsible for IIoT or for that matter IoT itself because that simply isn't accurate! About 4 years ago I hosted a workshop at Santa Clara University on M2M and Internet of Things. In that event, we were starting recognize the different but merging definitions of connected things and visions were formulated to unify them. Soon after we saw the proliferation of single board computers and then the availability of eval boards with modules for connecting 'things' that are so common these days.

Back to your story, here is an interesting anecdote with respect to Japan's 'closed' approach to IoT. Three years ago I met with VP of R&D of a major Japanese company starting with the letter N that had visions for connecting things with emphasis at first on energy monitoring for residential units. The company's approach was bogged down in protecting some work on databases and the vision was showing complicated middleware thrown into the 'stack' which I found redundant and added unneeded latency. Needless to say I did not partner with the company on the project. Your article reinforces that experience!

MP Divakar

tb100
User Rank
Author
Zigbee
tb100   11/20/2015 2:40:47 PM
NO RATINGS
I remember Tron! I never saw any Tron hardware available in the US, though (not that I looked very hard for it).

The first I noticed a home automation product in the US more advanced than the old X10 system was Zigbee, which started in 1998. It was very impressive but when it came out licensing was extremely expensive, so it seemed like it was meant for home automation for the wealthy's mansions.

Over the years Zigbee and later Zwave became much cheaper, and we started connecting the Internet to everything, including ZigBee and Zwave hubs, so it all morphs into the Internet of Things for home automation.  Nest (now Google) has jumped on the bandwagen with their 'Thread' system.  Most of the home automation systems I see now days rely on one or more combinations of these standards.

The 'maker' community may not have started this process, but they are certainly embracing these new products in lots of innovative ways, and I'm convinced that a lot of the new IOT products come from this mindset.

y_sasaki
User Rank
Author
Echonet Lite
y_sasaki   11/20/2015 2:01:45 PM
NO RATINGS
Well, there is at least one "open standard for IoT" initiated by Japanese electric industry called "Echonet Lite".

Echonet is originally initiated as standard "smart" applicance framework for PLC, back in year 2000. It was long before "IoT" became hot buzz-word.  Echonet Lite is released in 2011 as evolution of Echonet, re-targetted for Smart Home and Smart Meter.

It is fact that Echonet Lite is one of oldest proposed IoT framework, open standard already issued as ISO and IEC. And it is also true that we rarely hear word "Echonet" outside of Japan. As my personal feeling, Echonet does not have traction even in Japan, despite of 15-years of long history. Why?

I don't know the answer. One thing I feel is "double standard" tendency in Japanese companies. Even a company associated with open-standard project, it does not necessary mean the company is also activly materializing the standard as product. Often they are totally different, isolated groups who are participating standard project and who are actually developping products. In my opinion, it is rather be management failure then individual engineer's effort, that so many "initiated in Japan" open-standard had failed miserably.

Bert22306
User Rank
Author
Or a lot of these publicized IoT efforts won't amount to much
Bert22306   11/19/2015 6:12:05 PM
NO RATINGS
I continue to disagree with an approach to this topic which attempts to make IoT something new and different. So as a result, I'm probably more sympathetic with the Japanese view.

It was instructive, for example, to see H.264 compression mentioned. Because that codec, along with all other standards-based codecs, are an example of the elements of IoT that this industry has been developing over the decades. Certainly the IETF efforts are all part of it too. Sure, codecs for severely resource-constrained sensors, such as one might expect as sensors become smaller and much more ubiquitous, will be increasingly in demand. But surely, that's also nothing new. More of the same.

The IETF has had a working group involved with ad-hoc networks, others addressing version 6 of the protocol, another with IPv6 running over energy-constrained media, and so on. Many of thes efforts began long before anyone coined the term IoT. Because people have been working this general problem for years and years. Even Japanese engineers have.

And this "cloud" business also becomes more vague and ethereal than it needs to be. Take image processing required of a simple camera, used as an example in the article. Is it going to be done in "the cloud," rather than being done on board? Well, maybe in some cases, maybe not in others. You might not want to rely on "cloud" connectivity for your every picture. And what is "the cloud" anyway? Is it something new and different? No. It would be either servers sitting in centers all over the world, just like the web, or it could be a huge number of edge devices sharing a small part of their CPU cycles, sort of like Napster (that long ago already). So even there, you are not necessarily taking the processing out of edge devices. You might be distributing the processing to MORE edge devices.

And any other variant in between.

Sometimes, often actually, much-publicized, centrally controlled mega-projects end up being a little ho-hum. This happens because "necessity is the mother of invention." Vaguely defined concepts, not so much. Big, centrally-run mega programs can end up spinning wheels without a whole heck of a lot to show for it.

Les_Slater
User Rank
Author
Fundamentally a partitioning issue
Les_Slater   11/19/2015 6:06:07 PM
NO RATINGS
The Aggregate Model suggested by Sakamura is a particular case of classical partitioning. There has always been a near sightedness in the engineering community of how problems are resolved as resource technology changes. The cost of communications approaching zero is certainly one of those resources that should be thought differently of, storage has also gone down that path.

What is needed is a formality of problem geometry where, among other factors, what ifs can be played with, with a back-end generating architectures with various costs like manufacturing, energy consumption, etc.

junko.yoshida
User Rank
Author
Re: Japan is just being Japan...
junko.yoshida   11/19/2015 4:24:44 PM
NO RATINGS
Thanks for chiming in. I think you're too kind to Japan in your comment, however.

With all due respect, things like industrial IoT did not start from makers' movement. Japan had a good chance to contribute to industrial IoT by leveraging her experience with Toyota's Kanban system or Kamatsu'KOMTRAX. But Japanese have been too miopic, focused on their "closed" system in my humble opinion. 

denseWaveRunner
User Rank
Rookie
Japan is just being Japan...
denseWaveRunner   11/19/2015 4:19:31 PM
NO RATINGS
IoT didn't really emerge from the "embedded" community, it emerged from the "maker" community.  Two different mindsets.  In any event I think Japan is more known for product improvements and quality than it has been as a market innovator, so blasting a community for not being an innovator isn't really justified.

Most Recent Comments
David Ashton
 
Smartalix
 
Sang-Yun
 
Sang-Yun
 
realjjj
 
DMcCunney
 
David Ashton
 
rick merritt
 
rick merritt
Like Us on Facebook
EE Times on Twitter
EE Times Twitter Feed